R 무방향성 데이터분석_연관규칙

install.packages('arulesViz') library(arulesViz) data('Groceries') summary(Groceries) head(Groceries) #트랜젝션 데이터라 바로 볼 수X 데이터프레임 변환 필요. gro=as(Groceries,'data.frame') ##R연관규칙 생성 및 주요 규칙확인 rules=apriori(Groceries,parameter=list(support=0.001,confidence=0.5)) #apriori가 A살 때 B산다는 규칙을 만들어주는 함수 rules inspect(head(sort(rules,by='lift'),3)) #lift값으로 정렬해서 상위 3개 확인 inspect(head(sort(rules,by='lift'),5)) plot(rules) subrules2=head(sort(rules,by='lift'),10) plot(subrules2,method = 'graph',control=list(type='items')) gro=as(Groceries,'data.frame') View(gro) install.packages('arules') library(arules) update.packages() data(Adult) summary(Adult) Adult.df <- as(Adult, 'data.frame') Adult.df rules.adult<-apriori(Adult) inspect(head(rules.adult)) #rhs rights hands sides left가 먼저 A를 살때 B살 확률에서 A가 left hands rules.adult1=apriori(Adult,parameter=list(support=0.1,confidence=0.6),                      appearance = list(rhs=c('income=small','income=large'),                                        default='lhs'),control=list(verbose=F)) rules.adult1.sorted=sort(rules.adult1,by='lift') inspect(head(rules.adult1.sorted)) plot(rules.adult1.sorted,method='scatterplot') plot(rules.adult1.sorted,method='graph',control=list(type='items',alpha=0.5)) a10=head(sort(rules.adult1,by='lift'),10) plot(a10,method='graph',control=list(type='items',alpha=1)) ?plot titan.df=as.data.frame(Titanic) head(titan.df) summary(Titanic) titanic=NULL for(i in 1:4){titanic = cbind(titanic, rep(as.character(titan.df[,i]),                                            titan.df$Freq))} titanic=as.data.frame(titanic) names(titanic)=names(titan.df)[1:4] summary(titanic) head(titanic);tail(titanic) rules.all=apriori(titanic)  #apriori로 만들어주고 inspect로 본다 options(digits=3) inspect(rules.all) #규칙의 우변(rhs)가 생존여부와 관계된 규칙 #설정값 변경 최소부분집합크기=2, 최소지지도=0.005, 최소신뢰도=0.8 rules=apriori(titanic,control = list(verbose=F), parameter=list(minlen=2,supp=0.005,conf=0.8),               appearance=list(rhs=c("Survived=No",'Survived=Yes'),                               default='lhs')) rules.sorted=sort(rules,by='lift')  #향상도(lift)기준으로 정렬 inspect(rules.sorted)  #규칙확인 #중복되는 규칙 찾기 subset.matrix=is.subset(rules.sorted,rules.sorted) subset.matrix[lower.tri(subset.matrix,diag=T)]<-FALSE redundant=colSums(subset.matrix,na.rm=T)>=1 which(redundant) rules.pruned<-rules.sorted[!redundant] plot(rules.all) #디폴트 값 measure=c('support','confidence'),shading='lift' plot(rules.all,shading='order') #규칙번호에 따라 음영 부여 plot(rules.all, method='grouped') #음영이 진할 수록 리프트가 높음 plot(rules.sorted) plot(rules.sorted,method='grouped') plot(rules.sorted,method='paracoord',control=list(reorder=T)) #평행좌표그림 #interactive 그림 plot(rules.sorted,measure=c('support','lift'),shading='confidence',interactive=T) #행렬-기반 시각화 plot(rules.sorted,method='matrix',measure='lift') plot(rules.sorted,method='matrix',measure=c('lift','confidence'),control = list(reorder=TRUE)) plot(rules.sorted,method='matrix3D',measure='lift',control = list(reorder=TRUE)) ##파일로저장 #write #연습문제 adult78<-apriori(Adult,parameter=list(support=0.7,confidence=0.8)) mybasket=read.transactions('mybasket.csv') summary(mybasket) head(mybasket) str(mybasket) itemFrequency(mybasket) itemFrequencyPlot(mybasket) myb=apriori(mybasket) inspect(tail(myb)) my10<-apriori(mybasket,parameter=list(support=0.1,confidence=0.0),) sort(itemFrequency(my10),decreasing = T) sort(itemFrequency(mybasket)) rules=apriori(my10,control = list(verbose=F),               appearance=list(rhs=("bakery"),                               lhs=('snack'))) inspect9(my10, ) #----------------------------------------------------물어보기 mybasket.trans <- read.transactions("mybasket.csv",format="basket",sep=",") summary(mybasket.trans) itemFrequency(mybasket.trans)  mybasket.rules <- apriori(mybasket.trans,parameter=list(support=0.1,confidence=0.0))  summary(mybasket.rules) sort(itemFrequency(mybasket.trans),decreasing = TRUE) inspect(subset(mybasket.rules,subset=lhs %in% "clothes"& rhs %in% "snack")) inspect(subset(mybasket.rules,subst=lhs%in% "clothes" & lhs %in% "snack")) install.packages('sna') library(sna) amatrix=rbind(c(0,1,1,1,0),c(1,0,0,1,0),c(1,0,0,0,0),c(1,1,0,0,1),c(0,0,0,1,0)) amatrix gden(amatrix) #gden 네트워크 밀도 : 0.5 50%만 연결되었다는 뜻 degree(amatrix,gmode='graph') #degree 연결중심성 : 링크 구하는 것 closeness(amatrix,gmode='graph') #거리의 평균 >거점물류센터에서 많이 쓰임 betweenness(amatrix,gmode='graph') gplot(amatrix) gplot(amatrix,mode='circle') plot.sociomatrix(amatrix)