R_Studio(關聯)Apriori演算法尋找頻繁項集的方法
阿新 • • 發佈：2018-11-03
使用Apriori演算法尋找頻繁項集
#匯入arules包 install.packages("arules") library ( arules ) setwd('D:\\data') Gary<- read.transactions("Garytxt.txt", format = "basket", sep=",") #設定頻繁項集元素個數大於等於2 GarySize<-size(Gary) Gary_u<-Gary[GarySize>1] #檢視部分規則Gary.Rsummary(Gary_u) #支援度0.4,置信度0.5,過濾掉lhs為空的規則 Gary_u=apriori(Gary_u,parameter=list(support=0.4,confidence=0.5,minlen=2)) #Gary_u=apriori(Gary_u,parameter=list(support=0.4,confidence=0.5),appearance=list(rhs=c("a"),default="lhs")) #檢視部分規則 inspect(Gary_u)
實現過程
匯入arules包
資料預處理
#匯入arules包 install.packages("arules") library ( arules ) setwd('D:\\data') Gary<- read.transactions("Garytxt.txt", format = "basket", sep=",") #設定頻繁項集元素個數大於等於2 GarySize<-size(Gary) Gary_u<-Gary[GarySize>1] #檢視部分規則
檢視Gary_u中的資料
> summary(Gary_u) transactions as itemMatrixin sparse format with 6 rows (elements/itemsets/transactions) and 5 columns (items) and a density of 0.7666667 most frequent items: B E A C D (Other) 6 5 4 4 4 0 element (itemset/transaction) length distribution: sizes 3 4 5 2 3 1 Min. 1st Qu. Median Mean 3rd Qu. Max. 3.000 3.250 4.000 3.833 4.000 5.000 includes extended item information - examples: labels 1 A 2 B 3 C
生成關聯規則
> #Gary_u=apriori(Gary_u,parameter=list(support=0.4,confidence=0.5),appearance=list(rhs=c("a"),default="lhs")) > > > #檢視部分規則 > inspect(Gary_u) lhs rhs support confidence lift count [1] {C} => {E} 0.5000000 0.7500000 0.900 3 [2] {E} => {C} 0.5000000 0.6000000 0.900 3 [3] {C} => {B} 0.6666667 1.0000000 1.000 4 [4] {B} => {C} 0.6666667 0.6666667 1.000 4 [5] {D} => {A} 0.5000000 0.7500000 1.125 3 [6] {A} => {D} 0.5000000 0.7500000 1.125 3 [7] {D} => {E} 0.5000000 0.7500000 0.900 3 [8] {E} => {D} 0.5000000 0.6000000 0.900 3 [9] {D} => {B} 0.6666667 1.0000000 1.000 4 [10] {B} => {D} 0.6666667 0.6666667 1.000 4 [11] {A} => {E} 0.6666667 1.0000000 1.200 4 [12] {E} => {A} 0.6666667 0.8000000 1.200 4 [13] {A} => {B} 0.6666667 1.0000000 1.000 4 [14] {B} => {A} 0.6666667 0.6666667 1.000 4 [15] {E} => {B} 0.8333333 1.0000000 1.000 5 [16] {B} => {E} 0.8333333 0.8333333 1.000 5 [17] {C,E} => {B} 0.5000000 1.0000000 1.000 3 [18] {B,C} => {E} 0.5000000 0.7500000 0.900 3 [19] {B,E} => {C} 0.5000000 0.6000000 0.900 3 [20] {A,D} => {E} 0.5000000 1.0000000 1.200 3 [21] {D,E} => {A} 0.5000000 1.0000000 1.500 3 [22] {A,E} => {D} 0.5000000 0.7500000 1.125 3 [23] {A,D} => {B} 0.5000000 1.0000000 1.000 3 [24] {B,D} => {A} 0.5000000 0.7500000 1.125 3 [25] {A,B} => {D} 0.5000000 0.7500000 1.125 3 [26] {D,E} => {B} 0.5000000 1.0000000 1.000 3 [27] {B,D} => {E} 0.5000000 0.7500000 0.900 3 [28] {B,E} => {D} 0.5000000 0.6000000 0.900 3 [29] {A,E} => {B} 0.6666667 1.0000000 1.000 4 [30] {A,B} => {E} 0.6666667 1.0000000 1.200 4 [31] {B,E} => {A} 0.6666667 0.8000000 1.200 4 [32] {A,D,E} => {B} 0.5000000 1.0000000 1.000 3 [33] {A,B,D} => {E} 0.5000000 1.0000000 1.200 3 [34] {B,D,E} => {A} 0.5000000 1.0000000 1.500 3 [35] {A,B,E} => {D} 0.5000000 0.7500000 1.125 3