In skysign/KoreaAddressAPI: To Validate, Parse and Correct Korea Address

library(dplyr)

주소 데이터 불러오기

pathPrefix = '../cleaned/'
YYYYMMPrefix = '202004/'
pathPostFix = 'JiBeon_Korea.csv'
path = paste0(pathPrefix, YYYYMMPrefix, pathPostFix)
df = read.csv(path, fileEncoding = 'UTF-8')

head(df)

모든 컬럼의 타입을 chraracter로 변경하기

df$BeopJeongDongKoDeu = as.character(df$BeopJeongDongKoDeu) # 법정동코드 10자리 , 앞 두자리는 시도, 3~5자리는 구
df$SiDoMyeong = as.character(df$SiDoMyeong) #시, 도
df$SiGunGuMyeong = as.character(df$SiGunGuMyeong) #구
df$BeopJeongEupMyeonDongMyeong = as.character(df$BeopJeongEupMyeonDongMyeong) #동
df$JiBeonBonBeon = as.character(df$JiBeonBonBeon.BeonJi) #지번본번
df$JiBeonBuBeon = as.character(df$JiBeonBuBeon.Ho) #지번부번

head(df)

코드,시,군,동, 번지만 분리하여 데이터 프레임 생성

KAAreas = c('SiDoMyeong', 'SiGunGuMyeong', 'BeopJeongEupMyeonDongMyeong','JiBeonBonBeon','JiBeonBuBeon')
columns = c('BeopJeongDongKoDeu', KAAreas)

df2 <- df[columns]
head(df2)

df2['code_root'] = '0'
df2['code_Si'] = substr(df2$BeopJeongDongKoDeu, 1, 2) # 법정동 코드에서 시 분리
df2['code_Gu'] = substr(df2$BeopJeongDongKoDeu, 3, 5) # 법정동 코드에서 구 분리
df2['code_Dong'] = substr(df2$BeopJeongDongKoDeu, 6, 8) # 법정동 코드에서 동 분리
df2['code_BonBeon'] = paste0(df2$code_Dong, df2$JiBeonBonBeon) # 법정동 코드 + 지번본번
df2['code_BuBeon'] = df2$JiBeonBuBeon # 법정동 코드 + 지번본번 + 지번부번
head(df2)

siColumns = c('code_root', 'code_Si', 'SiDoMyeong')
guColumns = c('code_Si', 'code_Gu', 'SiGunGuMyeong')
dongColumns = c('code_Gu', 'code_Dong', 'BeopJeongEupMyeonDongMyeong')
bonBeonColumns = c('code_Dong','code_BonBeon','JiBeonBonBeon')
buBeonColumns = c('code_BonBeon','code_BuBeon','JiBeonBuBeon')

dfSi = df2[siColumns]
dfGu = df2[guColumns]
dfDong = df2[dongColumns]
dfBonBeon = df2[bonBeonColumns]
dfBuBeon = df2[buBeonColumns]

dfSi = unique(dfSi)
dfGu = unique(dfGu)
dfDong = unique(dfDong)
dfBonBeon = unique(dfBonBeon)
dfBuBeon = unique(dfBuBeon)

newColumns = c('parent', 'child', 'HaengJeongGuYeok')

colnames(dfSi) = newColumns
colnames(dfGu) = newColumns
colnames(dfDong) = newColumns
colnames(dfBonBeon) = newColumns
colnames(dfBuBeon) = newColumns

head(dfSi)
head(dfGu)
head(dfDong)
head(dfBonBeon)
head(dfBuBeon)

R package에 넣을 데이터프레임 rda에 저장

dfsRec = list(dfSi, dfGu, dfDong, dfBonBeon, dfBuBeon)
pathPrefix = '../data/'
pathPostFix = 'dfs.rda'
path = paste0(pathPrefix, pathPostFix)

save(dfsRec, file = path)


pathPostFix = 'KRAreas.rda'
path = paste0(pathPrefix, pathPostFix)
save(KAAreas, file = path)

rda 파일을 최적화해서, 용량을 작게 만들기

path = '../data'

tools::resaveRdaFiles(path)

skysign/KoreaAddressAPI documentation built on July 20, 2020, 9:08 p.m.