PREPARATION
setwd("~/Google Drive/UM/Smart Services/Thesis/Thesis/Code/Naive Bayes/4.Feature Set 3/Negations")
#install.packages("naivebayes")
library(naivebayes)
## Warning: package 'naivebayes' was built under R version 3.4.3
library(dplyr)
## Warning: Installed Rcpp (0.12.16) different from Rcpp used to build dplyr (0.12.11).
## Please reinstall dplyr to avoid random crashes or undefined behavior.
##
## Attaching package: 'dplyr'
## The following objects are masked from 'package:stats':
##
## filter, lag
## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union
library(psych)
library(ggplot2)
##
## Attaching package: 'ggplot2'
## The following objects are masked from 'package:psych':
##
## %+%, alpha
library(e1071)
library(readxl)
#Import Labels
Labels <- read_excel("~/Google Drive/UM/Smart Services/Thesis/Thesis/Code/Naive Bayes/1.Labels/Source Data.xlsx")
Label <- Labels$Score
#Import Features
Features <- read.csv("~/Google Drive/UM/Smart Services/Thesis/Thesis/Code/Naive Bayes/4.Feature Set 3/Negations/Feature Set 3 TF.csv")
Features <- Features[-1]
RECODE LABELS FOR ONE-VS-ALL
#Class 2
Label2 <- list()
for(i in 1:1000){
if(Label[i]==3| Label[i]==4){
Label2[i] <- 1
}else{
Label2[i] <- 0
}
}
#As Factor
Label2 <- as.factor(unlist(Label2))
#Class 3
Label3 <- list()
for(i in 1:1000){
if(Label[i]==5| Label[i]==6){
Label3[i] <- 1
}else{
Label3[i] <- 0
}
}
#As Factor
Label3 <- as.factor(unlist(Label3))
#Class 4
Label4 <- list()
for(i in 1:1000){
if(Label[i]==7| Label[i]==8){
Label4[i] <- 1
}else{
Label4[i] <- 0
}
}
#As Factor
Label4 <- as.factor(unlist(Label4))
#Class 5
Label5 <- list()
for(i in 1:1000){
if(Label[i]==9| Label[i]==10){
Label5[i] <- 1
}else{
Label5[i] <- 0
}
}
#As Factor
Label5 <- as.factor(unlist(Label5))
#All Labels
All <- list()
for(i in 1:1000){
if(Label[i]==9| Label[i]==10){
All[i] <- 5
}else if(Label[i]==7| Label[i]==8){
All[i] <- 4
}else if(Label[i]==5| Label[i]==6){
All[i] <- 3
}else{
All[i] <- 2
}
}
#As Factor
All <- as.factor(unlist(All))
#Control
Control.df <- data.frame(matrix(seq(1,1000),ncol=1,nrow=1000))
Control.df$Actual <- Label
Control.df$All <- All
Control.df$Label2 <- Label2
Control.df$Label3 <- Label3
Control.df$Label4 <- Label4
Control.df$Label5 <- Label5
Control.df[1:10,2:7]
## Actual All Label2 Label3 Label4 Label5
## 1 3 2 1 0 0 0
## 2 8 4 0 0 1 0
## 3 7 4 0 0 1 0
## 4 4 2 1 0 0 0
## 5 7 4 0 0 1 0
## 6 7 4 0 0 1 0
## 7 5 3 0 1 0 0
## 8 10 5 0 0 0 1
## 9 7 4 0 0 1 0
## 10 8 4 0 0 1 0
TRANSFORM FEATURES TO FACTOR VARIABLES
#Transform Integer to Factor
for(i in 1:417){
Features[,i] <- as.factor(Features[,i])
}
str(Features)
## 'data.frame': 1000 obs. of 417 variables:
## $ amaz_jj : Factor w/ 2 levels "0","1": 1 1 1 1 2 1 1 1 1 1 ...
## $ arriv_jj : Factor w/ 2 levels "0","1": 2 1 1 1 1 1 1 1 1 1 ...
## $ bad_jj : Factor w/ 3 levels "0","1","2": 1 1 1 2 1 1 1 1 1 1 ...
## $ basic_jj : Factor w/ 3 levels "0","1","2": 1 1 2 1 1 1 1 1 1 1 ...
## $ beauti_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ befor_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ best_jjs : Factor w/ 2 levels "0","1": 2 1 1 1 1 1 1 1 1 1 ...
## $ big_jj : Factor w/ 3 levels "0","1","2": 3 1 1 1 1 1 1 1 1 3 ...
## $ build_jj : Factor w/ 3 levels "0","1","2": 1 1 1 2 1 1 1 1 1 1 ...
## $ central_jj : Factor w/ 4 levels "0","1","2","3": 1 1 1 1 1 1 1 1 1 1 ...
## $ clean_jj : Factor w/ 5 levels "0","1","2","3",..: 1 1 1 2 1 1 2 1 2 1 ...
## $ clear_jj : Factor w/ 2 levels "0","1": 1 1 1 1 2 1 1 1 1 1 ...
## $ close_jj : Factor w/ 3 levels "0","1","2": 2 1 1 1 1 1 1 1 1 1 ...
## $ cold_jj : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 1 1 ...
## $ difficult_jj : Factor w/ 3 levels "0","1","2": 1 1 2 1 1 1 1 1 1 1 ...
## $ due_jj : Factor w/ 3 levels "0","1","2": 2 1 1 1 1 1 1 1 1 1 ...
## $ earl_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ easi_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ english_jj : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 1 1 ...
## $ enough_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ excel_jj : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 1 1 ...
## $ extra_jj : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 1 1 ...
## $ first_jj : Factor w/ 3 levels "0","1","2": 1 2 1 1 1 1 1 1 1 1 ...
## $ free_jj : Factor w/ 4 levels "0","1","2","5": 1 1 1 1 1 1 1 1 1 1 ...
## $ fresh_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ friend_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 2 1 1 ...
## $ front_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ full_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ general_jj : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 1 1 ...
## $ good_jj : Factor w/ 6 levels "0","1","2","3",..: 1 1 2 1 1 2 1 3 1 4 ...
## $ great_jj : Factor w/ 6 levels "0","1","2","3",..: 1 4 1 2 1 3 1 1 1 1 ...
## $ guest_jjs : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ high_jj : Factor w/ 3 levels "0","1","2": 3 1 1 1 1 1 1 2 1 1 ...
## $ hot_jj : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 1 1 ...
## $ huge_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ littl_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ locat_jj : Factor w/ 2 levels "0","1": 1 1 1 2 1 1 1 1 1 1 ...
## $ london_jj : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 1 1 ...
## $ loud_jj : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 1 1 ...
## $ main_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 2 1 1 1 1 ...
## $ major_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ modern_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 2 1 1 1 1 ...
## $ much_jj : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 1 2 ...
## $ new_jj : Factor w/ 3 levels "0","1","2": 2 2 1 1 1 1 1 1 1 1 ...
## $ next_jj : Factor w/ 4 levels "0","1","2","3": 4 1 1 2 1 1 1 1 1 1 ...
## $ nice_jj : Factor w/ 5 levels "0","1","2","3",..: 1 1 2 4 1 1 1 1 1 1 ...
## $ nois_jj : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 1 1 ...
## $ noisi_jj : Factor w/ 2 levels "0","1": 1 1 1 2 1 1 1 1 1 1 ...
## $ ok_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ old_jj : Factor w/ 3 levels "0","1","2": 1 1 1 2 1 1 1 1 1 1 ...
## $ onli_jj : Factor w/ 3 levels "0","1","2": 2 1 1 1 1 1 1 1 1 1 ...
## $ open_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ overal_jj : Factor w/ 2 levels "0","1": 1 1 1 2 1 1 1 1 1 1 ...
## $ particular_jj : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 1 1 ...
## $ perfect_jj : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 1 1 ...
## $ pillow_jj : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 1 1 ...
## $ pleasant_jj : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 1 1 ...
## $ poor_jj : Factor w/ 3 levels "0","1","3": 1 1 1 1 1 1 1 1 1 1 ...
## $ public_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ quiet_jj : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 2 1 1 1 ...
## $ realli_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ recept_jj : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 1 1 ...
## $ safe_jj : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 1 1 ...
## $ second_jj : Factor w/ 3 levels "0","1","2": 1 2 1 2 1 1 1 1 1 1 ...
## $ select_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ servic_jj : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 1 2 ...
## $ short_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ shower_jjr : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 1 1 ...
## $ sleep_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ small_jj : Factor w/ 4 levels "0","1","2","3": 2 1 1 1 1 1 1 1 1 1 ...
## $ spacious_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 2 1 1 1 ...
## $ special_jj : Factor w/ 2 levels "0","1": 2 1 1 1 1 1 1 1 1 1 ...
## $ standard_jj : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 1 1 ...
## $ stay_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ steep_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ super_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ sure_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ underground_jj: Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ upgrad_jj : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 1 1 ...
## $ veri_jj : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 2 1 ...
## $ warm_jj : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ whole_jj : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 1 1 ...
## $ ask_vb : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ bed_vbd : Factor w/ 3 levels "0","1","2": 1 1 1 2 1 1 1 1 1 2 ...
## $ build_vb : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ came_vbd : Factor w/ 3 levels "0","1","2": 2 1 1 3 1 1 1 1 1 1 ...
## $ check_vb : Factor w/ 2 levels "0","1": 2 1 1 1 1 1 1 1 1 1 ...
## $ definit_vb : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ done_vbn : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ expens_vbz : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ gave_vbd : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ get_vb : Factor w/ 3 levels "0","1","2": 1 1 1 1 3 1 1 1 1 1 ...
## $ given_vbn : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 1 1 ...
## $ go_vb : Factor w/ 3 levels "0","1","2": 1 1 2 1 1 1 1 1 1 1 ...
## $ go_vbp : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 1 1 ...
## $ got_vbd : Factor w/ 4 levels "0","1","2","3": 3 1 1 1 1 1 1 1 1 1 ...
## $ like_vb : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ love_vb : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 1 1 ...
## $ made_vbd : Factor w/ 3 levels "0","1","2": 3 1 1 1 1 1 2 1 1 1 ...
## [list output truncated]
PARTITIONING TRAINING & VALIDATION
#Features
set.seed(1234)
ind <- sample(2,nrow(Features),replace = T, prob =c(0.8,0.2))
train <- Features[ind == 1,]
test <- Features[ind ==2,]
Labels
train.labels.2 <- Label2[ind == 1]
test.labels.2 <- Label2[ind ==2]
train.labels.3 <- Label3[ind == 1]
test.labels.3 <- Label3[ind ==2]
train.labels.4 <- Label4[ind == 1]
test.labels.4 <- Label4[ind ==2]
train.labels.5 <- Label5[ind == 1]
test.labels.5 <- Label5[ind ==2]
train.labels <- All[ind == 1]
test.labels <- All[ind ==2]
NAIVE BAYES MODEL
NB2 <- naiveBayes(x = train,y = train.labels.2)
NB3 <- naiveBayes(x = train,y = train.labels.3)
NB4 <- naiveBayes(x = train,y = train.labels.4)
NB5 <- naiveBayes(x = train,y = train.labels.5)
PREDICTIONS
NB.Pred2 <- predict(NB2, test,type ="raw")
NB.Pred3 <- predict(NB3, test,type ="raw")
NB.Pred4 <- predict(NB4, test,type ="raw")
NB.Pred5 <- predict(NB5, test,type ="raw")
VOTING
Voting.df <- data.frame(NB.Pred2, NB.Pred3,NB.Pred4,NB.Pred5)
colnames(Voting.df) <- c("Class 2: 0","Class2: 1","Class 3: 0","Class3: 1","Class 4: 0","Class4: 1","Class 5: 0","Class5: 1")
head(Voting.df)
## Class 2: 0 Class2: 1 Class 3: 0 Class3: 1 Class 4: 0 Class4: 1
## 1 0.9999997 2.675233e-07 0.80096246 1.990375e-01 0.19916559 0.8008344
## 2 1.0000000 3.166137e-10 0.99998524 1.475687e-05 0.02838203 0.9716180
## 3 1.0000000 2.069964e-09 0.99999444 5.559505e-06 0.30343888 0.6965611
## 4 1.0000000 8.664322e-12 0.97971270 2.028730e-02 0.88432958 0.1156704
## 5 0.9999940 6.013239e-06 0.84543275 1.545672e-01 0.68562162 0.3143784
## 6 1.0000000 2.077360e-10 0.08857613 9.114239e-01 0.63518382 0.3648162
## Class 5: 0 Class5: 1
## 1 0.9993310 0.0006689738
## 2 0.9278382 0.0721618438
## 3 0.8733417 0.1266582896
## 4 0.9880090 0.0119909750
## 5 0.8881522 0.1118477796
## 6 0.9762619 0.0237381484
Transformed.Voting.df <- Voting.df[seq(2,8,2)]
colnames(Transformed.Voting.df) <- c("2","3","4","5")
head(Transformed.Voting.df)
## 2 3 4 5
## 1 2.675233e-07 1.990375e-01 0.8008344 0.0006689738
## 2 3.166137e-10 1.475687e-05 0.9716180 0.0721618438
## 3 2.069964e-09 5.559505e-06 0.6965611 0.1266582896
## 4 8.664322e-12 2.028730e-02 0.1156704 0.0119909750
## 5 6.013239e-06 1.545672e-01 0.3143784 0.1118477796
## 6 2.077360e-10 9.114239e-01 0.3648162 0.0237381484
Evaluation <- Transformed.Voting.df
Index <- as.numeric(apply(Transformed.Voting.df,MARGIN = 1,which.max))
Index <- Index+1
Evaluation$Vote <- Index
Evaluation$Actual <- test.labels
head(Evaluation,100)
## 2 3 4 5 Vote Actual
## 1 2.675233e-07 1.990375e-01 8.008344e-01 6.689738e-04 4 4
## 2 3.166137e-10 1.475687e-05 9.716180e-01 7.216184e-02 4 5
## 3 2.069964e-09 5.559505e-06 6.965611e-01 1.266583e-01 4 5
## 4 8.664322e-12 2.028730e-02 1.156704e-01 1.199098e-02 4 4
## 5 6.013239e-06 1.545672e-01 3.143784e-01 1.118478e-01 4 4
## 6 2.077360e-10 9.114239e-01 3.648162e-01 2.373815e-02 3 4
## 7 2.721574e-17 1.475385e-10 2.295659e-05 2.673768e-01 5 5
## 8 2.912077e-18 1.704426e-03 9.808296e-01 3.745072e-05 4 3
## 9 6.798799e-09 8.254181e-09 3.524340e-01 6.505342e-01 5 5
## 10 9.696937e-01 1.290606e-05 4.669131e-02 4.205123e-04 2 3
## 11 1.272082e-09 6.973153e-07 6.859449e-01 6.432005e-01 4 4
## 12 5.400734e-06 2.154586e-03 1.434814e-01 2.402013e-01 5 3
## 13 2.939470e-06 1.387033e-03 4.069730e-02 8.827676e-01 5 5
## 14 4.724952e-05 1.219650e-02 4.415675e-01 1.159785e-01 4 4
## 15 4.539994e-05 1.608008e-02 6.862808e-01 1.576366e-02 4 4
## 16 1.740397e-09 1.445590e-01 1.836835e-01 2.206960e-01 5 5
## 17 5.084245e-10 9.922126e-07 9.634503e-01 5.569846e-03 4 4
## 18 2.306055e-07 2.996792e-03 1.353135e-01 9.471752e-01 5 5
## 19 1.231405e-03 2.206166e-01 6.569930e-01 4.702652e-03 4 4
## 20 1.483969e-09 2.193868e-04 8.049292e-01 5.553026e-01 4 2
## 21 2.692747e-07 4.620966e-03 2.278633e-02 9.284500e-02 5 4
## 22 3.816975e-13 3.777650e-04 5.363262e-02 9.818958e-01 5 4
## 23 5.151466e-06 1.237017e-01 9.230272e-01 3.887990e-08 4 3
## 24 5.336899e-10 1.095135e-06 3.738782e-01 9.437257e-01 5 5
## 25 4.955548e-05 6.050062e-02 4.661872e-01 7.188795e-02 4 4
## 26 1.788526e-07 4.084973e-03 1.356286e-01 3.779042e-01 5 5
## 27 2.831903e-08 7.435689e-02 1.921995e-01 1.661951e-01 4 4
## 28 1.085425e-04 1.592574e-03 1.755787e-02 9.359655e-01 5 5
## 29 9.486017e-10 9.369207e-05 3.846772e-01 1.195603e-03 4 3
## 30 6.951738e-01 3.946409e-01 4.216636e-01 1.069745e-08 2 3
## 31 1.227268e-07 2.802748e-04 6.726957e-03 9.969041e-01 5 5
## 32 9.026714e-13 2.084471e-05 3.303876e-04 9.997262e-01 5 5
## 33 7.797822e-14 3.060547e-03 3.924259e-01 8.814967e-01 5 5
## 34 6.793970e-09 4.186680e-01 1.145076e-02 5.253080e-03 3 3
## 35 2.020065e-05 1.106212e-02 5.745327e-01 6.818087e-02 4 4
## 36 1.525443e-06 3.453870e-01 5.107300e-01 2.068065e-02 4 4
## 37 2.632160e-04 2.222339e-01 8.219692e-01 1.618427e-04 4 5
## 38 4.837668e-01 9.999496e-01 3.050258e-01 9.255879e-07 3 3
## 39 7.103571e-15 4.154914e-05 1.233747e-03 9.995293e-01 5 5
## 40 6.136685e-10 4.618884e-03 3.793464e-01 5.436195e-03 4 3
## 41 1.110864e-11 1.752326e-05 1.049722e-03 9.991446e-01 5 5
## 42 3.901475e-15 4.419211e-04 9.953740e-01 1.041779e-03 4 4
## 43 7.200515e-17 1.784191e-12 3.715882e-02 3.865429e-01 5 5
## 44 3.015828e-06 3.234302e-02 4.552776e-01 4.613826e-01 5 3
## 45 5.428764e-07 7.660904e-01 8.866711e-01 8.389208e-03 4 5
## 46 1.953617e-03 3.568343e-03 2.858975e-01 1.044007e-01 4 4
## 47 2.318883e-06 3.738255e-02 2.888682e-01 7.698700e-01 5 5
## 48 1.398017e-08 1.407311e-02 4.521026e-01 8.125927e-01 5 3
## 49 3.434258e-08 1.116484e-03 6.279426e-02 9.922357e-01 5 5
## 50 1.558448e-05 4.200791e-02 1.627100e-01 6.907761e-01 5 5
## 51 3.961392e-09 1.338125e-02 5.975197e-02 9.766056e-01 5 4
## 52 6.670099e-07 2.848391e-03 6.853720e-01 3.831335e-01 4 5
## 53 5.792663e-10 5.653325e-03 7.052876e-03 9.959940e-01 5 4
## 54 2.158453e-07 1.440488e-04 9.835146e-02 9.595349e-01 5 4
## 55 1.606017e-06 3.007892e-01 2.561494e-01 5.014773e-01 5 4
## 56 2.431434e-07 9.048242e-03 7.710988e-03 9.904860e-01 5 5
## 57 6.315644e-06 1.170419e-02 1.712894e-01 9.362977e-01 5 2
## 58 4.067628e-08 4.167343e-05 1.038293e-01 9.859016e-01 5 5
## 59 6.383347e-09 3.937900e-04 2.609186e-02 9.969459e-01 5 4
## 60 1.316709e-04 7.233155e-01 7.773220e-01 4.099669e-03 4 4
## 61 1.790895e-06 8.754085e-03 2.345527e-01 8.791960e-01 5 5
## 62 6.916047e-07 3.207526e-02 1.956033e-01 8.702572e-01 5 5
## 63 1.401894e-03 4.968459e-02 8.727334e-01 5.937466e-02 4 4
## 64 1.014250e-07 2.188757e-04 1.538936e-01 9.735561e-01 5 5
## 65 8.106443e-10 2.411916e-03 4.723202e-02 9.889463e-01 5 5
## 66 6.299388e-07 1.085407e-02 2.867514e-01 8.640251e-01 5 5
## 67 3.373317e-06 6.029595e-02 2.051094e-01 7.267095e-01 5 3
## 68 1.471392e-05 9.780468e-02 1.618833e-01 7.116257e-01 5 3
## 69 1.701029e-07 9.273921e-02 2.144578e-01 3.912291e-01 5 4
## 70 8.700922e-13 1.529405e-02 7.568398e-01 1.147484e-01 4 4
## 71 4.318598e-07 9.242558e-05 7.864270e-02 9.887497e-01 5 5
## 72 3.448049e-08 9.368701e-04 2.580787e-01 9.452959e-01 5 5
## 73 1.241832e-04 1.700348e-02 1.869995e-01 7.259597e-01 5 2
## 74 2.624285e-05 5.611764e-02 1.006712e-01 8.733851e-01 5 4
## 75 9.545338e-06 2.964948e-04 8.313120e-01 3.980987e-01 4 4
## 76 1.034673e-14 6.348808e-05 8.730653e-01 3.915619e-01 4 5
## 77 1.567446e-11 2.084947e-04 5.663591e-02 2.721853e-02 4 3
## 78 2.784769e-12 4.635684e-06 9.478930e-01 3.420601e-01 4 4
## 79 1.256080e-10 7.333994e-05 1.300173e-03 2.634075e-02 5 5
## 80 6.201003e-08 5.194204e-05 5.951090e-02 9.964426e-01 5 5
## 81 4.331137e-11 5.879594e-05 2.441585e-01 8.884648e-01 5 5
## 82 4.584571e-11 3.169496e-05 4.668022e-01 5.047134e-01 5 4
## 83 1.473284e-12 5.380884e-05 3.214161e-05 2.394445e-03 5 5
## 84 1.656416e-07 8.102866e-03 5.092416e-03 9.914755e-01 5 5
## 85 1.366942e-09 1.046638e-01 3.698988e-01 1.207251e-01 4 5
## 86 2.999313e-08 9.495602e-01 7.756890e-01 4.693106e-06 3 3
## 87 2.514278e-07 3.271435e-01 2.650582e-01 1.596808e-01 3 4
## 88 2.041704e-09 1.385791e-07 4.400738e-03 4.680133e-02 5 5
## 89 9.266970e-16 9.100058e-08 7.568886e-03 9.998359e-01 5 5
## 90 5.970452e-14 6.452446e-04 2.344751e-01 9.601094e-01 5 3
## 91 8.486429e-12 4.236109e-02 3.003388e-02 9.684760e-01 5 3
## 92 1.351007e-08 1.361592e-03 3.568931e-01 1.313921e-01 4 5
## 93 1.859004e-16 8.210513e-02 9.077830e-01 3.817813e-02 4 4
## 94 6.646038e-10 1.259851e-13 1.100460e-02 7.556664e-01 5 5
## 95 4.193540e-07 8.747394e-03 5.764348e-01 3.415027e-01 4 5
## 96 2.262784e-15 1.358090e-05 7.994193e-01 7.707409e-01 4 5
## 97 5.308388e-10 4.772464e-02 6.137567e-03 9.621784e-01 5 5
## 98 1.131472e-07 2.615343e-02 1.620603e-02 8.066643e-01 5 5
## 99 2.777013e-01 1.317438e-02 1.912110e-02 4.127989e-01 5 2
## 100 6.898207e-09 2.490847e-01 4.923336e-01 3.439048e-01 4 5
CM <- table(Evaluation$Actual,Evaluation$Vote)
CM
##
## 2 3 4 5
## 2 0 0 3 4
## 3 2 5 6 12
## 4 0 4 33 25
## 5 0 2 19 93
#Proportions
Overall <- length(Evaluation$Actual)
Length2 <- length(which(Evaluation$Actual==2))
Length3 <- length(which(Evaluation$Actual==3))
Length4 <- length(which(Evaluation$Actual==4))
Length5 <- length(which(Evaluation$Actual==5))
#Accuracy
Accuracy <- sum(diag(CM))/sum(CM)
#Precision
Precision <- diag(CM)/rowSums(CM)
Precision <- (Precision[1]*Length2+Precision[2]*Length3+Precision[3]*Length4+Precision[4]*Length5)/208
#Recall
Recall <- diag(CM)/colSums(CM)
Recall <- (Recall[1]*Length2+Recall[2]*Length3+Recall[3]*Length4+Recall[4]*Length5)/208
Accuracy
## [1] 0.6298077
Precision
## 2
## 0.6298077
Recall
## 2
## 0.5962693