library(bitops) library(RCurl) library(KoNLP) library(rJava) library(tm) library(wordcloud) library(XLConnect) setwd ("D:/Users/Hyo/Cs-Ds/CS/MusicStudy/data") rm(list=ls()) music<- file.path("km.xlsx") music <- readWorksheetFromFile(music, sheet="1970s") lyrics<- Corpus(VectorSource(music$lyrics)) result.text <- lyrics result.text <- gsub("\n", " ", result.text) result.text <- gsub("\r", " ", result.text) result.text <- gsub("별들", "별", result.text) result.text <- gsub("기둥이", "기둥", result.text) result.text <- gsub("마나님", "마누라", result.text) result.text <- gsub("딸들아", "딸", result.text) result.text <- gsub("따님", "딸", result.text) result.text <- gsub("님이여", "님", result.text) result.text <- gsub("누군가", "누구", result.text) result.text <- gsub("누군지", "누구", result.text) result.text <- gsub("누굴", "누구", result.text) result.text <- gsub("내게로", "내", result.text) result.text <- gsub("내게서", "내", result.text) result.text <- gsub("나는야", "나", result.text) result.text <- gsub("먹구름아", "먹구름", result.text) result.text <- gsub("말씀", "말", result.text) result.text <- gsub("말해", "말", result.text) result.text <- gsub("무지갯", "무지개", result.text) result.text <- gsub("홀로이", "혼자", result.text) result.text <- gsub("물결", "물", result.text) result.text <- gsub("바람결", "바람", result.text) result.text <- gsub("바람불어", "바람", result.text) result.text <- gsub("사람들", "사람", result.text) result.text <- gsub("사람들아", "사람", result.text) result.text <- gsub("사랑이었기", "사랑", result.text) result.text <- gsub("사랑이었기에", "사랑", result.text) result.text <- gsub("사위감", "사위", result.text) result.text <- gsub("어디론지", "어디", result.text) result.text <- gsub("어딘가", "어디", result.text) result.text <- gsub("어디선가", "어디", result.text) result.text <- gsub("어머님", "어머니", result.text) result.text <- gsub("여울지는", "여울", result.text) result.text <- gsub("우리꺼라면", "우리", result.text) result.text <- gsub("우리들", "우리", result.text) result.text <- gsub("호숫가", "호수", result.text) result.text <- gsub("햇볕", "햇살", result.text) result.text <- gsub("햇빛", "햇살", result.text) result.text <- gsub("햇살", "햇살 ", result.text) result.text <- gsub("최진사네", "최진사", result.text) result.text <- gsub("☜", " ", result.text) result.text <- gsub("♧", " ", result.text) result.text <- gsub("♨", " ", result.text) result.text <- gsub("♬", " ", result.text) result.text <- gsub("①", " ", result.text) result.text <- gsub("②", " ", result.text) result.text <- gsub("③", " ", result.text) result.text <- gsub("⊙", " ", result.text) result.text <- gsub("☞", " ", result.text) result.text <- gsub(",", ", ", result.text) result.text <- gsub("\\.", "\\. ", result.text) result.text <- gsub("\\+", " ", result.text) result.text <- gsub("\\-", " ", result.text) result.text <- gsub("\\:", " ", result.text) result.text <- gsub("\\(", " ", result.text) result.text <- gsub("\\)", " ", result.text) result.text <- gsub(" \n", "\n", result.text) result.text <- gsub("=", " ", result.text) result.text <- gsub("~", " ", result.text) result.text <- gsub("^_^", " ", result.text) result.text <- gsub("^ ", "", result.text) result.text <- gsub(" $", "", result.text) removeTwitSign <- function(x) { gsub("@[[:graph:]]*","",x) } removeURL <- function(x) { gsub("http://[[:graph:]]*","",x) } removeEnter <- function(x) { gsub("\n","",x) } exNouns <- function(x) { paste(extractNoun(x), collapse=" ")} # NA -> "" 로 변환 result.text[is.na(result.text)] <- "" result.text <- gsub("[[:punct:]]", "", result.text) result.text <- gsub(" $", "", result.text) useSejongDic() mergeUserDic(data.frame(c( "개여울", "가시내", "꽃가마", "윙크", "꼴", "마로니에", "먹구름", "서울", "그녀", "그대", "이쁜이", "이곳", "집시", "가고파", "기둥", "꿈길", "둘러메다", "불다", "서귀포", "서늘하다", "서해", "싸늘하다", "아이", "토라지다" ), c("ncn"))) result_nouns <- sapply(result.text, exNouns) myCorpus <- Corpus(VectorSource(result_nouns)) myCorpus <- tm_map(myCorpus, removePunctuation) myCorpus <- tm_map(myCorpus, removeNumbers) myCorpus <- tm_map(myCorpus, tolower) myStopwords <- c(stopwords('english'), "rt", "토라져", "토라졌을까", "토라지면", "가려마", "거지", "기둥지요", "간다네", "간다", "가에", "가노라", "가노라면", "가노", "가고파", "괴고", "괜스레", "껄껄껄", "나즈막히", "날으네", "넘겨주소", "넙죽", "남아", "내고", "내리소서", "닐리리야", "닐리리", "뚜루", "듯", "되었다네", "되", "들이", "랄랄라랄라", "랄랄라", "랄랄", "랄라라랄라", "랄라", "랄랄랄라", "랄랄랄랄라", "로운", "루루루", "루루", "라", "라라라라", "라라라라라", "라라라라랄랄라", "라라랄랄라", "라랄라", "라랄라라", "라랄랄라", "라오", "랄랄라랄랄라", "랄랄랄", "루루루루", "르", "못할게", "모른다네", "만났소", "만나", "마져", "마주", "모른다네", "마다", "만큼", "잊으라시면", "워우워우워", "이래", "이럴까", "왜냐고", "오질않네", "예예예예", "예예예", "여릿", "에헤야데헤야", "에헤야", "어쩌다", "어솨야", "어떠냐고", "아아아", "아롱대는", "예예예", "어서와대헤야", "얼씨구나", "없다", "없다네", "없을걸", "원치", "싶소", "싸늘", "스리", "스무", "스톨라", "소근대는", "살겠소", "살고파요", "살으렵니다", "뺏는고", "뺏는", "봤다나요", "뵙고", "부르리까", "보이잖니", "보이", "보았소", "보고", "보고파질", "변치말자", "부세요", "부세","반기", "버릴래", "품으렵니다", "품파", "품품품", "절씨구나", "주려무나", "좋을시고", "짧다란", "짧기만한", "터른", "토록", "필리리이", "처럼", "하게", "하고파", "하고파도", "하고픈", "했다네", "했었오", "헬로아", "흠마흠마흠마", "흠마흠마흠마흠마예", "해볼걸", "해서", "해도", "함", "할걸", "핼쓱한", "하렴아", "하리", "하리까", "하소", "하오", "한단", "한데", "라라라" ) myCorpus <-tm_map(myCorpus, removeWords, myStopwords) inspect(myCorpus[1:5]) myTdm2 <- TermDocumentMatrix(myCorpus, control=list(wordLengths=c(2,Inf))) myTdm1 <- TermDocumentMatrix(myCorpus, control=list(wordLengths=c(1,Inf))) mat2 <- as.data.frame(as.matrix(myTdm2)) mat1 <- as.data.frame(as.matrix(myTdm1)) write.table(mat2, file="_lyrics_70s_2.txt", col.names=FALSE, row.names=TRUE,sep="\t") write.table(mat1, file="_lyrics_70s_1.txt", col.names=FALSE, row.names=TRUE,sep="\t") myTdm <- myTdm1 pal <- brewer.pal(8,"Dark2") # 폰트 세팅. 띄어쓰기나 대소문자에 민감하다는 점에 주의 # 맑은고딕 : windowsFonts(malgun=windowsFont("맑은 고딕")) # 나눔고딕 : windowsFonts(malgun=windowsFont("나눔고딕")) windowsFonts(malgun=windowsFont("맑은 고딕")) m <- as.matrix(myTdm) # calculate the frequency of words v <- sort(rowSums(m), decreasing=TRUE) myNames <- names(v) k <- which(names(v)=="apple") myNames[k] <- "apple" d <- data.frame(word=myNames, freq=v) #wordcloud(d$word, d$freq, scale=c(4,0.5), min.freq=3, random.order=F, rot.per=.1, family="malgun") wordcloud(d$word, d$freq, scale=c(7,0.8), min.freq=5, random.order=F, rot.per=.1, colors=pal, family="malgun") dev.copy(png,"1970s-1.png",width=8,height=6,units="in",res=200) dev.off() myTdm <- myTdm2 pal <- brewer.pal(8,"Dark2") # 폰트 세팅. 띄어쓰기나 대소문자에 민감하다는 점에 주의 # 맑은고딕 : windowsFonts(malgun=windowsFont("맑은 고딕")) # 나눔고딕 : windowsFonts(malgun=windowsFont("나눔고딕")) windowsFonts(malgun=windowsFont("맑은 고딕")) m <- as.matrix(myTdm) # calculate the frequency of words v <- sort(rowSums(m), decreasing=TRUE) myNames <- names(v) k <- which(names(v)=="apple") myNames[k] <- "apple" d <- data.frame(word=myNames, freq=v) #wordcloud(d$word, d$freq, scale=c(4,0.5), min.freq=3, random.order=F, rot.per=.1, family="malgun") wordcloud(d$word, d$freq, scale=c(7,0.8), min.freq=5, random.order=F, rot.per=.1, colors=pal, family="malgun") dev.copy(png,"1970s-2.png",width=8,height=6,units="in",res=200) dev.off()