随机森林代码

library(randomForest)
totaldata<-read.table("/export/tangjie/leak_change.txt",header = FALSE,sep = '\t')
names(totaldata)<-c('pin','p_book','n_book','m_book','office','music','luxury','fitness','adult_items','trip','user_level',
'gendar','mobile_pay_cnt','valid_credit_card_cnt','complain_cnt','ord_midnight_cnt','ord_night_cnt',
'ord_worktime_cnt','rej_cnt','rej_pct','rtn_cnt','rtn_pct','allprice_lv','login_pct'
,'complain_pct','mobile_pay_pct','pop_ord_cnt_pct','cart_ord_pct','trv_amt_pct',
'jsd_pct','city_rank','jr_jk','jr_others','tag','used_in1year','ent_amt','zt_ord_pct','follow_cnt','follower_cnt',
'following_cnt','follow_score','comment_sc','zx_ord_pct','df_ord_pct','lt_game_pct',
'sw_ord_pct','dis_pct','cnt_stab_cv','ord_cnt_per_mon_12','mult_ords','jq_dq_cnt_pct',
'jingdou_jqdq_cnt_pct','tiny_ord_pct'
)
str(totaldata)
totaldata$allprice_lv  =as.numeric(totaldata$allprice_lv)
totaldata$trv_amt_pct  =as.numeric(totaldata$trv_amt_pct)
ind<-sample(2,nrow(totaldata),replace=TRUE,prob=c(0.7,0.3))
traindata<- totaldata[ind==1,]
testdata<- totaldata[ind==2,]
rf <- randomForest(tag ~ p_book + n_book + m_book + office +
music + luxury + fitness + adult_items + trip + user_level +
gendar + mobile_pay_cnt + valid_credit_card_cnt + complain_cnt
+ ord_midnight_cnt + ord_night_cnt + ord_worktime_cnt + rej_cnt +
rej_pct + rtn_cnt + rtn_pct + allprice_lv + login_pct + complain_pct
+ mobile_pay_pct + pop_ord_cnt_pct + cart_ord_pct + trv_amt_pct + jsd_pct
+ city_rank + jr_jk + jr_others + used_in1year + ent_amt + zt_ord_pct +
follow_cnt + follower_cnt + follow_score + comment_sc + zx_ord_pct + df_ord_pct
+ lt_game_pct + sw_ord_pct + dis_pct + cnt_stab_cv + ord_cnt_per_mon_12 + mult_ords
+ jq_dq_cnt_pct + jingdou_jqdq_cnt_pct + tiny_ord_pct, data=traindata, ntree=100, proximity=TRUE)
rf.pred <- predict(rf,testdata)
table(observed=testdata$tag,predicted=rf.pred )