2020_SPRING_RUSH/week2/week2_2_CCTV_in_Seoul.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 🦉COSADAMA Introduction to Data Science Study\n",
    "\n",
    "- 일자: 2020-03-23\n",
    "- 작성자: 박하람 \n",
    "- 참고자료: 파이썬 머신러닝 완벽가이드(권철민), 비쥬얼프로그래밍 수업(김승태), 파이썬으로 데이터 주무르기(민형기)\n",
    "\n",
    "## 2. [프로젝트] 서울시 구별 CCTV 현황 분석\n",
    "\n",
    "이번 주차부터는 본격적으로 프로젝트를 통해 데이터 분석에 대한 기초 감각을 익히려 합니다. 직접 데이터를 다루면서 지난 시간에 학습한 pandas, numpy, matplotlib를 다루어보아야 확실하게 쓸 수 있더라구요. 그냥 보기만 하면 휘리릭 날아가 버립니다😅 오늘은 **서울시 구별 CCTV 현황 분석**을 해보려 합니다. 자잘한 것에 집중하기 보다 큰 그림을 보면서 이렇게 데이터 분석 과정이 이루어지는구나 라고 접근해주시면 좋을 것 같습니다. \n",
    "\n",
    "**서울시 구별 CCTV 현황 분석**에서는 단순히 어디에서 CCTV가 많이 설치됐는지부터 시작해서 구별 인구 대비 비율을 확인하는 것까지 진행합니다. 특히 인구 현황을 보면서 구별 인구 현황도 확인하고, 구별 CCTV 현황을 시각화하는 것까지 해볼 겁니다. 이번 프로젝트를 통해서는 pandas와 Matplotlib의 기초를 배울 수 있습니다. \n",
    "\n",
    "### 2.1 문제 정의하기\n",
    "\n",
    "본격적인 프로젝트에 앞서서 꼭 생각해봐야 할 중요한 지점이 있습니다. 우리는 판다스와 넘파이, 맷플랏립은 왜 배워야 할까요? 이 기술이 무엇에 필요한 걸까요? 무엇 때문에 우리가 이것을 이렇게 힘들게 배워야 하는 걸까요? 제 말은 결론적으로 배보다 배꼽이 크면 안된다라는 것을 말씀드리고 싶은 겁니다. 데이터를 분석하는 필요한 프로그래밍 지식들은 엄밀히 말하면 **도구**에 불과해요. 제일 선행되어야 하는 것은 **\"내가 무엇을 해결하고 싶으냐? 무엇이 문제점이냐?\"**입니다! 내가 어떠한 문제를 해결하고 싶다는 생각이 들면, 그 때부터 그 문제를 해결하기위해 **어떤 데이터를 수집**해야 하는지, 그 데이터를 어떻게 **가공**해야 할 것인지, 가공한 데이터를 어떻게 **분석하고 의미화할 것인지**, 그래서 나타난 결과가 **어떻게 문제를 해결**할 것인지를 알 수 있게 되는거죠. \n",
    "\n",
    "이후에 머신러닝 프로젝트를 하더라도 마찬가지에요. 제가 지역별 건강 격차를 알고 싶어요, 그리고 그 건강 격차가 미래에 어떻게 될 지 예측을 해보고 싶어요. 그러면 우선 지역별 건강 격차를 드러낼 수 있는 데이터를 수집해야 해요. 아주 단순히 지역별 병원의 수라고 해봅시다. 그러면 지역별 병원의 수는 어떻게 구할 수 있는지부터 고민해야죠. 지역별 병원의 수를 구했으면 이 데이터를 어떻게 머신러닝에 맞게 가공을 해야 하는지, 또 나는 지역별 병원의 '수'를 예측하고 싶은데 그럼 또 어떤 알고리즘을 적용해야 할지 등등등 모든 것이 내가 알고 싶고 검증하고 싶은 **문제 정의**로부터 시작됩니다. \n",
    "\n",
    "#### Q. 서울시 구별 CCTV 현황 분석을 하는 이유? \n",
    "물론 제가 이 프로젝트를 한다고 하니까 분석하지만, 그 전에 그냥 상상해봅시다. 우리는 다 사회학도들이니까 사회에 대한 궁금증이 많은 사람들이잖아요?😉 (이렇게 상상해야 나중에 써먹지 않겠습니까!!!) 예를 들어 제가 (진짜 대충 생각한건데) '지역구가 부유할수록 CCTV를 더 많이 설치할 것이다.'라는 가설을 검증하고 싶어요. 오, 전체 대한민국은 너무 넓으니까 우선 서울시만 가지고 검증을 해봐야겠다, 이거 어디에서 데이터 구할 수 있지...? 아하, 서울 데이터 광장에서 구할 수 있구나! 그럼 또 뭐가 필요하지? 엇, CCTV 수만 가지고 이것을 제대로 검증할 수 있을까? 지역구의 땅이 넓어서 많이 설치하는 건 아닐까? 사람의 수가 많아서 많이 설치되어 있는 건 아닐까? 등등 여러가지 통제변수들도 생각해보면서 데이터로 검증해봐야겠다는 생각을 하는 겁니다. \n",
    "\n",
    "### 2.2 데이터 수집하기 \n",
    "\n",
    "그래서 결론적으로 여러분은 **서울시 자치구 연도별 CCTV 설치 현황**과 **서울인구** 공공데이터를 수집합니다. 서울인구 데이터는 지역구 인구 대비 CCTV 설치수까지 알아보려고 구하는 겁니다. 갑자기 과정이 생략되었는데요, 빈공간을 상상해보시고 이제 진짜로 프로젝트를 해보도록 하죠! \n",
    "\n",
    "- [서울시 자치구 연도별 CCTV 설치 현황](https://data.seoul.go.kr/dataList/OA-2734/C/1/datasetView.do)\n",
    "- [서울시 주민등록인구 통계](https://data.seoul.go.kr/dataList/419/S/2/datasetView.do)\n",
    "\n",
    "주교재에 적힌 사이트와 현재 사이트가 달라지고 데이터도 개편되어서 위의 링크에서 다운받으려고 하면 그대로 따라할 수 없을 거에요. 그래서 저는 코사다마 구글 드라이브 week2 - 참고자료 폴더에 두개의 데이터를 넣어두었으니까 쓰시면 됩니다. 그래도 앞으로 써먹어야 하니까 위의 링크들 들어가셔서 이리저리 구경해보세요! \n",
    "\n",
    "### 2.3 데이터 가공하기 \n",
    "\n",
    "파일이 하나는 csv(콤마로 구분된 텍스트 파일)이고, 다른 하나는 엑셀입니다. 파이썬으로 이 둘을 모두 읽어볼 수 있는 방법을 익혀볼게요. pandas가 데이터 분석에 편리한 이유는 이렇게 단순한 코드 하나만으로 엑셀과 csv와 같이 다양한 파일을 읽어낼 수 있는 모듈을 가지고 있어요. \n",
    "\n",
    "#### pandas와 numpy import하기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd    \n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### csv와 엑셀 파일 읽기\n",
    "\n",
    "판다스와 넘파이를 임포트 해주고, 먼저는 서울시 자치구 연도별 CCTV 설치 현황 csv 파일부터 읽어봅니다. \n",
    "\n",
    "- .read_csv()로 간단히 읽어낼 수 있습니다!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>기관명</th>\n",
       "      <th>소계</th>\n",
       "      <th>2013년도 이전</th>\n",
       "      <th>2014년</th>\n",
       "      <th>2015년</th>\n",
       "      <th>2016년</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>강남구</td>\n",
       "      <td>2780</td>\n",
       "      <td>1292</td>\n",
       "      <td>430</td>\n",
       "      <td>584</td>\n",
       "      <td>932</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>강동구</td>\n",
       "      <td>773</td>\n",
       "      <td>379</td>\n",
       "      <td>99</td>\n",
       "      <td>155</td>\n",
       "      <td>377</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>강북구</td>\n",
       "      <td>748</td>\n",
       "      <td>369</td>\n",
       "      <td>120</td>\n",
       "      <td>138</td>\n",
       "      <td>204</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>강서구</td>\n",
       "      <td>884</td>\n",
       "      <td>388</td>\n",
       "      <td>258</td>\n",
       "      <td>184</td>\n",
       "      <td>81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>관악구</td>\n",
       "      <td>1496</td>\n",
       "      <td>846</td>\n",
       "      <td>260</td>\n",
       "      <td>390</td>\n",
       "      <td>613</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   기관명    소계  2013년도 이전  2014년  2015년  2016년\n",
       "0  강남구  2780       1292    430    584    932\n",
       "1  강동구   773        379     99    155    377\n",
       "2  강북구   748        369    120    138    204\n",
       "3  강서구   884        388    258    184     81\n",
       "4  관악구  1496        846    260    390    613"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "CCTV_Seoul = pd.read_csv('01. CCTV_in_Seoul.csv', \n",
    "                        encoding='UTF-8')\n",
    "CCTV_Seoul.head()  # 데이터 앞의 5줄만 보여주세요!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "항상 데이터들을 살펴보는 것을 잊지마세요. 데이터가 무엇을 의미하는지도 중요하니까, 대충 아 강남구가 다른 곳에 비해서 많이 설치되어있네. 추이도 볼 수 있는데, 설치수가 줄어든 것도 보이고 늘어난 곳도 보이네요. \n",
    "\n",
    "그런데 구의 column이름이 '기관명'으로 되어 있어서 이걸 '구별'로 바꿔볼게요!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>구별</th>\n",
       "      <th>소계</th>\n",
       "      <th>2013년도 이전</th>\n",
       "      <th>2014년</th>\n",
       "      <th>2015년</th>\n",
       "      <th>2016년</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>강남구</td>\n",
       "      <td>2780</td>\n",
       "      <td>1292</td>\n",
       "      <td>430</td>\n",
       "      <td>584</td>\n",
       "      <td>932</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>강동구</td>\n",
       "      <td>773</td>\n",
       "      <td>379</td>\n",
       "      <td>99</td>\n",
       "      <td>155</td>\n",
       "      <td>377</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>강북구</td>\n",
       "      <td>748</td>\n",
       "      <td>369</td>\n",
       "      <td>120</td>\n",
       "      <td>138</td>\n",
       "      <td>204</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>강서구</td>\n",
       "      <td>884</td>\n",
       "      <td>388</td>\n",
       "      <td>258</td>\n",
       "      <td>184</td>\n",
       "      <td>81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>관악구</td>\n",
       "      <td>1496</td>\n",
       "      <td>846</td>\n",
       "      <td>260</td>\n",
       "      <td>390</td>\n",
       "      <td>613</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    구별    소계  2013년도 이전  2014년  2015년  2016년\n",
       "0  강남구  2780       1292    430    584    932\n",
       "1  강동구   773        379     99    155    377\n",
       "2  강북구   748        369    120    138    204\n",
       "3  강서구   884        388    258    184     81\n",
       "4  관악구  1496        846    260    390    613"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "CCTV_Seoul.rename(columns={CCTV_Seoul.columns[0] : '구별'}, \n",
    "                  inplace=True)\n",
    "CCTV_Seoul.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "inplace=True 옵션을 쓸 때는 조심하기! 그럼 서울인구수가 담긴 엑셀 데이터도 읽어볼게요. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>기간</th>\n",
       "      <th>자치구</th>\n",
       "      <th>세대</th>\n",
       "      <th>인구</th>\n",
       "      <th>인구.1</th>\n",
       "      <th>인구.2</th>\n",
       "      <th>인구.3</th>\n",
       "      <th>인구.4</th>\n",
       "      <th>인구.5</th>\n",
       "      <th>인구.6</th>\n",
       "      <th>인구.7</th>\n",
       "      <th>인구.8</th>\n",
       "      <th>세대당인구</th>\n",
       "      <th>65세이상고령자</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>기간</td>\n",
       "      <td>자치구</td>\n",
       "      <td>세대</td>\n",
       "      <td>합계</td>\n",
       "      <td>합계</td>\n",
       "      <td>합계</td>\n",
       "      <td>한국인</td>\n",
       "      <td>한국인</td>\n",
       "      <td>한국인</td>\n",
       "      <td>등록외국인</td>\n",
       "      <td>등록외국인</td>\n",
       "      <td>등록외국인</td>\n",
       "      <td>세대당인구</td>\n",
       "      <td>65세이상고령자</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>기간</td>\n",
       "      <td>자치구</td>\n",
       "      <td>세대</td>\n",
       "      <td>계</td>\n",
       "      <td>남자</td>\n",
       "      <td>여자</td>\n",
       "      <td>계</td>\n",
       "      <td>남자</td>\n",
       "      <td>여자</td>\n",
       "      <td>계</td>\n",
       "      <td>남자</td>\n",
       "      <td>여자</td>\n",
       "      <td>세대당인구</td>\n",
       "      <td>65세이상고령자</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>2017.1/4</td>\n",
       "      <td>합계</td>\n",
       "      <td>4202888</td>\n",
       "      <td>10197604</td>\n",
       "      <td>5000005</td>\n",
       "      <td>5197599</td>\n",
       "      <td>9926968</td>\n",
       "      <td>4871560</td>\n",
       "      <td>5055408</td>\n",
       "      <td>270636</td>\n",
       "      <td>128445</td>\n",
       "      <td>142191</td>\n",
       "      <td>2.36</td>\n",
       "      <td>1321458</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>2017.1/4</td>\n",
       "      <td>종로구</td>\n",
       "      <td>72654</td>\n",
       "      <td>162820</td>\n",
       "      <td>79675</td>\n",
       "      <td>83145</td>\n",
       "      <td>153589</td>\n",
       "      <td>75611</td>\n",
       "      <td>77978</td>\n",
       "      <td>9231</td>\n",
       "      <td>4064</td>\n",
       "      <td>5167</td>\n",
       "      <td>2.11</td>\n",
       "      <td>25425</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>2017.1/4</td>\n",
       "      <td>중구</td>\n",
       "      <td>59481</td>\n",
       "      <td>133240</td>\n",
       "      <td>65790</td>\n",
       "      <td>67450</td>\n",
       "      <td>124312</td>\n",
       "      <td>61656</td>\n",
       "      <td>62656</td>\n",
       "      <td>8928</td>\n",
       "      <td>4134</td>\n",
       "      <td>4794</td>\n",
       "      <td>2.09</td>\n",
       "      <td>20764</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         기간  자치구       세대        인구     인구.1     인구.2     인구.3     인구.4  \\\n",
       "0        기간  자치구       세대        합계       합계       합계      한국인      한국인   \n",
       "1        기간  자치구       세대         계       남자       여자        계       남자   \n",
       "2  2017.1/4   합계  4202888  10197604  5000005  5197599  9926968  4871560   \n",
       "3  2017.1/4  종로구    72654    162820    79675    83145   153589    75611   \n",
       "4  2017.1/4   중구    59481    133240    65790    67450   124312    61656   \n",
       "\n",
       "      인구.5    인구.6    인구.7    인구.8  세대당인구  65세이상고령자  \n",
       "0      한국인   등록외국인   등록외국인   등록외국인  세대당인구  65세이상고령자  \n",
       "1       여자       계      남자      여자  세대당인구  65세이상고령자  \n",
       "2  5055408  270636  128445  142191   2.36   1321458  \n",
       "3    77978    9231    4064    5167   2.11     25425  \n",
       "4    62656    8928    4134    4794   2.09     20764  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pop_Seoul = pd.read_excel('01. population_in_Seoul.xls', \n",
    "                          encoding='utf-8')\n",
    "pop_Seoul.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "잉? 근데 column 제목들이 마음에 안들죠? index 1번째 줄에 있는 것으로 타이틀을 변경하고 싶고, 필요없는 데이터들을 모두 지우고 싶어요. 그럴 때는 아래와 같이 하라고 주교재에 나와 있습니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\"\\npop_Seoul = pd.read_excel('01. population_in_Seoul.xls', header= 2, parse_cols = 'B, D, G, J, N', encoding='utf-8')\\npop_Seoul.head()\\n\""
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# why isn't it working?\n",
    "'''\n",
    "pop_Seoul = pd.read_excel('01. population_in_Seoul.xls', header= 2, parse_cols = 'B, D, G, J, N', encoding='utf-8')\n",
    "pop_Seoul.head()\n",
    "'''"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "이것을 프린트하면 에러가 떠요.\n",
    "으잉, 이거 왜 안되는거야! 책이랑 똑같이 했는데! 판다스가 계속해서 업그레이드 하기 때문에 그래요. 그럴 때는 우리에게 치트키가 있잖아요? "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# If you want to know parameters you don't know, Here is a tip! \n",
    "pd.read_excel?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>자치구</th>\n",
       "      <th>계</th>\n",
       "      <th>계.1</th>\n",
       "      <th>계.2</th>\n",
       "      <th>65세이상고령자</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>합계</td>\n",
       "      <td>10197604.0</td>\n",
       "      <td>9926968.0</td>\n",
       "      <td>270636.0</td>\n",
       "      <td>1321458.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>종로구</td>\n",
       "      <td>162820.0</td>\n",
       "      <td>153589.0</td>\n",
       "      <td>9231.0</td>\n",
       "      <td>25425.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>중구</td>\n",
       "      <td>133240.0</td>\n",
       "      <td>124312.0</td>\n",
       "      <td>8928.0</td>\n",
       "      <td>20764.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>용산구</td>\n",
       "      <td>244203.0</td>\n",
       "      <td>229456.0</td>\n",
       "      <td>14747.0</td>\n",
       "      <td>36231.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>성동구</td>\n",
       "      <td>311244.0</td>\n",
       "      <td>303380.0</td>\n",
       "      <td>7864.0</td>\n",
       "      <td>39997.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   자치구           계        계.1       계.2   65세이상고령자\n",
       "0   합계  10197604.0  9926968.0  270636.0  1321458.0\n",
       "1  종로구    162820.0   153589.0    9231.0    25425.0\n",
       "2   중구    133240.0   124312.0    8928.0    20764.0\n",
       "3  용산구    244203.0   229456.0   14747.0    36231.0\n",
       "4  성동구    311244.0   303380.0    7864.0    39997.0"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Wow, it's working!\n",
    "pop_Seoul = pd.read_excel('01. population_in_Seoul.xls', \n",
    "                          header= 2,                       # header는 row 2번째줄로!\n",
    "                          usecols = 'B, D, G, J, N',       # 요 columns만 가져온다!\n",
    "                          inplace=True, encoding='utf-8')  \n",
    "pop_Seoul.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "이제 제대로 나왔네요. 그런데 columns 이름이 마음에 들지 않아요. 그리고 이 표를 잘 보세요. 우리가 나중에 구별 인구 대비 CCTV 설치수까지 비교를 해본다고 했잖아요. 그런데 CCTV 자료는 CCTV_Seoul에 있고, 서울인구 자료는 pop_Seoul에 있잖아요. 그러면 나중에 이 데이터를 합치게 될 거에요. 데이터를 합칠 때는 기준이 있는데, 저는 그 기준을 지역구로 하고 싶어요. 그래서 자치구를 '구별'로 바꿔주고 계, 계1, 계2, 65세이상고령자를 변수의 특성이 더 잘 나타나는 데이터로 바꾸어주고 싶어요. 그러면 아래와 같이 하면 됩니다!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>구별</th>\n",
       "      <th>인구수</th>\n",
       "      <th>한국인</th>\n",
       "      <th>외국인</th>\n",
       "      <th>고령자</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>합계</td>\n",
       "      <td>10197604.0</td>\n",
       "      <td>9926968.0</td>\n",
       "      <td>270636.0</td>\n",
       "      <td>1321458.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>종로구</td>\n",
       "      <td>162820.0</td>\n",
       "      <td>153589.0</td>\n",
       "      <td>9231.0</td>\n",
       "      <td>25425.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>중구</td>\n",
       "      <td>133240.0</td>\n",
       "      <td>124312.0</td>\n",
       "      <td>8928.0</td>\n",
       "      <td>20764.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>용산구</td>\n",
       "      <td>244203.0</td>\n",
       "      <td>229456.0</td>\n",
       "      <td>14747.0</td>\n",
       "      <td>36231.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>성동구</td>\n",
       "      <td>311244.0</td>\n",
       "      <td>303380.0</td>\n",
       "      <td>7864.0</td>\n",
       "      <td>39997.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    구별         인구수        한국인       외국인        고령자\n",
       "0   합계  10197604.0  9926968.0  270636.0  1321458.0\n",
       "1  종로구    162820.0   153589.0    9231.0    25425.0\n",
       "2   중구    133240.0   124312.0    8928.0    20764.0\n",
       "3  용산구    244203.0   229456.0   14747.0    36231.0\n",
       "4  성동구    311244.0   303380.0    7864.0    39997.0"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pop_Seoul = pop_Seoul.rename(columns={pop_Seoul.columns[0] : '구별',\n",
    "                                      pop_Seoul.columns[1] : '인구수',\n",
    "                                      pop_Seoul.columns[2] : '한국인',\n",
    "                                      pop_Seoul.columns[3] : '외국인',\n",
    "                                      pop_Seoul.columns[4] : '고령자'})\n",
    "pop_Seoul.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "책에는 이와 다르게 inplace 옵션을 사용하는데요, 저는 제 개인적인 취향은 inplace보다 이렇게 변수를 다시 정의해주는게 더 좋은 것 같더라구요. 본인이 원하는대로 하시면 됩니다. \n",
    "\n",
    "그럼 이렇게 CSV와 엑셀 파일에 들어있는 데이터를 파이썬으로 불러오고, 정리까지 해보았습니다! \n",
    "\n",
    "### 2.4 탐색적 데이터 분석하기\n",
    "\n",
    "그럼 데이터를 잘 불러왔으니 데이터를 한번 살펴봐야겠죠? "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>구별</th>\n",
       "      <th>소계</th>\n",
       "      <th>2013년도 이전</th>\n",
       "      <th>2014년</th>\n",
       "      <th>2015년</th>\n",
       "      <th>2016년</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>강남구</td>\n",
       "      <td>2780</td>\n",
       "      <td>1292</td>\n",
       "      <td>430</td>\n",
       "      <td>584</td>\n",
       "      <td>932</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>강동구</td>\n",
       "      <td>773</td>\n",
       "      <td>379</td>\n",
       "      <td>99</td>\n",
       "      <td>155</td>\n",
       "      <td>377</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>강북구</td>\n",
       "      <td>748</td>\n",
       "      <td>369</td>\n",
       "      <td>120</td>\n",
       "      <td>138</td>\n",
       "      <td>204</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>강서구</td>\n",
       "      <td>884</td>\n",
       "      <td>388</td>\n",
       "      <td>258</td>\n",
       "      <td>184</td>\n",
       "      <td>81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>관악구</td>\n",
       "      <td>1496</td>\n",
       "      <td>846</td>\n",
       "      <td>260</td>\n",
       "      <td>390</td>\n",
       "      <td>613</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    구별    소계  2013년도 이전  2014년  2015년  2016년\n",
       "0  강남구  2780       1292    430    584    932\n",
       "1  강동구   773        379     99    155    377\n",
       "2  강북구   748        369    120    138    204\n",
       "3  강서구   884        388    258    184     81\n",
       "4  관악구  1496        846    260    390    613"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "CCTV_Seoul.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 데이터 정렬하기 \n",
    "\n",
    "CCTV가 어디에 많이 있고 없는지를 알고 싶어요. 그러면 '소계'가 CCTV 수니까 이를 정렬시켜볼게요.\n",
    "\n",
    "- .sort_values(by='기준 칼럼', ascending=True): 기준 칼럼을 기준으로 오름차순 정렬하기 "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>구별</th>\n",
       "      <th>소계</th>\n",
       "      <th>2013년도 이전</th>\n",
       "      <th>2014년</th>\n",
       "      <th>2015년</th>\n",
       "      <th>2016년</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>도봉구</td>\n",
       "      <td>485</td>\n",
       "      <td>238</td>\n",
       "      <td>159</td>\n",
       "      <td>42</td>\n",
       "      <td>386</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12</td>\n",
       "      <td>마포구</td>\n",
       "      <td>574</td>\n",
       "      <td>314</td>\n",
       "      <td>118</td>\n",
       "      <td>169</td>\n",
       "      <td>379</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>송파구</td>\n",
       "      <td>618</td>\n",
       "      <td>529</td>\n",
       "      <td>21</td>\n",
       "      <td>68</td>\n",
       "      <td>463</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>24</td>\n",
       "      <td>중랑구</td>\n",
       "      <td>660</td>\n",
       "      <td>509</td>\n",
       "      <td>121</td>\n",
       "      <td>177</td>\n",
       "      <td>109</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>23</td>\n",
       "      <td>중구</td>\n",
       "      <td>671</td>\n",
       "      <td>413</td>\n",
       "      <td>190</td>\n",
       "      <td>72</td>\n",
       "      <td>348</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     구별   소계  2013년도 이전  2014년  2015년  2016년\n",
       "9   도봉구  485        238    159     42    386\n",
       "12  마포구  574        314    118    169    379\n",
       "17  송파구  618        529     21     68    463\n",
       "24  중랑구  660        509    121    177    109\n",
       "23   중구  671        413    190     72    348"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "CCTV_Seoul.sort_values(by='소계', ascending=True).head(5) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "데이터를 살펴보는 것이니 따로 변수에 담아줄 필요는 없어요. 보니까 도봉구, 마포구, 송파구, 중랑구, 중구가 CCTV 수로 하위 TOP5네요. 그리고 옆에 인덱스를 보면 구별에 따라서 인덱스가 적혀져 있어요. 이렇게 판다스는 고유한 인덱스가 변하지 않고 따라온다는 사실 덕분에 편하게 데이터를 가공할 수 있어요. \n",
    "\n",
    "그리고 강남 3구 중의 하나인 송파구가 CCTV가 가장 적은 구 중의 하나라는 것인데, 강남 3구면 송파구에는 예산이 다른 구보다는 있지 않을까?(정확히 몰라요, 걍 생각해보는 것이라 검증을 해봐야지요) 👉 CCTV를 많이 설치할 수 있을 텐데..? 등의 이런 생각을 해볼 수 있겠죠. 이렇게 생각해 보는 과정이 **탐색적 데이터 분석** 과정입니다. \n",
    "\n",
    "반대로 CCTV가 많이 설치되어 있는 TOP5 구들을 봅시다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>구별</th>\n",
       "      <th>소계</th>\n",
       "      <th>2013년도 이전</th>\n",
       "      <th>2014년</th>\n",
       "      <th>2015년</th>\n",
       "      <th>2016년</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>강남구</td>\n",
       "      <td>2780</td>\n",
       "      <td>1292</td>\n",
       "      <td>430</td>\n",
       "      <td>584</td>\n",
       "      <td>932</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>양천구</td>\n",
       "      <td>2034</td>\n",
       "      <td>1843</td>\n",
       "      <td>142</td>\n",
       "      <td>30</td>\n",
       "      <td>467</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14</td>\n",
       "      <td>서초구</td>\n",
       "      <td>1930</td>\n",
       "      <td>1406</td>\n",
       "      <td>157</td>\n",
       "      <td>336</td>\n",
       "      <td>398</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>21</td>\n",
       "      <td>은평구</td>\n",
       "      <td>1873</td>\n",
       "      <td>1138</td>\n",
       "      <td>224</td>\n",
       "      <td>278</td>\n",
       "      <td>468</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>20</td>\n",
       "      <td>용산구</td>\n",
       "      <td>1624</td>\n",
       "      <td>1368</td>\n",
       "      <td>218</td>\n",
       "      <td>112</td>\n",
       "      <td>398</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     구별    소계  2013년도 이전  2014년  2015년  2016년\n",
       "0   강남구  2780       1292    430    584    932\n",
       "18  양천구  2034       1843    142     30    467\n",
       "14  서초구  1930       1406    157    336    398\n",
       "21  은평구  1873       1138    224    278    468\n",
       "20  용산구  1624       1368    218    112    398"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "CCTV_Seoul.sort_values(by='소계', ascending=False).head(5) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "상위 5구로는 강남구, 양천구, 서초구, 은평구, 용산구가 나오게 됩니다. 강남구와 서초구가 탑에 속해있으니, 내가 세운 가설이 맞을 수도 있겠군 등 이렇게 생각해보는 거죠..! \n",
    "\n",
    "#### 데이터 연산하기 \n",
    "\n",
    "그 다음 살펴볼 것이 **3년간 CCTV 증가율**이에요. CCTV 데이터에 연도 데이터도 함께 있으니까 최근 증가율도 살펴볼 수 있어요. \n",
    "- (최근 3년간 CCTV 증가율) = (2014년 + 2015년 + 2016년) / (2013년도 이전)* 100 "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>구별</th>\n",
       "      <th>소계</th>\n",
       "      <th>2013년도 이전</th>\n",
       "      <th>2014년</th>\n",
       "      <th>2015년</th>\n",
       "      <th>2016년</th>\n",
       "      <th>최근증가율</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>22</td>\n",
       "      <td>종로구</td>\n",
       "      <td>1002</td>\n",
       "      <td>464</td>\n",
       "      <td>314</td>\n",
       "      <td>211</td>\n",
       "      <td>630</td>\n",
       "      <td>248.922414</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>도봉구</td>\n",
       "      <td>485</td>\n",
       "      <td>238</td>\n",
       "      <td>159</td>\n",
       "      <td>42</td>\n",
       "      <td>386</td>\n",
       "      <td>246.638655</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12</td>\n",
       "      <td>마포구</td>\n",
       "      <td>574</td>\n",
       "      <td>314</td>\n",
       "      <td>118</td>\n",
       "      <td>169</td>\n",
       "      <td>379</td>\n",
       "      <td>212.101911</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>노원구</td>\n",
       "      <td>1265</td>\n",
       "      <td>542</td>\n",
       "      <td>57</td>\n",
       "      <td>451</td>\n",
       "      <td>516</td>\n",
       "      <td>188.929889</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>강동구</td>\n",
       "      <td>773</td>\n",
       "      <td>379</td>\n",
       "      <td>99</td>\n",
       "      <td>155</td>\n",
       "      <td>377</td>\n",
       "      <td>166.490765</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     구별    소계  2013년도 이전  2014년  2015년  2016년       최근증가율\n",
       "22  종로구  1002        464    314    211    630  248.922414\n",
       "9   도봉구   485        238    159     42    386  246.638655\n",
       "12  마포구   574        314    118    169    379  212.101911\n",
       "8   노원구  1265        542     57    451    516  188.929889\n",
       "1   강동구   773        379     99    155    377  166.490765"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 최근 3년간 CCTV 설치 증가율\n",
    "CCTV_Seoul['최근증가율'] = ((CCTV_Seoul['2014년'] + CCTV_Seoul['2015년'] \n",
    "                       + CCTV_Seoul['2016년']) / CCTV_Seoul['2013년도 이전']) * 100\n",
    "CCTV_Seoul.sort_values(by='최근증가율', ascending=False).head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "엇, 보니까 종로구가 정말 많이 증가했네요. 도봉구, 마포구도 그렇구요. 그럼 왜 이렇게 CCTV 설치율이 최근 사이에 늘어났을까도 사회학적으로 고민해볼 수 있겠죠! \n",
    "\n",
    "#### 데이터 확인하기 \n",
    "- .unique(): 한번 이상 나타난 데이터를 알려줘요. \n",
    "\n",
    "구별 칼럼에서 한 번 이상 나온 값이 무엇이 있을까 볼게요."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['합계', '종로구', '중구', '용산구', '성동구', '광진구', '동대문구', '중랑구', '성북구',\n",
       "       '강북구', '도봉구', '노원구', '은평구', '서대문구', '마포구', '양천구', '강서구', '구로구',\n",
       "       '금천구', '영등포구', '동작구', '관악구', '서초구', '강남구', '송파구', '강동구', nan],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pop_Seoul['구별'].unique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 결측치 처리하기\n",
    "\n",
    "엇, 그런데 우리가 넘파이에서 배운 nan이 있어요. 이런 것을 결측치라고 합니다. **머신러닝을 돌릴 때나 데이터 분석을 할 때나 결측치가 있으면 예측이나 분석의 정확도가 떨어지고, 의도하지 않은 결과가 나올 수 있어요.** 그래서 이렇게 결측치가 있는 경우에는 NaN이 어디에 있는지를 알아서 **적절하게 처리**를 해줘야 해요."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>구별</th>\n",
       "      <th>인구수</th>\n",
       "      <th>한국인</th>\n",
       "      <th>외국인</th>\n",
       "      <th>고령자</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>26</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     구별  인구수  한국인  외국인  고령자\n",
       "26  NaN  NaN  NaN  NaN  NaN"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pop_Seoul[pop_Seoul['구별'].isnull()] "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>구별</th>\n",
       "      <th>인구수</th>\n",
       "      <th>한국인</th>\n",
       "      <th>외국인</th>\n",
       "      <th>고령자</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>합계</td>\n",
       "      <td>10197604.0</td>\n",
       "      <td>9926968.0</td>\n",
       "      <td>270636.0</td>\n",
       "      <td>1321458.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>종로구</td>\n",
       "      <td>162820.0</td>\n",
       "      <td>153589.0</td>\n",
       "      <td>9231.0</td>\n",
       "      <td>25425.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>중구</td>\n",
       "      <td>133240.0</td>\n",
       "      <td>124312.0</td>\n",
       "      <td>8928.0</td>\n",
       "      <td>20764.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>용산구</td>\n",
       "      <td>244203.0</td>\n",
       "      <td>229456.0</td>\n",
       "      <td>14747.0</td>\n",
       "      <td>36231.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>성동구</td>\n",
       "      <td>311244.0</td>\n",
       "      <td>303380.0</td>\n",
       "      <td>7864.0</td>\n",
       "      <td>39997.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    구별         인구수        한국인       외국인        고령자\n",
       "0   합계  10197604.0  9926968.0  270636.0  1321458.0\n",
       "1  종로구    162820.0   153589.0    9231.0    25425.0\n",
       "2   중구    133240.0   124312.0    8928.0    20764.0\n",
       "3  용산구    244203.0   229456.0   14747.0    36231.0\n",
       "4  성동구    311244.0   303380.0    7864.0    39997.0"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pop_Seoul.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "아하, 이게 인덱스 26에만 있구나. 그리고 합계가 필요없어서 인덱스가 0, 26인 row를 지우려고 해요.\n",
    "저는 결측치를 처리하는 방법으로 저는 이 데이터를 데이터에서 제거하기로 한거죠."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>구별</th>\n",
       "      <th>인구수</th>\n",
       "      <th>한국인</th>\n",
       "      <th>외국인</th>\n",
       "      <th>고령자</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>종로구</td>\n",
       "      <td>162820.0</td>\n",
       "      <td>153589.0</td>\n",
       "      <td>9231.0</td>\n",
       "      <td>25425.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>중구</td>\n",
       "      <td>133240.0</td>\n",
       "      <td>124312.0</td>\n",
       "      <td>8928.0</td>\n",
       "      <td>20764.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>용산구</td>\n",
       "      <td>244203.0</td>\n",
       "      <td>229456.0</td>\n",
       "      <td>14747.0</td>\n",
       "      <td>36231.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>성동구</td>\n",
       "      <td>311244.0</td>\n",
       "      <td>303380.0</td>\n",
       "      <td>7864.0</td>\n",
       "      <td>39997.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>광진구</td>\n",
       "      <td>372164.0</td>\n",
       "      <td>357211.0</td>\n",
       "      <td>14953.0</td>\n",
       "      <td>42214.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>동대문구</td>\n",
       "      <td>369496.0</td>\n",
       "      <td>354079.0</td>\n",
       "      <td>15417.0</td>\n",
       "      <td>54173.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>중랑구</td>\n",
       "      <td>414503.0</td>\n",
       "      <td>409882.0</td>\n",
       "      <td>4621.0</td>\n",
       "      <td>56774.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>성북구</td>\n",
       "      <td>461260.0</td>\n",
       "      <td>449773.0</td>\n",
       "      <td>11487.0</td>\n",
       "      <td>64692.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>강북구</td>\n",
       "      <td>330192.0</td>\n",
       "      <td>326686.0</td>\n",
       "      <td>3506.0</td>\n",
       "      <td>54813.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>10</td>\n",
       "      <td>도봉구</td>\n",
       "      <td>348646.0</td>\n",
       "      <td>346629.0</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>51312.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11</td>\n",
       "      <td>노원구</td>\n",
       "      <td>569384.0</td>\n",
       "      <td>565565.0</td>\n",
       "      <td>3819.0</td>\n",
       "      <td>71941.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12</td>\n",
       "      <td>은평구</td>\n",
       "      <td>494388.0</td>\n",
       "      <td>489943.0</td>\n",
       "      <td>4445.0</td>\n",
       "      <td>72334.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>13</td>\n",
       "      <td>서대문구</td>\n",
       "      <td>327163.0</td>\n",
       "      <td>314982.0</td>\n",
       "      <td>12181.0</td>\n",
       "      <td>48161.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14</td>\n",
       "      <td>마포구</td>\n",
       "      <td>389649.0</td>\n",
       "      <td>378566.0</td>\n",
       "      <td>11083.0</td>\n",
       "      <td>48765.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15</td>\n",
       "      <td>양천구</td>\n",
       "      <td>479978.0</td>\n",
       "      <td>475949.0</td>\n",
       "      <td>4029.0</td>\n",
       "      <td>52975.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>강서구</td>\n",
       "      <td>603772.0</td>\n",
       "      <td>597248.0</td>\n",
       "      <td>6524.0</td>\n",
       "      <td>72548.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>구로구</td>\n",
       "      <td>447874.0</td>\n",
       "      <td>416487.0</td>\n",
       "      <td>31387.0</td>\n",
       "      <td>56833.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>금천구</td>\n",
       "      <td>255082.0</td>\n",
       "      <td>236353.0</td>\n",
       "      <td>18729.0</td>\n",
       "      <td>32970.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>19</td>\n",
       "      <td>영등포구</td>\n",
       "      <td>402985.0</td>\n",
       "      <td>368072.0</td>\n",
       "      <td>34913.0</td>\n",
       "      <td>52413.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>20</td>\n",
       "      <td>동작구</td>\n",
       "      <td>412520.0</td>\n",
       "      <td>400456.0</td>\n",
       "      <td>12064.0</td>\n",
       "      <td>56013.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>21</td>\n",
       "      <td>관악구</td>\n",
       "      <td>525515.0</td>\n",
       "      <td>507203.0</td>\n",
       "      <td>18312.0</td>\n",
       "      <td>68082.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>22</td>\n",
       "      <td>서초구</td>\n",
       "      <td>450310.0</td>\n",
       "      <td>445994.0</td>\n",
       "      <td>4316.0</td>\n",
       "      <td>51733.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>23</td>\n",
       "      <td>강남구</td>\n",
       "      <td>570500.0</td>\n",
       "      <td>565550.0</td>\n",
       "      <td>4950.0</td>\n",
       "      <td>63167.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>24</td>\n",
       "      <td>송파구</td>\n",
       "      <td>667483.0</td>\n",
       "      <td>660584.0</td>\n",
       "      <td>6899.0</td>\n",
       "      <td>72506.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>25</td>\n",
       "      <td>강동구</td>\n",
       "      <td>453233.0</td>\n",
       "      <td>449019.0</td>\n",
       "      <td>4214.0</td>\n",
       "      <td>54622.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      구별       인구수       한국인      외국인      고령자\n",
       "1    종로구  162820.0  153589.0   9231.0  25425.0\n",
       "2     중구  133240.0  124312.0   8928.0  20764.0\n",
       "3    용산구  244203.0  229456.0  14747.0  36231.0\n",
       "4    성동구  311244.0  303380.0   7864.0  39997.0\n",
       "5    광진구  372164.0  357211.0  14953.0  42214.0\n",
       "6   동대문구  369496.0  354079.0  15417.0  54173.0\n",
       "7    중랑구  414503.0  409882.0   4621.0  56774.0\n",
       "8    성북구  461260.0  449773.0  11487.0  64692.0\n",
       "9    강북구  330192.0  326686.0   3506.0  54813.0\n",
       "10   도봉구  348646.0  346629.0   2017.0  51312.0\n",
       "11   노원구  569384.0  565565.0   3819.0  71941.0\n",
       "12   은평구  494388.0  489943.0   4445.0  72334.0\n",
       "13  서대문구  327163.0  314982.0  12181.0  48161.0\n",
       "14   마포구  389649.0  378566.0  11083.0  48765.0\n",
       "15   양천구  479978.0  475949.0   4029.0  52975.0\n",
       "16   강서구  603772.0  597248.0   6524.0  72548.0\n",
       "17   구로구  447874.0  416487.0  31387.0  56833.0\n",
       "18   금천구  255082.0  236353.0  18729.0  32970.0\n",
       "19  영등포구  402985.0  368072.0  34913.0  52413.0\n",
       "20   동작구  412520.0  400456.0  12064.0  56013.0\n",
       "21   관악구  525515.0  507203.0  18312.0  68082.0\n",
       "22   서초구  450310.0  445994.0   4316.0  51733.0\n",
       "23   강남구  570500.0  565550.0   4950.0  63167.0\n",
       "24   송파구  667483.0  660584.0   6899.0  72506.0\n",
       "25   강동구  453233.0  449019.0   4214.0  54622.0"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pop_Seoul = pop_Seoul.drop([0, 26])\n",
    "pop_Seoul.head(25)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "저는 inplace를 싫어해서 그냥 다시 변수를 정의해주었습니다. 이렇게 되니까 잘 제거되었는지 보고싶죠? "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>구별</th>\n",
       "      <th>인구수</th>\n",
       "      <th>한국인</th>\n",
       "      <th>외국인</th>\n",
       "      <th>고령자</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [구별, 인구수, 한국인, 외국인, 고령자]\n",
       "Index: []"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pop_Seoul[pop_Seoul['구별'].isnull()] "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "오, 없네요! 결측치를 잘 제거했습니다.\n",
    "\n",
    "#### 데이터 탐색하기 \n",
    "\n",
    "이 작가는 고령자비율과 외국인비율을 계산했어요. 고령자비율과 외국인비율이 CCTV 설치율과 영향이 있는가 보고 싶은가봐요. 그러면 우리도 한번 고령자비율과 외국인 비율을 계산해볼게요!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>구별</th>\n",
       "      <th>인구수</th>\n",
       "      <th>한국인</th>\n",
       "      <th>외국인</th>\n",
       "      <th>고령자</th>\n",
       "      <th>외국인비율</th>\n",
       "      <th>고령자비율</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>종로구</td>\n",
       "      <td>162820.0</td>\n",
       "      <td>153589.0</td>\n",
       "      <td>9231.0</td>\n",
       "      <td>25425.0</td>\n",
       "      <td>5.669451</td>\n",
       "      <td>15.615404</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>중구</td>\n",
       "      <td>133240.0</td>\n",
       "      <td>124312.0</td>\n",
       "      <td>8928.0</td>\n",
       "      <td>20764.0</td>\n",
       "      <td>6.700690</td>\n",
       "      <td>15.583909</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>용산구</td>\n",
       "      <td>244203.0</td>\n",
       "      <td>229456.0</td>\n",
       "      <td>14747.0</td>\n",
       "      <td>36231.0</td>\n",
       "      <td>6.038828</td>\n",
       "      <td>14.836427</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>성동구</td>\n",
       "      <td>311244.0</td>\n",
       "      <td>303380.0</td>\n",
       "      <td>7864.0</td>\n",
       "      <td>39997.0</td>\n",
       "      <td>2.526635</td>\n",
       "      <td>12.850689</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>광진구</td>\n",
       "      <td>372164.0</td>\n",
       "      <td>357211.0</td>\n",
       "      <td>14953.0</td>\n",
       "      <td>42214.0</td>\n",
       "      <td>4.017852</td>\n",
       "      <td>11.342849</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    구별       인구수       한국인      외국인      고령자     외국인비율      고령자비율\n",
       "1  종로구  162820.0  153589.0   9231.0  25425.0  5.669451  15.615404\n",
       "2   중구  133240.0  124312.0   8928.0  20764.0  6.700690  15.583909\n",
       "3  용산구  244203.0  229456.0  14747.0  36231.0  6.038828  14.836427\n",
       "4  성동구  311244.0  303380.0   7864.0  39997.0  2.526635  12.850689\n",
       "5  광진구  372164.0  357211.0  14953.0  42214.0  4.017852  11.342849"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pop_Seoul['외국인비율'] = pop_Seoul['외국인'] / pop_Seoul['인구수'] * 100 #외국인비율 칼럼 생성\n",
    "pop_Seoul['고령자비율'] = pop_Seoul['고령자'] / pop_Seoul['인구수'] * 100 #고령자비율 칼럼 생성\n",
    "pop_Seoul.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "이렇게 하면 데이터를 잘 알수 없으니까 또 오름차순과 내림차순을 해서 알아보도록 하죠."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>구별</th>\n",
       "      <th>인구수</th>\n",
       "      <th>한국인</th>\n",
       "      <th>외국인</th>\n",
       "      <th>고령자</th>\n",
       "      <th>외국인비율</th>\n",
       "      <th>고령자비율</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>24</td>\n",
       "      <td>송파구</td>\n",
       "      <td>667483.0</td>\n",
       "      <td>660584.0</td>\n",
       "      <td>6899.0</td>\n",
       "      <td>72506.0</td>\n",
       "      <td>1.033584</td>\n",
       "      <td>10.862599</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>강서구</td>\n",
       "      <td>603772.0</td>\n",
       "      <td>597248.0</td>\n",
       "      <td>6524.0</td>\n",
       "      <td>72548.0</td>\n",
       "      <td>1.080540</td>\n",
       "      <td>12.015794</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>23</td>\n",
       "      <td>강남구</td>\n",
       "      <td>570500.0</td>\n",
       "      <td>565550.0</td>\n",
       "      <td>4950.0</td>\n",
       "      <td>63167.0</td>\n",
       "      <td>0.867660</td>\n",
       "      <td>11.072217</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11</td>\n",
       "      <td>노원구</td>\n",
       "      <td>569384.0</td>\n",
       "      <td>565565.0</td>\n",
       "      <td>3819.0</td>\n",
       "      <td>71941.0</td>\n",
       "      <td>0.670725</td>\n",
       "      <td>12.634883</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>21</td>\n",
       "      <td>관악구</td>\n",
       "      <td>525515.0</td>\n",
       "      <td>507203.0</td>\n",
       "      <td>18312.0</td>\n",
       "      <td>68082.0</td>\n",
       "      <td>3.484582</td>\n",
       "      <td>12.955291</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     구별       인구수       한국인      외국인      고령자     외국인비율      고령자비율\n",
       "24  송파구  667483.0  660584.0   6899.0  72506.0  1.033584  10.862599\n",
       "16  강서구  603772.0  597248.0   6524.0  72548.0  1.080540  12.015794\n",
       "23  강남구  570500.0  565550.0   4950.0  63167.0  0.867660  11.072217\n",
       "11  노원구  569384.0  565565.0   3819.0  71941.0  0.670725  12.634883\n",
       "21  관악구  525515.0  507203.0  18312.0  68082.0  3.484582  12.955291"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pop_Seoul.sort_values(by='인구수', ascending=False).head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "인구수를 기준으로 내림차순을 해주면, 송파구가 인구가 제일 많네요! "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>구별</th>\n",
       "      <th>인구수</th>\n",
       "      <th>한국인</th>\n",
       "      <th>외국인</th>\n",
       "      <th>고령자</th>\n",
       "      <th>외국인비율</th>\n",
       "      <th>고령자비율</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>19</td>\n",
       "      <td>영등포구</td>\n",
       "      <td>402985.0</td>\n",
       "      <td>368072.0</td>\n",
       "      <td>34913.0</td>\n",
       "      <td>52413.0</td>\n",
       "      <td>8.663598</td>\n",
       "      <td>13.006191</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>구로구</td>\n",
       "      <td>447874.0</td>\n",
       "      <td>416487.0</td>\n",
       "      <td>31387.0</td>\n",
       "      <td>56833.0</td>\n",
       "      <td>7.007998</td>\n",
       "      <td>12.689506</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>금천구</td>\n",
       "      <td>255082.0</td>\n",
       "      <td>236353.0</td>\n",
       "      <td>18729.0</td>\n",
       "      <td>32970.0</td>\n",
       "      <td>7.342345</td>\n",
       "      <td>12.925255</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>21</td>\n",
       "      <td>관악구</td>\n",
       "      <td>525515.0</td>\n",
       "      <td>507203.0</td>\n",
       "      <td>18312.0</td>\n",
       "      <td>68082.0</td>\n",
       "      <td>3.484582</td>\n",
       "      <td>12.955291</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>동대문구</td>\n",
       "      <td>369496.0</td>\n",
       "      <td>354079.0</td>\n",
       "      <td>15417.0</td>\n",
       "      <td>54173.0</td>\n",
       "      <td>4.172440</td>\n",
       "      <td>14.661322</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      구별       인구수       한국인      외국인      고령자     외국인비율      고령자비율\n",
       "19  영등포구  402985.0  368072.0  34913.0  52413.0  8.663598  13.006191\n",
       "17   구로구  447874.0  416487.0  31387.0  56833.0  7.007998  12.689506\n",
       "18   금천구  255082.0  236353.0  18729.0  32970.0  7.342345  12.925255\n",
       "21   관악구  525515.0  507203.0  18312.0  68082.0  3.484582  12.955291\n",
       "6   동대문구  369496.0  354079.0  15417.0  54173.0  4.172440  14.661322"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pop_Seoul.sort_values(by='외국인', ascending=False).head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "외국인을 기준으로 하면, 영등포구에 외국인이 많이 살고 있군요!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>구별</th>\n",
       "      <th>인구수</th>\n",
       "      <th>한국인</th>\n",
       "      <th>외국인</th>\n",
       "      <th>고령자</th>\n",
       "      <th>외국인비율</th>\n",
       "      <th>고령자비율</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>19</td>\n",
       "      <td>영등포구</td>\n",
       "      <td>402985.0</td>\n",
       "      <td>368072.0</td>\n",
       "      <td>34913.0</td>\n",
       "      <td>52413.0</td>\n",
       "      <td>8.663598</td>\n",
       "      <td>13.006191</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>금천구</td>\n",
       "      <td>255082.0</td>\n",
       "      <td>236353.0</td>\n",
       "      <td>18729.0</td>\n",
       "      <td>32970.0</td>\n",
       "      <td>7.342345</td>\n",
       "      <td>12.925255</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>구로구</td>\n",
       "      <td>447874.0</td>\n",
       "      <td>416487.0</td>\n",
       "      <td>31387.0</td>\n",
       "      <td>56833.0</td>\n",
       "      <td>7.007998</td>\n",
       "      <td>12.689506</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>중구</td>\n",
       "      <td>133240.0</td>\n",
       "      <td>124312.0</td>\n",
       "      <td>8928.0</td>\n",
       "      <td>20764.0</td>\n",
       "      <td>6.700690</td>\n",
       "      <td>15.583909</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>용산구</td>\n",
       "      <td>244203.0</td>\n",
       "      <td>229456.0</td>\n",
       "      <td>14747.0</td>\n",
       "      <td>36231.0</td>\n",
       "      <td>6.038828</td>\n",
       "      <td>14.836427</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      구별       인구수       한국인      외국인      고령자     외국인비율      고령자비율\n",
       "19  영등포구  402985.0  368072.0  34913.0  52413.0  8.663598  13.006191\n",
       "18   금천구  255082.0  236353.0  18729.0  32970.0  7.342345  12.925255\n",
       "17   구로구  447874.0  416487.0  31387.0  56833.0  7.007998  12.689506\n",
       "2     중구  133240.0  124312.0   8928.0  20764.0  6.700690  15.583909\n",
       "3    용산구  244203.0  229456.0  14747.0  36231.0  6.038828  14.836427"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pop_Seoul.sort_values(by='외국인비율', ascending=False).head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "외국인비율로도 보니까 영등포구가 제일이네요."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>구별</th>\n",
       "      <th>인구수</th>\n",
       "      <th>한국인</th>\n",
       "      <th>외국인</th>\n",
       "      <th>고령자</th>\n",
       "      <th>외국인비율</th>\n",
       "      <th>고령자비율</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>강서구</td>\n",
       "      <td>603772.0</td>\n",
       "      <td>597248.0</td>\n",
       "      <td>6524.0</td>\n",
       "      <td>72548.0</td>\n",
       "      <td>1.080540</td>\n",
       "      <td>12.015794</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>24</td>\n",
       "      <td>송파구</td>\n",
       "      <td>667483.0</td>\n",
       "      <td>660584.0</td>\n",
       "      <td>6899.0</td>\n",
       "      <td>72506.0</td>\n",
       "      <td>1.033584</td>\n",
       "      <td>10.862599</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12</td>\n",
       "      <td>은평구</td>\n",
       "      <td>494388.0</td>\n",
       "      <td>489943.0</td>\n",
       "      <td>4445.0</td>\n",
       "      <td>72334.0</td>\n",
       "      <td>0.899091</td>\n",
       "      <td>14.631019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11</td>\n",
       "      <td>노원구</td>\n",
       "      <td>569384.0</td>\n",
       "      <td>565565.0</td>\n",
       "      <td>3819.0</td>\n",
       "      <td>71941.0</td>\n",
       "      <td>0.670725</td>\n",
       "      <td>12.634883</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>21</td>\n",
       "      <td>관악구</td>\n",
       "      <td>525515.0</td>\n",
       "      <td>507203.0</td>\n",
       "      <td>18312.0</td>\n",
       "      <td>68082.0</td>\n",
       "      <td>3.484582</td>\n",
       "      <td>12.955291</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     구별       인구수       한국인      외국인      고령자     외국인비율      고령자비율\n",
       "16  강서구  603772.0  597248.0   6524.0  72548.0  1.080540  12.015794\n",
       "24  송파구  667483.0  660584.0   6899.0  72506.0  1.033584  10.862599\n",
       "12  은평구  494388.0  489943.0   4445.0  72334.0  0.899091  14.631019\n",
       "11  노원구  569384.0  565565.0   3819.0  71941.0  0.670725  12.634883\n",
       "21  관악구  525515.0  507203.0  18312.0  68082.0  3.484582  12.955291"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pop_Seoul.sort_values(by='고령자', ascending=False).head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "고령자 수자체로만 보면 강서구에 제일 많습니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>구별</th>\n",
       "      <th>인구수</th>\n",
       "      <th>한국인</th>\n",
       "      <th>외국인</th>\n",
       "      <th>고령자</th>\n",
       "      <th>외국인비율</th>\n",
       "      <th>고령자비율</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>강북구</td>\n",
       "      <td>330192.0</td>\n",
       "      <td>326686.0</td>\n",
       "      <td>3506.0</td>\n",
       "      <td>54813.0</td>\n",
       "      <td>1.061806</td>\n",
       "      <td>16.600342</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>종로구</td>\n",
       "      <td>162820.0</td>\n",
       "      <td>153589.0</td>\n",
       "      <td>9231.0</td>\n",
       "      <td>25425.0</td>\n",
       "      <td>5.669451</td>\n",
       "      <td>15.615404</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>중구</td>\n",
       "      <td>133240.0</td>\n",
       "      <td>124312.0</td>\n",
       "      <td>8928.0</td>\n",
       "      <td>20764.0</td>\n",
       "      <td>6.700690</td>\n",
       "      <td>15.583909</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>용산구</td>\n",
       "      <td>244203.0</td>\n",
       "      <td>229456.0</td>\n",
       "      <td>14747.0</td>\n",
       "      <td>36231.0</td>\n",
       "      <td>6.038828</td>\n",
       "      <td>14.836427</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>13</td>\n",
       "      <td>서대문구</td>\n",
       "      <td>327163.0</td>\n",
       "      <td>314982.0</td>\n",
       "      <td>12181.0</td>\n",
       "      <td>48161.0</td>\n",
       "      <td>3.723221</td>\n",
       "      <td>14.720797</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      구별       인구수       한국인      외국인      고령자     외국인비율      고령자비율\n",
       "9    강북구  330192.0  326686.0   3506.0  54813.0  1.061806  16.600342\n",
       "1    종로구  162820.0  153589.0   9231.0  25425.0  5.669451  15.615404\n",
       "2     중구  133240.0  124312.0   8928.0  20764.0  6.700690  15.583909\n",
       "3    용산구  244203.0  229456.0  14747.0  36231.0  6.038828  14.836427\n",
       "13  서대문구  327163.0  314982.0  12181.0  48161.0  3.723221  14.720797"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pop_Seoul.sort_values(by='고령자비율', ascending=False).head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "고령자비율로만 보면 강북구에 많네요! \n",
    "\n",
    "#### 데이터 병합하기\n",
    "\n",
    "이렇게 데이터를 하나씩 보면서 인구 데이터와 CCTV 데이터의 특징을 살펴봤어요. 그런데 아직 우리는 인구 대비 CCTV 현황을 보지 못했죠? 이 데이터는 따로 존재하기 때문에 두 데이터를 병합해야 해요. 판다스에서 **merge()**를 배웠으니, 이것을 사용하여 데이터를 옆으로 병합해 보겠습니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>구별</th>\n",
       "      <th>소계</th>\n",
       "      <th>2013년도 이전</th>\n",
       "      <th>2014년</th>\n",
       "      <th>2015년</th>\n",
       "      <th>2016년</th>\n",
       "      <th>최근증가율</th>\n",
       "      <th>인구수</th>\n",
       "      <th>한국인</th>\n",
       "      <th>외국인</th>\n",
       "      <th>고령자</th>\n",
       "      <th>외국인비율</th>\n",
       "      <th>고령자비율</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>강남구</td>\n",
       "      <td>2780</td>\n",
       "      <td>1292</td>\n",
       "      <td>430</td>\n",
       "      <td>584</td>\n",
       "      <td>932</td>\n",
       "      <td>150.619195</td>\n",
       "      <td>570500.0</td>\n",
       "      <td>565550.0</td>\n",
       "      <td>4950.0</td>\n",
       "      <td>63167.0</td>\n",
       "      <td>0.867660</td>\n",
       "      <td>11.072217</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>강동구</td>\n",
       "      <td>773</td>\n",
       "      <td>379</td>\n",
       "      <td>99</td>\n",
       "      <td>155</td>\n",
       "      <td>377</td>\n",
       "      <td>166.490765</td>\n",
       "      <td>453233.0</td>\n",
       "      <td>449019.0</td>\n",
       "      <td>4214.0</td>\n",
       "      <td>54622.0</td>\n",
       "      <td>0.929765</td>\n",
       "      <td>12.051638</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>강북구</td>\n",
       "      <td>748</td>\n",
       "      <td>369</td>\n",
       "      <td>120</td>\n",
       "      <td>138</td>\n",
       "      <td>204</td>\n",
       "      <td>125.203252</td>\n",
       "      <td>330192.0</td>\n",
       "      <td>326686.0</td>\n",
       "      <td>3506.0</td>\n",
       "      <td>54813.0</td>\n",
       "      <td>1.061806</td>\n",
       "      <td>16.600342</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>강서구</td>\n",
       "      <td>884</td>\n",
       "      <td>388</td>\n",
       "      <td>258</td>\n",
       "      <td>184</td>\n",
       "      <td>81</td>\n",
       "      <td>134.793814</td>\n",
       "      <td>603772.0</td>\n",
       "      <td>597248.0</td>\n",
       "      <td>6524.0</td>\n",
       "      <td>72548.0</td>\n",
       "      <td>1.080540</td>\n",
       "      <td>12.015794</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>관악구</td>\n",
       "      <td>1496</td>\n",
       "      <td>846</td>\n",
       "      <td>260</td>\n",
       "      <td>390</td>\n",
       "      <td>613</td>\n",
       "      <td>149.290780</td>\n",
       "      <td>525515.0</td>\n",
       "      <td>507203.0</td>\n",
       "      <td>18312.0</td>\n",
       "      <td>68082.0</td>\n",
       "      <td>3.484582</td>\n",
       "      <td>12.955291</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    구별    소계  2013년도 이전  2014년  2015년  2016년       최근증가율       인구수       한국인  \\\n",
       "0  강남구  2780       1292    430    584    932  150.619195  570500.0  565550.0   \n",
       "1  강동구   773        379     99    155    377  166.490765  453233.0  449019.0   \n",
       "2  강북구   748        369    120    138    204  125.203252  330192.0  326686.0   \n",
       "3  강서구   884        388    258    184     81  134.793814  603772.0  597248.0   \n",
       "4  관악구  1496        846    260    390    613  149.290780  525515.0  507203.0   \n",
       "\n",
       "       외국인      고령자     외국인비율      고령자비율  \n",
       "0   4950.0  63167.0  0.867660  11.072217  \n",
       "1   4214.0  54622.0  0.929765  12.051638  \n",
       "2   3506.0  54813.0  1.061806  16.600342  \n",
       "3   6524.0  72548.0  1.080540  12.015794  \n",
       "4  18312.0  68082.0  3.484582  12.955291  "
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# data_result라는 새로운 변수에 담아줄게요! \n",
    "data_result = pd.merge(CCTV_Seoul, pop_Seoul, on='구별')\n",
    "data_result.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "이제 앞에서 column이름은 구별로 공통적으로 처리했는지 아시겠죠? 두 데이터의 공통 컬럼인 '구별'로 merge를 하면 공통된 key를 가진 데이터만 남게 됩니다. \n",
    "\n",
    "#### 데이터 삭제하기 \n",
    "\n",
    "그러면 이제 분석에 필요없는 칼럼들을 지워봅시다. 보니까 '2013년도 이전','2014년','2015년','2016년'은 필요하지 않네요! **drop()**을 사용하여 지워보겠습니다. (책에 나와있는 del 사용은 지양하길 권장합니다.) drop()은 기본이 axis=0이니까 column을 삭제하려면 axis=1로 해야겠네요!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>구별</th>\n",
       "      <th>소계</th>\n",
       "      <th>최근증가율</th>\n",
       "      <th>인구수</th>\n",
       "      <th>한국인</th>\n",
       "      <th>외국인</th>\n",
       "      <th>고령자</th>\n",
       "      <th>외국인비율</th>\n",
       "      <th>고령자비율</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>강남구</td>\n",
       "      <td>2780</td>\n",
       "      <td>150.619195</td>\n",
       "      <td>570500.0</td>\n",
       "      <td>565550.0</td>\n",
       "      <td>4950.0</td>\n",
       "      <td>63167.0</td>\n",
       "      <td>0.867660</td>\n",
       "      <td>11.072217</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>강동구</td>\n",
       "      <td>773</td>\n",
       "      <td>166.490765</td>\n",
       "      <td>453233.0</td>\n",
       "      <td>449019.0</td>\n",
       "      <td>4214.0</td>\n",
       "      <td>54622.0</td>\n",
       "      <td>0.929765</td>\n",
       "      <td>12.051638</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>강북구</td>\n",
       "      <td>748</td>\n",
       "      <td>125.203252</td>\n",
       "      <td>330192.0</td>\n",
       "      <td>326686.0</td>\n",
       "      <td>3506.0</td>\n",
       "      <td>54813.0</td>\n",
       "      <td>1.061806</td>\n",
       "      <td>16.600342</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>강서구</td>\n",
       "      <td>884</td>\n",
       "      <td>134.793814</td>\n",
       "      <td>603772.0</td>\n",
       "      <td>597248.0</td>\n",
       "      <td>6524.0</td>\n",
       "      <td>72548.0</td>\n",
       "      <td>1.080540</td>\n",
       "      <td>12.015794</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>관악구</td>\n",
       "      <td>1496</td>\n",
       "      <td>149.290780</td>\n",
       "      <td>525515.0</td>\n",
       "      <td>507203.0</td>\n",
       "      <td>18312.0</td>\n",
       "      <td>68082.0</td>\n",
       "      <td>3.484582</td>\n",
       "      <td>12.955291</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    구별    소계       최근증가율       인구수       한국인      외국인      고령자     외국인비율  \\\n",
       "0  강남구  2780  150.619195  570500.0  565550.0   4950.0  63167.0  0.867660   \n",
       "1  강동구   773  166.490765  453233.0  449019.0   4214.0  54622.0  0.929765   \n",
       "2  강북구   748  125.203252  330192.0  326686.0   3506.0  54813.0  1.061806   \n",
       "3  강서구   884  134.793814  603772.0  597248.0   6524.0  72548.0  1.080540   \n",
       "4  관악구  1496  149.290780  525515.0  507203.0  18312.0  68082.0  3.484582   \n",
       "\n",
       "       고령자비율  \n",
       "0  11.072217  \n",
       "1  12.051638  \n",
       "2  16.600342  \n",
       "3  12.015794  \n",
       "4  12.955291  "
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_result = data_result.drop(['2013년도 이전','2014년','2015년','2016년'],1)\n",
    "data_result.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "아, 판다스 너무 좋아요~💖 진짜 데이터 가공하기 너무 좋지 않나요,,, 감탄을 해야 합니다, 여러분ㅎㅎㅎ 코드 한줄로 이렇게 가공이 쉽다니,,,ㅠㅜ \n",
    "\n",
    "#### 인덱스 만들기 \n",
    "후에 우리가 표로 시각화해서 나타낼 것인데요, 보통 표는 x축과 y축으로 되어있죠. 그래서 표를 나타날 때 인덱스와 컬럼명이 x나 y축으로 들어가게 되어요. (잘 이해가 안 간다면 넘어가시고 시각화를 한 다음 다시 돌아와보세요!) 그렇기 때문에 저는 index를 구 이름으로 바꾸어보겠습니다. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>소계</th>\n",
       "      <th>최근증가율</th>\n",
       "      <th>인구수</th>\n",
       "      <th>한국인</th>\n",
       "      <th>외국인</th>\n",
       "      <th>고령자</th>\n",
       "      <th>외국인비율</th>\n",
       "      <th>고령자비율</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>구별</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>강남구</td>\n",
       "      <td>2780</td>\n",
       "      <td>150.619195</td>\n",
       "      <td>570500.0</td>\n",
       "      <td>565550.0</td>\n",
       "      <td>4950.0</td>\n",
       "      <td>63167.0</td>\n",
       "      <td>0.867660</td>\n",
       "      <td>11.072217</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>강동구</td>\n",
       "      <td>773</td>\n",
       "      <td>166.490765</td>\n",
       "      <td>453233.0</td>\n",
       "      <td>449019.0</td>\n",
       "      <td>4214.0</td>\n",
       "      <td>54622.0</td>\n",
       "      <td>0.929765</td>\n",
       "      <td>12.051638</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>강북구</td>\n",
       "      <td>748</td>\n",
       "      <td>125.203252</td>\n",
       "      <td>330192.0</td>\n",
       "      <td>326686.0</td>\n",
       "      <td>3506.0</td>\n",
       "      <td>54813.0</td>\n",
       "      <td>1.061806</td>\n",
       "      <td>16.600342</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>강서구</td>\n",
       "      <td>884</td>\n",
       "      <td>134.793814</td>\n",
       "      <td>603772.0</td>\n",
       "      <td>597248.0</td>\n",
       "      <td>6524.0</td>\n",
       "      <td>72548.0</td>\n",
       "      <td>1.080540</td>\n",
       "      <td>12.015794</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>관악구</td>\n",
       "      <td>1496</td>\n",
       "      <td>149.290780</td>\n",
       "      <td>525515.0</td>\n",
       "      <td>507203.0</td>\n",
       "      <td>18312.0</td>\n",
       "      <td>68082.0</td>\n",
       "      <td>3.484582</td>\n",
       "      <td>12.955291</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       소계       최근증가율       인구수       한국인      외국인      고령자     외국인비율  \\\n",
       "구별                                                                      \n",
       "강남구  2780  150.619195  570500.0  565550.0   4950.0  63167.0  0.867660   \n",
       "강동구   773  166.490765  453233.0  449019.0   4214.0  54622.0  0.929765   \n",
       "강북구   748  125.203252  330192.0  326686.0   3506.0  54813.0  1.061806   \n",
       "강서구   884  134.793814  603772.0  597248.0   6524.0  72548.0  1.080540   \n",
       "관악구  1496  149.290780  525515.0  507203.0  18312.0  68082.0  3.484582   \n",
       "\n",
       "         고령자비율  \n",
       "구별              \n",
       "강남구  11.072217  \n",
       "강동구  12.051638  \n",
       "강북구  16.600342  \n",
       "강서구  12.015794  \n",
       "관악구  12.955291  "
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_result.set_index('구별', inplace=True)\n",
    "data_result.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 상관계수 분석\n",
    "\n",
    "우리는 간단한 상관계수 분석을 해볼거에요. 사회통계론으로 상관계수 분석에 대해서 아실 것 같은데요, 상관관계는 관계를 알고자 하는 두 변수가 모두 연속변수일 경우에만 계산이 가능하고, 이는 인과관계를 분석하는게 아닌 두 연속변수가 관련이 있는지 없는지를 알아보는 기술통계 분석이에요. 두 변수를 x,y라고 하면 상관관계 계수(coefficient)는 x값과 y값을 모두 z값으로 표준화해서 Zx * Zy를 모두 더한다음, n(총 개수)으로 나눠준 것이 상관계수에요. 다른 것은 다됐고 여기서 상관계수는 넘파이가 다 계산해 줄 테니까 우리는 여기서 **상관계수가 무엇을 의미**하는지만 알면 돼요!\n",
    "\n",
    "- 0 ~ 0.2 : 매우 약한 상관관계 \n",
    "- 0.2 ~ 0.4 : 약한 상관관계 \n",
    "- 0.4 ~ 0.6 : 어느 정도의 상관관계 \n",
    "- 0.6 ~ 0.8 : 강한 상관관계 \n",
    "- 0.8 ~ 1.0 : 매우 강한 상관관계 \n",
    "\n",
    "상관계수의 범위는 -1부터 1까지 음수가 나오면 음의 상관관계, 양수가 나오면 양의 상관관계라는 것까지만 알면 됩니다. \n",
    "\n",
    "그럼 다수의 데이터 중 상관계수가 가장 큰 값은 데이터를 비교해볼게요. **numpy의 corrcoef 명령**을 사용하면 상관계수를 알아서 계산해줍니다. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 1.        , -0.28078554],\n",
       "       [-0.28078554,  1.        ]])"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 약한 음의 상관관계 \n",
    "np.corrcoef(data_result['고령자비율'], data_result['소계'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 1.        , -0.13607433],\n",
       "       [-0.13607433,  1.        ]])"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 상관관계가 거의 없다고 할 수 있음. \n",
    "np.corrcoef(data_result['외국인비율'], data_result['소계'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1.        , 0.30634228],\n",
       "       [0.30634228, 1.        ]])"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 0.3이어서 약한 상관관계 \n",
    "np.corrcoef(data_result['인구수'], data_result['소계'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "엇, 보니까 외국인비율은 거의 상관관계가 없다고 할 수 있고, 여기서 그래도 유의미한 것은 '인구수'네요! 그럼 CCTV와 인구수의 관계를 더 들여다볼게요!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>소계</th>\n",
       "      <th>최근증가율</th>\n",
       "      <th>인구수</th>\n",
       "      <th>한국인</th>\n",
       "      <th>외국인</th>\n",
       "      <th>고령자</th>\n",
       "      <th>외국인비율</th>\n",
       "      <th>고령자비율</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>구별</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>강남구</td>\n",
       "      <td>2780</td>\n",
       "      <td>150.619195</td>\n",
       "      <td>570500.0</td>\n",
       "      <td>565550.0</td>\n",
       "      <td>4950.0</td>\n",
       "      <td>63167.0</td>\n",
       "      <td>0.867660</td>\n",
       "      <td>11.072217</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>양천구</td>\n",
       "      <td>2034</td>\n",
       "      <td>34.671731</td>\n",
       "      <td>479978.0</td>\n",
       "      <td>475949.0</td>\n",
       "      <td>4029.0</td>\n",
       "      <td>52975.0</td>\n",
       "      <td>0.839413</td>\n",
       "      <td>11.036964</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>서초구</td>\n",
       "      <td>1930</td>\n",
       "      <td>63.371266</td>\n",
       "      <td>450310.0</td>\n",
       "      <td>445994.0</td>\n",
       "      <td>4316.0</td>\n",
       "      <td>51733.0</td>\n",
       "      <td>0.958451</td>\n",
       "      <td>11.488308</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>은평구</td>\n",
       "      <td>1873</td>\n",
       "      <td>85.237258</td>\n",
       "      <td>494388.0</td>\n",
       "      <td>489943.0</td>\n",
       "      <td>4445.0</td>\n",
       "      <td>72334.0</td>\n",
       "      <td>0.899091</td>\n",
       "      <td>14.631019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>용산구</td>\n",
       "      <td>1624</td>\n",
       "      <td>53.216374</td>\n",
       "      <td>244203.0</td>\n",
       "      <td>229456.0</td>\n",
       "      <td>14747.0</td>\n",
       "      <td>36231.0</td>\n",
       "      <td>6.038828</td>\n",
       "      <td>14.836427</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       소계       최근증가율       인구수       한국인      외국인      고령자     외국인비율  \\\n",
       "구별                                                                      \n",
       "강남구  2780  150.619195  570500.0  565550.0   4950.0  63167.0  0.867660   \n",
       "양천구  2034   34.671731  479978.0  475949.0   4029.0  52975.0  0.839413   \n",
       "서초구  1930   63.371266  450310.0  445994.0   4316.0  51733.0  0.958451   \n",
       "은평구  1873   85.237258  494388.0  489943.0   4445.0  72334.0  0.899091   \n",
       "용산구  1624   53.216374  244203.0  229456.0  14747.0  36231.0  6.038828   \n",
       "\n",
       "         고령자비율  \n",
       "구별              \n",
       "강남구  11.072217  \n",
       "양천구  11.036964  \n",
       "서초구  11.488308  \n",
       "은평구  14.631019  \n",
       "용산구  14.836427  "
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_result.sort_values(by='소계', ascending=False).head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "오호라, 여기에서 가장 CCTV가 많이 설치된 구는 강남구, 양천구, 서초구 순이네요. 그럼 인구수를 한번 볼까요? "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>소계</th>\n",
       "      <th>최근증가율</th>\n",
       "      <th>인구수</th>\n",
       "      <th>한국인</th>\n",
       "      <th>외국인</th>\n",
       "      <th>고령자</th>\n",
       "      <th>외국인비율</th>\n",
       "      <th>고령자비율</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>구별</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>송파구</td>\n",
       "      <td>618</td>\n",
       "      <td>104.347826</td>\n",
       "      <td>667483.0</td>\n",
       "      <td>660584.0</td>\n",
       "      <td>6899.0</td>\n",
       "      <td>72506.0</td>\n",
       "      <td>1.033584</td>\n",
       "      <td>10.862599</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>강서구</td>\n",
       "      <td>884</td>\n",
       "      <td>134.793814</td>\n",
       "      <td>603772.0</td>\n",
       "      <td>597248.0</td>\n",
       "      <td>6524.0</td>\n",
       "      <td>72548.0</td>\n",
       "      <td>1.080540</td>\n",
       "      <td>12.015794</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>강남구</td>\n",
       "      <td>2780</td>\n",
       "      <td>150.619195</td>\n",
       "      <td>570500.0</td>\n",
       "      <td>565550.0</td>\n",
       "      <td>4950.0</td>\n",
       "      <td>63167.0</td>\n",
       "      <td>0.867660</td>\n",
       "      <td>11.072217</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>노원구</td>\n",
       "      <td>1265</td>\n",
       "      <td>188.929889</td>\n",
       "      <td>569384.0</td>\n",
       "      <td>565565.0</td>\n",
       "      <td>3819.0</td>\n",
       "      <td>71941.0</td>\n",
       "      <td>0.670725</td>\n",
       "      <td>12.634883</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>관악구</td>\n",
       "      <td>1496</td>\n",
       "      <td>149.290780</td>\n",
       "      <td>525515.0</td>\n",
       "      <td>507203.0</td>\n",
       "      <td>18312.0</td>\n",
       "      <td>68082.0</td>\n",
       "      <td>3.484582</td>\n",
       "      <td>12.955291</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       소계       최근증가율       인구수       한국인      외국인      고령자     외국인비율  \\\n",
       "구별                                                                      \n",
       "송파구   618  104.347826  667483.0  660584.0   6899.0  72506.0  1.033584   \n",
       "강서구   884  134.793814  603772.0  597248.0   6524.0  72548.0  1.080540   \n",
       "강남구  2780  150.619195  570500.0  565550.0   4950.0  63167.0  0.867660   \n",
       "노원구  1265  188.929889  569384.0  565565.0   3819.0  71941.0  0.670725   \n",
       "관악구  1496  149.290780  525515.0  507203.0  18312.0  68082.0  3.484582   \n",
       "\n",
       "         고령자비율  \n",
       "구별              \n",
       "송파구  10.862599  \n",
       "강서구  12.015794  \n",
       "강남구  11.072217  \n",
       "노원구  12.634883  \n",
       "관악구  12.955291  "
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_result.sort_values(by='인구수', ascending=False).head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "아하, 인구수로 보니 송파구, 강서구, 강남구 순으로 인구수가 많네요. \n",
    "\n",
    "### 2.5 matplotlib으로 시각화하기 \n",
    "\n",
    "우리 결과들을 더 직관적으로 보기 위해서 matplotlib를 사용하여 그래프로 만들어봅시다. 어떤 그래프를 선택할 것이냐는 데이터에 따라서 달라져요. **데이터의 특성을 잘 나타낼 수 있는 그래프를 선택할 수 있어야 합니다.** 그래서 앞서서 데이터를 쭈욱 살펴봤던 것이 그래서 매우 중요해요. 아니면 반대로 데이터의 특성을 잘 모르겠을 때, 시각화해서 표로 표시한다면 데이터의 특성을 좀 더 잘 알게 되기도 합니다. (이것은 나중에 더 설명해보기로 하구요🐼) \n",
    "\n",
    "우선, matplotlib를 inline으로 깔아주고, 표에 한글이 나올 수 있도록 복붙만 해봅시다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt \n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "import platform \n",
    "\n",
    "from matplotlib import font_manager, rc \n",
    "plt.rcParams['axes.unicode_minus'] = False\n",
    "\n",
    "if platform.system() == 'Darwin':\n",
    "    rc('font', family='AppleGothic')\n",
    "elif platform.system() == 'Windows':\n",
    "    path = \"c:/Windows/Fonts/malgun.ttf\"\n",
    "    font_name = font_manager.FontProperties(fname=path).get_name()\n",
    "    rc('font', family=font_name)\n",
    "else:\n",
    "    print('Unknown system')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>소계</th>\n",
       "      <th>최근증가율</th>\n",
       "      <th>인구수</th>\n",
       "      <th>한국인</th>\n",
       "      <th>외국인</th>\n",
       "      <th>고령자</th>\n",
       "      <th>외국인비율</th>\n",
       "      <th>고령자비율</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>구별</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>강남구</td>\n",
       "      <td>2780</td>\n",
       "      <td>150.619195</td>\n",
       "      <td>570500.0</td>\n",
       "      <td>565550.0</td>\n",
       "      <td>4950.0</td>\n",
       "      <td>63167.0</td>\n",
       "      <td>0.867660</td>\n",
       "      <td>11.072217</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>강동구</td>\n",
       "      <td>773</td>\n",
       "      <td>166.490765</td>\n",
       "      <td>453233.0</td>\n",
       "      <td>449019.0</td>\n",
       "      <td>4214.0</td>\n",
       "      <td>54622.0</td>\n",
       "      <td>0.929765</td>\n",
       "      <td>12.051638</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>강북구</td>\n",
       "      <td>748</td>\n",
       "      <td>125.203252</td>\n",
       "      <td>330192.0</td>\n",
       "      <td>326686.0</td>\n",
       "      <td>3506.0</td>\n",
       "      <td>54813.0</td>\n",
       "      <td>1.061806</td>\n",
       "      <td>16.600342</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>강서구</td>\n",
       "      <td>884</td>\n",
       "      <td>134.793814</td>\n",
       "      <td>603772.0</td>\n",
       "      <td>597248.0</td>\n",
       "      <td>6524.0</td>\n",
       "      <td>72548.0</td>\n",
       "      <td>1.080540</td>\n",
       "      <td>12.015794</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>관악구</td>\n",
       "      <td>1496</td>\n",
       "      <td>149.290780</td>\n",
       "      <td>525515.0</td>\n",
       "      <td>507203.0</td>\n",
       "      <td>18312.0</td>\n",
       "      <td>68082.0</td>\n",
       "      <td>3.484582</td>\n",
       "      <td>12.955291</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       소계       최근증가율       인구수       한국인      외국인      고령자     외국인비율  \\\n",
       "구별                                                                      \n",
       "강남구  2780  150.619195  570500.0  565550.0   4950.0  63167.0  0.867660   \n",
       "강동구   773  166.490765  453233.0  449019.0   4214.0  54622.0  0.929765   \n",
       "강북구   748  125.203252  330192.0  326686.0   3506.0  54813.0  1.061806   \n",
       "강서구   884  134.793814  603772.0  597248.0   6524.0  72548.0  1.080540   \n",
       "관악구  1496  149.290780  525515.0  507203.0  18312.0  68082.0  3.484582   \n",
       "\n",
       "         고령자비율  \n",
       "구별              \n",
       "강남구  11.072217  \n",
       "강동구  12.051638  \n",
       "강북구  16.600342  \n",
       "강서구  12.015794  \n",
       "관악구  12.955291  "
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_result.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### bar 그래프 그리기\n",
    "\n",
    "우선, 직관적으로 구별 CCTV 설치수를 알고 싶어요. 그래서 저는 **바 그래프**를 한번 선택해보았습니다. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data_result['소계'].plot(kind='barh', grid=True, figsize=(10,10))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "숫자만 볼 때보다는 나은데 그래도 뒤죽박죽하니 정렬하면 더 잘 알수 있겠죠."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data_result['소계'].sort_values().plot(kind='barh', grid=True, figsize=(10,10))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "이렇게 sort_values()를 메서드 체이닝으로 연결해주어 정렬해줍니다. 와, 이렇게 bar 그래프로 보니까 강남구가 정말 많기는 많네요..! 그리고 그리드를 설정해줘서 강남구가 도봉구의 5배 이상이 된다는 것 또한 쉽게 알 수 있네요. 보니까 강남구가 월등하게 CCTV 설치수가 많고, 그 다음에는 양천구, 서초구, 은평구 순으로 많습니다. 가장 수가 적은 곳은 도봉구이고, 중구, 중랑구, 송파구, 마포구가 CCTV 설치수가 상대적으로 낮은 곳이네요. \n",
    "\n",
    "그렇지만 CCTV 수가 많은 것만 보면 안되죠. 지역구가 다른 곳에 비해 상대적으로 넓다거나, 인구 수가 많아서 CCTV를 더 많이 설치했을 수도 있잖아요. 그래서 우리는 인구 대비 CCTV 비율을 계산해서 다시 시각화해볼게요."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data_result['CCTV비율'] = data_result['소계'] / data_result['인구수'] * 100\n",
    "data_result['CCTV비율'].sort_values().plot(kind='barh', grid=True, figsize=(10,10))\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "오, 보니까 용산구는 CCTV 수도 상위였지만 인구 대비 수로 보면 월등하네요. 종로구는 CCTV 수로는 중위정도 됐는데, 인구 대비 비율로는 상위권에 속해네요ㅎㅎ 오, 시각화 하니까 이런 것까지 알 수 있네요! \n",
    "\n",
    "##### 산점도 그리기 \n",
    "\n",
    "그러면 나아가서 인구수와 CCTV 수를 산점도로 그려볼게요. **scatter**를 사용하면 산점도를 그릴 수 있고, s는 마커의 크기입니다.(마커가 뭔지 모른다면 숫자를 늘렸다 줄였다 해보세요!)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(6,6))\n",
    "plt.scatter(data_result['인구수'], data_result['소계'], s=50)\n",
    "plt.xlabel('인구수')\n",
    "plt.ylabel('CCTV')\n",
    "plt.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "각각의 코드가 무엇을 의미하는지 모른다면 matplotlib의 scatter를 좀더 보고오세요^^ 인구수를 보면 x축이 100,000부터 시작해 700,000까지 있죠? 기억해두시면 이따가 추세선 그릴 때 또 쓰일 거에요.\n",
    "\n",
    "##### 추세선 그리기 - 선형 회귀선\n",
    "(사회통계론을 듣지 않았거나 이 부분을 읽어도 이해가 가지 않는다면, 코드만 복붙해서 실행해보고 넘어가면 됩니다!)\n",
    "\n",
    "오, 이제는 여기에 대충 경향을 보고 싶어요. 아까 상관계수가 어느 정도 약하게 있다고 나왔잖아요. 그래서 요 데이터를 가지고 경향성을 나타낼 수 있는 하나의 직선을 만들어줄 겁니다. 우리가 넘파이 실습과제에서 하나의 선을 그렸듯이 넘파이의 **polyfit 함수**를 사용해서 만들어줄 거에요. polyfit()은 **n차 다항식으로 회귀분석을 할 때 사용하는 함수**에요.\n",
    "\n",
    "- np.polyfit(x, y, n): x는 x에 들어갈 데이터, y는 y에 들어갈 데이터, n는 다항식의 차수\n",
    "\n",
    "우리가 그리고 싶은 추세선은 직선이라 y=ax+b라고 하죠. 직선은 1차원이니까 n=1이고, polyfit의 결과값은 [a,b]가 나오게 되어요. 즉, 기울기와 절편값이 나오는거죠. 우리가 사통론에서 요 a,b값을 구하려고, 일일이 하나씩 다 계산해서 정말 힘들게 계산했잖아요. 그런데 이렇게 polyfit()을 사용하면 바로 기울기와 절편값을 줍니다. 2,3,4 등 n차원까지 모두 가능해요."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1.30916415e-03, 6.45066497e+02])"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fp1 = np.polyfit(data_result['인구수'], data_result['소계'],1)\n",
    "fp1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "그럼 이건 주어진 데이터를 가지고 만들어낸 1차 방정식의 기울기와 절편값이에요. 그 다음에는 **poly1d()**를 가지고 직선을 만들어볼게요. \n",
    "\n",
    "- [poly1d()란?](https://docs.scipy.org/doc/numpy/reference/generated/numpy.poly1d.html)\n",
    "\n",
    "이것을 보시고 이해를 해보세요. poly1d는 다항식(polynomial function)을 정의해주는 함수에요. 즉, 우리가 poly1d() 안에 [a,b]를 넣는다면, ax+b라는 1차 방정식을 넣어주는거죠. 만약 poly1d([1,2,3])이라면 x^2+2x+3을 만들어주는 것이구요. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "poly1d([1.30916415e-03, 6.45066497e+02])"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f1 = np.poly1d(fp1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "f1 자체가 방정식이니까, 직선을 만드려면 x값을 공급해줘야해요. 그 x값은 **np.linspace**로 만들어 줄 거에요. \n",
    "\n",
    "- np.linspace(start, stop, num): num은 데이터 개수\n",
    "\n",
    "np.linspace 함수는 start부터 stop의 범위에서 num개를 균일한 간격으로 데이터를 생성하고 배열을 만드는 함수에요. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "fx = np.linspace(100000, 700000, 100)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "그러니까 정리를 하면, x에는 fx를 공급해주고, y로는 다항식을 만들어 준 f1을 넣어주는 것이죠. 그래서 점선의 일차원 그래프를 만들어보면! "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,10))\n",
    "plt.scatter(data_result['인구수'], data_result['소계'], s=50)\n",
    "plt.plot(fx, f1(fx), ls='dashed', lw=3, color='g')\n",
    "plt.xlabel('인구수')\n",
    "plt.ylabel('CCTV')\n",
    "plt.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "이렇게 추세선 완성! 오, CCTV와 인구수에 이런 추세선이 그려지는 군요. \n",
    "\n",
    "여기에다가 저자는 두 가지 장치를 더했어요. 하나는 직선의 경향에서 멀어진 구일수록 색이 변한다는 것과, 다른 하나는 경향에서 멀어진 구의 이름을 표시하고 싶다는 거에요. 그래서 이를 표시하기위해 오차를 계산하는 코드를 만들었어요. \n",
    "\n",
    "- 오차 = |(CCTV수) - f1(인구수)|\n",
    "\n",
    "엇, 이거 우리 회귀분석할 때 선형 회귀식을 만들 때 봤죠? 아까 우리가 추세선을 만들었잖아요. 그 추세선이 바로 **회귀식**이랍니다. **오차는 회귀선을 통해서 나온 y값과 실제 y값의 차이**잖아요. 그래서 요거를 이용해서 오차를 계산해 data_result에다 새로운 컬럼으로 넣어주어봤습니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>소계</th>\n",
       "      <th>최근증가율</th>\n",
       "      <th>인구수</th>\n",
       "      <th>한국인</th>\n",
       "      <th>외국인</th>\n",
       "      <th>고령자</th>\n",
       "      <th>외국인비율</th>\n",
       "      <th>고령자비율</th>\n",
       "      <th>CCTV비율</th>\n",
       "      <th>오차</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>구별</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>강남구</td>\n",
       "      <td>2780</td>\n",
       "      <td>150.619195</td>\n",
       "      <td>570500.0</td>\n",
       "      <td>565550.0</td>\n",
       "      <td>4950.0</td>\n",
       "      <td>63167.0</td>\n",
       "      <td>0.867660</td>\n",
       "      <td>11.072217</td>\n",
       "      <td>0.487292</td>\n",
       "      <td>1388.055355</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>강동구</td>\n",
       "      <td>773</td>\n",
       "      <td>166.490765</td>\n",
       "      <td>453233.0</td>\n",
       "      <td>449019.0</td>\n",
       "      <td>4214.0</td>\n",
       "      <td>54622.0</td>\n",
       "      <td>0.929765</td>\n",
       "      <td>12.051638</td>\n",
       "      <td>0.170552</td>\n",
       "      <td>465.422892</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>강북구</td>\n",
       "      <td>748</td>\n",
       "      <td>125.203252</td>\n",
       "      <td>330192.0</td>\n",
       "      <td>326686.0</td>\n",
       "      <td>3506.0</td>\n",
       "      <td>54813.0</td>\n",
       "      <td>1.061806</td>\n",
       "      <td>16.600342</td>\n",
       "      <td>0.226535</td>\n",
       "      <td>329.342026</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>강서구</td>\n",
       "      <td>884</td>\n",
       "      <td>134.793814</td>\n",
       "      <td>603772.0</td>\n",
       "      <td>597248.0</td>\n",
       "      <td>6524.0</td>\n",
       "      <td>72548.0</td>\n",
       "      <td>1.080540</td>\n",
       "      <td>12.015794</td>\n",
       "      <td>0.146413</td>\n",
       "      <td>551.503155</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>관악구</td>\n",
       "      <td>1496</td>\n",
       "      <td>149.290780</td>\n",
       "      <td>525515.0</td>\n",
       "      <td>507203.0</td>\n",
       "      <td>18312.0</td>\n",
       "      <td>68082.0</td>\n",
       "      <td>3.484582</td>\n",
       "      <td>12.955291</td>\n",
       "      <td>0.284673</td>\n",
       "      <td>162.948104</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       소계       최근증가율       인구수       한국인      외국인      고령자     외국인비율  \\\n",
       "구별                                                                      \n",
       "강남구  2780  150.619195  570500.0  565550.0   4950.0  63167.0  0.867660   \n",
       "강동구   773  166.490765  453233.0  449019.0   4214.0  54622.0  0.929765   \n",
       "강북구   748  125.203252  330192.0  326686.0   3506.0  54813.0  1.061806   \n",
       "강서구   884  134.793814  603772.0  597248.0   6524.0  72548.0  1.080540   \n",
       "관악구  1496  149.290780  525515.0  507203.0  18312.0  68082.0  3.484582   \n",
       "\n",
       "         고령자비율    CCTV비율           오차  \n",
       "구별                                     \n",
       "강남구  11.072217  0.487292  1388.055355  \n",
       "강동구  12.051638  0.170552   465.422892  \n",
       "강북구  16.600342  0.226535   329.342026  \n",
       "강서구  12.015794  0.146413   551.503155  \n",
       "관악구  12.955291  0.284673   162.948104  "
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fp1 = np.polyfit(data_result['인구수'], data_result['소계'],1)\n",
    "\n",
    "f1 = np.poly1d(fp1)\n",
    "fx = np.linspace(100000, 700000, 100)\n",
    "\n",
    "# 오차 칼럼 만들어주기 \n",
    "data_result['오차'] = np.abs(data_result['소계'] - f1(data_result['인구수']))\n",
    "data_result.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "그럼 이것을 '오차'를 기준으로 데이터를 정렬해보도록 할게요. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>소계</th>\n",
       "      <th>최근증가율</th>\n",
       "      <th>인구수</th>\n",
       "      <th>한국인</th>\n",
       "      <th>외국인</th>\n",
       "      <th>고령자</th>\n",
       "      <th>외국인비율</th>\n",
       "      <th>고령자비율</th>\n",
       "      <th>CCTV비율</th>\n",
       "      <th>오차</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>구별</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>강남구</td>\n",
       "      <td>2780</td>\n",
       "      <td>150.619195</td>\n",
       "      <td>570500.0</td>\n",
       "      <td>565550.0</td>\n",
       "      <td>4950.0</td>\n",
       "      <td>63167.0</td>\n",
       "      <td>0.867660</td>\n",
       "      <td>11.072217</td>\n",
       "      <td>0.487292</td>\n",
       "      <td>1388.055355</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>송파구</td>\n",
       "      <td>618</td>\n",
       "      <td>104.347826</td>\n",
       "      <td>667483.0</td>\n",
       "      <td>660584.0</td>\n",
       "      <td>6899.0</td>\n",
       "      <td>72506.0</td>\n",
       "      <td>1.033584</td>\n",
       "      <td>10.862599</td>\n",
       "      <td>0.092587</td>\n",
       "      <td>900.911312</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>양천구</td>\n",
       "      <td>2034</td>\n",
       "      <td>34.671731</td>\n",
       "      <td>479978.0</td>\n",
       "      <td>475949.0</td>\n",
       "      <td>4029.0</td>\n",
       "      <td>52975.0</td>\n",
       "      <td>0.839413</td>\n",
       "      <td>11.036964</td>\n",
       "      <td>0.423769</td>\n",
       "      <td>760.563512</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>서초구</td>\n",
       "      <td>1930</td>\n",
       "      <td>63.371266</td>\n",
       "      <td>450310.0</td>\n",
       "      <td>445994.0</td>\n",
       "      <td>4316.0</td>\n",
       "      <td>51733.0</td>\n",
       "      <td>0.958451</td>\n",
       "      <td>11.488308</td>\n",
       "      <td>0.428594</td>\n",
       "      <td>695.403794</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>용산구</td>\n",
       "      <td>1624</td>\n",
       "      <td>53.216374</td>\n",
       "      <td>244203.0</td>\n",
       "      <td>229456.0</td>\n",
       "      <td>14747.0</td>\n",
       "      <td>36231.0</td>\n",
       "      <td>6.038828</td>\n",
       "      <td>14.836427</td>\n",
       "      <td>0.665020</td>\n",
       "      <td>659.231690</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       소계       최근증가율       인구수       한국인      외국인      고령자     외국인비율  \\\n",
       "구별                                                                      \n",
       "강남구  2780  150.619195  570500.0  565550.0   4950.0  63167.0  0.867660   \n",
       "송파구   618  104.347826  667483.0  660584.0   6899.0  72506.0  1.033584   \n",
       "양천구  2034   34.671731  479978.0  475949.0   4029.0  52975.0  0.839413   \n",
       "서초구  1930   63.371266  450310.0  445994.0   4316.0  51733.0  0.958451   \n",
       "용산구  1624   53.216374  244203.0  229456.0  14747.0  36231.0  6.038828   \n",
       "\n",
       "         고령자비율    CCTV비율           오차  \n",
       "구별                                     \n",
       "강남구  11.072217  0.487292  1388.055355  \n",
       "송파구  10.862599  0.092587   900.911312  \n",
       "양천구  11.036964  0.423769   760.563512  \n",
       "서초구  11.488308  0.428594   695.403794  \n",
       "용산구  14.836427  0.665020   659.231690  "
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_sort = data_result.sort_values(by='오차', ascending=False)\n",
    "df_sort.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "그럼 이제 텍스트와 color map을 입혀볼게요. \n",
    "\n",
    "- color는 plt.scatter()의 c 옵션값으로 넣어줬어요. \n",
    "- plt.text(x, y, s): x와 y는 (x,y)위치(the position to place the text), s는 넣어줄 텍스트(the text)\n",
    "\n",
    "우리 파이썬을 배웠기 때문에 아래의 for문 이해할 수 있어요💖 "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(14,10))\n",
    "plt.scatter(data_result['인구수'], data_result['소계'],\n",
    "           c=data_result['오차'], s=50)                   #c=marker color, s=marker size\n",
    "plt.plot(fx, f1(fx), ls='dashed', lw=3, color='g')       #lw=line width\n",
    "\n",
    "# 텍스트 만들어주기\n",
    "for n in range(10):\n",
    "    plt.text(df_sort['인구수'][n]*1.02, df_sort['소계'][n]*0.98,   \n",
    "            df_sort.index[n], fontsize=15)\n",
    "\n",
    "plt.xlabel('인구수')\n",
    "plt.ylabel('인구당비율')\n",
    "\n",
    "plt.colorbar()\n",
    "plt.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "df_sort는 오차가 높은 순으로 배열을 해놓았죠. 그래서 for문을 사용해 range(10), 즉 상위 10개만 text를 보이도록 해줬어요. \n",
    "\n",
    "- df_sort['인구수'][n]*1.02, df_sort['소계'][n]*0.98\n",
    "\n",
    "이렇게 적은 이유는 점의 바로 위에 이름이 올라가게 하지 말고, 오른쪽으로 이름이 위치할 수 있도록 한거에요. 이름이 위치한 곳이 모두 다 오른쪽 살짝 옆에 있죠? 이것을 for문으로 구현한 것이랍니다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.5 분석 결과 \n",
    "\n",
    "직선을 기준으로 위에 있는 '강남구', '양천구', '서초구', '은평구', '용산구'는 서울시 전체 지역의 일반적인 경향보다 CCTV가 많이 설치된 지역이네요. 그리고 '송파구','강서구','중랑구','마포구','도봉구'는 일반적인 경향보다 CCTV가 적게 설치된 지역이고요. 특히 '강남구'는 월등히 많은 CCTV가 설치됐지만, '송파구'는 인구수에 비해 너무나도 적은 수의 CCTV를 가지고 있네요!\n",
    "\n",
    "여기다 이제 우리는 사회학하는 친구들이니까 왜 이렇게 될까를 생각해봐야겠죠. 앞서 처음에 생각했던 '구별 예산 정도에 따라 CCTV 설치수가 다를꺼다'라는 것을 증명할 수는 없지만(예산 데이터가 없으니까요), 왜 저렇게 강남구는 아웃라이어로 있나라고 생각해볼 수 있겠죠. 이제 이 데이터에 대한 해석을 하는 것이 사회학자의 몫이겠죵😊"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 프로젝트를 끝내며 \n",
    "\n",
    "축하합니다, 여러분! 드디어 하나의 프로젝트가 끝이 났어요!!! (이틀째 꼴딱 새고 이 교안을 만들었는데, 드디어 만들었어요ㅎㅎ) \n",
    "\n",
    "이렇게 첫번째 프로젝트를 해보았네요. 끝에 추세선 만들기에서 많이 어려웠을 것 같은데, 최대한 이해를 해보시고 안되면 넘어가세요. 괜히 압도되지 말고 할 수 있는 것을 챙겨서 하면 된답니다💕 전체 과정을 한번 쭉 보세요. 거시적으로 보면서 하나하나씩 세부적으로 들어가면 왜 이 코드를 작성하고 있는지 알 수 있게 됩니다. 여러분에게 많이 도움이 되었길 바라구요. 모두모두 고생했습니다💗"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}