Created
June 5, 2017 10:29
-
-
Save chiral/4f7708dddd2bb4bc4889b31b9e7ce7c9 to your computer and use it in GitHub Desktop.
comparison between LogisticRegression and RandomForest with "Census Income" dataset. please refer to http://archive.ics.uci.edu/ml/datasets/Adult
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
library(randomForest) | |
library(ROCR) | |
init <- function() { | |
df1 <- read.csv("adult.data",header=F) | |
df2 <- read.csv("adult.test",header=F) | |
df2$V15 <- gsub("\\.$","",df2$V15) | |
df <- rbind(df1,df2) | |
write.csv(df,"adult.all",row.names=F) | |
return(c(nrow(df1),nrow(df2))) | |
} | |
#--comment out this block only the first time-- | |
#sz <- init() | |
#df <- read.csv("adult.all",header=T) | |
#train <- df[1:sz[1],] | |
#test <- df[sz[1]+(1:sz[2]),] | |
#rm(df) | |
#--comment out this block only the first time-- | |
model1 <- glm(V15 ~.,family=binomial(link='logit'),data=train) | |
result1 <- predict(model1,newdata=test[,-15],type="response") | |
pred1 <- prediction(result1,test[,15]) | |
perf1 <- performance(pred1,"prec","rec") | |
plot(perf1,colorize = TRUE) | |
model2 <- randomForest(V15 ~.,data=train,mtry=2) | |
result2 <- predict(model2,newdata=test[,-15],type="prob") | |
pred2 <- prediction(result2[,2],test[,15]) | |
perf2 <- performance(pred2,"prec","rec") | |
plot(perf2,add=TRUE,colorize = TRUE) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment