20년 서울시 행정구역별 전세 값 데이터 분석 & 시각화 with Python/Mapbox

Data Source

공공데이터 국토교통부_아파트 전월세 자료

data.go.kr/tcs/dss/selectApiDataDetailView.do?publicDataPk=15058017

공공데이터 포털

 

20년 서울시 행정구역별 평균 전세값 시각화 With Mapbox

 

 

 

 

20년 서울시 행정구역별 총 거래금액 Tree map 시각화 With Python

 

20년 총 거래금액(전세)을 기준으로 보았을때 확실히 강남 3구 전세 금액이 가장 큰 점을 확인 가능함

 

 

20년 서울시 행정구역별 평균 전세 금액 월별 변화

 

Python Code

데이터 추출과 시각화를 위한 파이썬 코드는 다음과 같다.

import pandas as pd
code=pd.read_html('https://inasie.github.io/%ED%94%84%EB%A1%9C%EA%B7%B8%EB%9E%98%EB%B0%8D/5/',encoding='utf-8')[0]
code2=code[code['법정동주소'].str.contains('서울특별시')][1:]

from urllib.request import urlopen
from urllib.parse import urlencode, unquote, quote_plus
import urllib
import requests
import pandas as pd
import xmltodict
import json
from datetime import datetime
import requests

api='~~~'
url=f'http://openapi.molit.go.kr:8081/OpenAPI_ToolInstallPackage/service/rest/RTMSOBJSvc/getRTMSDataSvcAptRent?serviceKey={api}&'

months=[]
for i in range(1,13):
    if i <10:
        months.append('2020'+'0'+str(i))
    else:
        months.append('2020'+str(i))
        
wol=[]
for code in code2['법정동코드']:
    for dates in months:
        queryParams =urlencode({quote_plus('LAWD_CD') : code, quote_plus('DEAL_YMD') : dates})
        url2 = url + queryParams
        response = urlopen(url2) 
        results = response.read().decode("utf-8")
        results_to_json = xmltodict.parse(results)
        data = json.loads(json.dumps(results_to_json))
        wol.extend(data['response']['body']['items']['item'])
        
df=pd.DataFrame(wol)
code2['법정동코드']=code2['법정동코드'].astype(str)
df2=df[df['월세금액']=='0']
df2['보증금액']=df2['보증금액'].str.replace(',','').astype(float)
jeon=pd.DataFrame(df2.groupby(['지역코드']).mean()['보증금액']).reset_index()
jeon2=pd.merge(left=jeon,right=code2,left_on=jeon['지역코드'],right_on=code2['법정동코드'],how='inner')

import json
state=json.load(open('C:/Users/banad/Downloads/TL_SCCO_SIG (1).zip.geojson', encoding='utf-8'))
for idx, sigun in enumerate(state['features']):
    code=sigun['properties']['SIG_CD']
    if code in jeon2['지역코드'].values:
        price= jeon2.loc[(jeon2['지역코드']==code),'보증금액'].iloc[0]
        addr= jeon2.loc[(jeon2['지역코드']==code),'법정동주소'].iloc[0]
        state['features'][idx]['properties']['보증금액(만 원)']=price
        state['features'][idx]['properties']['법정동주소']=addr
        
data=[x for x in state['features'] if len(x['properties'])>1 ]
state['features']=data

import mapboxgl
from mapboxgl.viz import *
import os
from mapboxgl.utils import create_color_stops
from mapboxgl.utils import create_numeric_stops

api='~'

center =[126.977123, 37.517169]

color_breaks = [10000,20000,30000,40000,50000,60000,70000,80000,90000]
color_stops =create_color_stops(color_breaks,colors='YlGnBu')
viz =ChoroplethViz(
    access_token=api,
    data=state,
    color_property='보증금액(만 원)',
    color_stops=color_stops,
    center=center,
    zoom=10,
    bearing=-11,
    pitch=43,
    height_property='보증금액(만 원)',
    height_stops=create_numeric_stops(color_breaks,0,10000),
    height_function_type='interpolate'
)
viz.show()  

import squarify
import numpy as np
import matplotlib.pyplot as plt

total=pd.DataFrame(df2.groupby(['지역코드']).sum()['보증금액']).reset_index()
total2=pd.merge(left=total,right=code2,left_on=total['지역코드'],right_on=code2['법정동코드'],how='inner')
total2['구']=total2['법정동주소'].apply(lambda x: x.split(' ')[1])

plt.rcParams["font.family"] = 'Malgun Gothic'

sizes=np.array(total2['보증금액'])
labels=total2['구']
colors = [plt.cm.Spectral(i/float(len(labels))) for i in range(len(labels))]
plt.figure(figsize=(12,8), dpi= 100)
squarify.plot(sizes=sizes, label=labels, color = colors, alpha=.5, edgecolor="black", linewidth=3, text_kwargs={'fontsize':10})
plt.axis('off')
plt.show()

df2['월']=df2['월'].apply(lambda x: '0'+str(x) if int(x)<10 else str(x) )
df2['년월']=df2['년']+df2['월']
df3=pd.merge(left=df2,right=code2,left_on=df2['지역코드'],right_on=code2['법정동코드'],how='left')
df4=pd.DataFrame(df3.groupby(['법정동주소','년월']).mean()['보증금액']).reset_index()
df4['주소']=df4['법정동주소'].apply(lambda x:x.split(' ')[1])

eng=pd.read_excel('C:/Users/banad/Downloads/seoul.xlsx')
df5=pd.merge(left=df4,right=eng,how='left',left_on=df4['주소'],right_on=eng['kor'])
df5=df5.rename(columns={'법정동주소':'addr','년월':'date','보증금액':'price'})
import plotly.express as px
fig = px.bar(df5, x="eng", y="price", color="price",
  animation_frame="date",  range_y=[0,100000])
fig.show()

jj=fig.to_json()
eng2=eng.set_index('eng')
eng3=eng2.to_dict()['kor']

for x,y in zip(list(eng3.keys()), list(eng3.values())):
    jj = jj.replace(x, y)