50k using Naive Bayes Model
# Libraries
library(naivebayes)
library(ggplot2)
## Warning: package 'ggplot2' was built under R version 3.5.1
library(caret)
## Warning: package 'caret' was built under R version 3.5.1
## Loading required package: lattice
library(psych)
## Warning: package 'psych' was built under R version 3.5.1
##
## Attaching package: 'psych'
## The following objects are masked from 'package:ggplot2':
##
## %+%, alpha
library(e1071)
## Warning: package 'e1071' was built under R version 3.5.1
# Data(Train)
train_sal <- read.csv(file.choose())
str(train_sal)
## 'data.frame': 30161 obs. of 14 variables:
## $ age : int 39 50 38 53 28 37 49 52 31 42 ...
## $ workclass : Factor w/ 7 levels " Federal-gov",..: 6 5 3 3 3 3 3 5 3 3 ...
## $ education : Factor w/ 16 levels " 10th"," 11th",..: 10 10 12 2 10 13 7 12 13 10 ...
## $ educationno : int 13 13 9 7 13 14 5 9 14 13 ...
## $ maritalstatus: Factor w/ 7 levels " Divorced"," Married-AF-spouse",..: 5 3 1 3 3 3 4 3 5 3 ...
## $ occupation : Factor w/ 14 levels " Adm-clerical",..: 1 4 6 6 10 4 8 4 10 4 ...
## $ relationship : Factor w/ 6 levels " Husband"," Not-in-family",..: 2 1 2 1 6 6 2 1 2 1 ...
## $ race : Factor w/ 5 levels " Amer-Indian-Eskimo",..: 5 5 5 3 3 5 3 5 5 5 ...
## $ sex : Factor w/ 2 levels " Female"," Male": 2 2 2 2 1 1 1 2 1 2 ...
## $ capitalgain : int 2174 0 0 0 0 0 0 0 14084 5178 ...
## $ capitalloss : int 0 0 0 0 0 0 0 0 0 0 ...
## $ hoursperweek : int 40 13 40 40 40 40 16 45 50 40 ...
## $ native : Factor w/ 40 levels " Cambodia"," Canada",..: 38 38 38 38 5 38 22 38 38 38 ...
## $ Salary : Factor w/ 2 levels " <=50K"," >50K": 1 1 1 1 1 1 1 2 2 2 ...
View(train_sal)
train_sal$educationno <- as.factor(train_sal$educationno)
class(train_sal)
## [1] "data.frame"
# Data(Test)
test_sal <- read.csv(file.choose())
str(test_sal)
## 'data.frame': 15060 obs. of 14 variables:
## $ age : int 25 38 28 44 34 63 24 55 65 36 ...
## $ workclass : Factor w/ 7 levels " Federal-gov",..: 3 3 2 3 3 5 3 3 3 1 ...
## $ education : Factor w/ 16 levels " 10th"," 11th",..: 2 12 8 16 1 15 16 6 12 10 ...
## $ educationno : int 7 9 12 10 6 15 10 4 9 13 ...
## $ maritalstatus: Factor w/ 7 levels " Divorced"," Married-AF-spouse",..: 5 3 3 3 5 3 5 3 3 3 ...
## $ occupation : Factor w/ 14 levels " Adm-clerical",..: 7 5 11 7 8 10 8 3 7 1 ...
## $ relationship : Factor w/ 6 levels " Husband"," Not-in-family",..: 4 1 1 1 2 1 5 1 1 1 ...
## $ race : Factor w/ 5 levels " Amer-Indian-Eskimo",..: 3 5 5 3 5 5 5 5 5 5 ...
## $ sex : Factor w/ 2 levels " Female"," Male": 2 2 2 2 2 2 1 2 2 2 ...
## $ capitalgain : int 0 0 0 7688 0 3103 0 0 6418 0 ...
## $ capitalloss : int 0 0 0 0 0 0 0 0 0 0 ...
## $ hoursperweek : int 40 50 40 40 30 32 40 10 40 40 ...
## $ native : Factor w/ 40 levels " Cambodia"," Canada",..: 38 38 38 38 38 38 38 38 38 38 ...
## $ Salary : Factor w/ 2 levels " <=50K"," >50K": 1 1 2 2 1 2 1 1 2 1 ...
View(test_sal)
test_sal$educationno <- as.factor(test_sal$educationno)
class(test_sal)
## [1] "data.frame"
#Visualization
# Plot and ggplot
ggplot(data=train_sal,aes(x=train_sal$Salary, y = train_sal$age, fill = train_sal$Salary)) +
geom_boxplot() +
ggtitle("Box Plot")
plot(train_sal$workclass,train_sal$Salary)
plot(train_sal$education,train_sal$Salary)
plot(train_sal$educationno,train_sal$Salary)
plot(train_sal$maritalstatus,train_sal$Salary)
plot(train_sal$occupation,train_sal$Salary)
plot(train_sal$relationship,train_sal$Salary)
plot(train_sal$race,train_sal$Salary)
plot(train_sal$sex,train_sal$Salary)
ggplot(data=train_sal,aes(x=train_sal$Salary, y = train_sal$capitalgain, fill = train_sal$Salary)) +
geom_boxplot() +
ggtitle("Box Plot")
ggplot(data=train_sal,aes(x=train_sal$Salary, y = train_sal$capitalloss, fill = train_sal$Salary)) +
geom_boxplot() +
ggtitle("Box Plot")
ggplot(data=train_sal,aes(x=train_sal$Salary, y = train_sal$hoursperweek, fill = train_sal$Salary)) +
geom_boxplot() +
ggtitle("Box Plot")
plot(train_sal$native,train_sal$Salary)
#Density Plot
ggplot(data=train_sal,aes(x = train_sal$age, fill = train_sal$Salary)) +
geom_density(alpha = 0.9, color = 'Violet')
ggtitle("Age - Density Plot")
## $title
## [1] "Age - Density Plot"
##
## attr(,"class")
## [1] "labels"
ggplot(data=train_sal,aes(x = train_sal$workclass, fill = train_sal$Salary)) +
geom_density(alpha = 0.9, color = 'Violet')
ggtitle("Workclass Density Plot")
## $title
## [1] "Workclass Density Plot"
##
## attr(,"class")
## [1] "labels"
ggplot(data=train_sal,aes(x = train_sal$education, fill = train_sal$Salary)) +
geom_density(alpha = 0.9, color = 'Violet')
ggtitle("education Density Plot")
## $title
## [1] "education Density Plot"
##
## attr(,"class")
## [1] "labels"
ggplot(data=train_sal,aes(x = train_sal$educationno, fill = train_sal$Salary)) +
geom_density(alpha = 0.9, color = 'Violet')
ggtitle("educationno Density Plot")
## $title
## [1] "educationno Density Plot"
##
## attr(,"class")
## [1] "labels"
ggplot(data=train_sal,aes(x = train_sal$maritalstatus, fill = train_sal$Salary)) +
geom_density(alpha = 0.9, color = 'Violet')
ggtitle("maritalstatus Density Plot")
## $title
## [1] "maritalstatus Density Plot"
##
## attr(,"class")
## [1] "labels"
ggplot(data=train_sal,aes(x = train_sal$occupation, fill = train_sal$Salary)) +
geom_density(alpha = 0.9, color = 'Violet')
## Warning: Groups with fewer than two data points have been dropped.
## Warning: Groups with fewer than two data points have been dropped.
ggtitle("occupation Density Plot")
## $title
## [1] "occupation Density Plot"
##
## attr(,"class")
## [1] "labels"
ggplot(data=train_sal,aes(x = train_sal$sex, fill = train_sal$Salary)) +
geom_density(alpha = 0.9, color = 'Violet')
ggtitle("sex Density Plot")
## $title
## [1] "sex Density Plot"
##
## attr(,"class")
## [1] "labels"
ggplot(data=train_sal,aes(x = train_sal$relationship, fill = train_sal$Salary)) +
geom_density(alpha = 0.9, color = 'Violet')
ggtitle("Relationship Density Plot")
## $title
## [1] "Relationship Density Plot"
##
## attr(,"class")
## [1] "labels"
ggplot(data=train_sal,aes(x = train_sal$race, fill = train_sal$Salary)) +
geom_density(alpha = 0.9, color = 'Violet')
ggtitle("Race Density Plot")
## $title
## [1] "Race Density Plot"
##
## attr(,"class")
## [1] "labels"
ggplot(data=train_sal,aes(x = train_sal$capitalgain, fill = train_sal$Salary)) +
geom_density(alpha = 0.9, color = 'Violet')
ggtitle("Capitalgain Density Plot")
## $title
## [1] "Capitalgain Density Plot"
##
## attr(,"class")
## [1] "labels"
ggplot(data=train_sal,aes(x = train_sal$capitalloss, fill = train_sal$Salary)) +
geom_density(alpha = 0.9, color = 'Violet')
ggtitle("Capitalloss Density Plot")
## $title
## [1] "Capitalloss Density Plot"
##
## attr(,"class")
## [1] "labels"
ggplot(data=train_sal,aes(x = train_sal$hoursperweek, fill = train_sal$Salary)) +
geom_density(alpha = 0.9, color = 'Violet')
ggtitle("Hoursperweek Density Plot")
## $title
## [1] "Hoursperweek Density Plot"
##
## attr(,"class")
## [1] "labels"
ggplot(data=train_sal,aes(x = train_sal$native, fill = train_sal$Salary)) +
geom_density(alpha = 0.9, color = 'Violet')
## Warning: Groups with fewer than two data points have been dropped.
ggtitle("native Density Plot")
## $title
## [1] "native Density Plot"
##
## attr(,"class")
## [1] "labels"
# Naive Bayes Model
Model <- naiveBayes(train_sal$Salary ~ ., data = train_sal)
Model
##
## Naive Bayes Classifier for Discrete Predictors
##
## Call:
## naiveBayes.default(x = X, y = Y, laplace = laplace)
##
## A-priori probabilities:
## Y
## <=50K >50K
## 0.7510693 0.2489307
##
## Conditional probabilities:
## age
## Y [,1] [,2]
## <=50K 36.60826 13.46489
## >50K 43.95911 10.26963
##
## workclass
## Y Federal-gov Local-gov Private Self-emp-inc
## <=50K 0.0255153843 0.0643623361 0.7685074825 0.0209243809
## >50K 0.0486148109 0.0811134790 0.6494405967 0.0799147576
## workclass
## Y Self-emp-not-inc State-gov Without-pay
## <=50K 0.0787975103 0.0412748863 0.0006180197
## >50K 0.0950985615 0.0458177944 0.0000000000
##
## education
## Y 10th 11th 12th 1st-4th 5th-6th
## <=50K 0.0335937845 0.0436586766 0.0153622037 0.0064009182 0.0121838167
## >50K 0.0078582845 0.0078582845 0.0038625466 0.0007991476 0.0015982952
## education
## Y 7th-8th 9th Assoc-acdm Assoc-voc Bachelors
## <=50K 0.0230433055 0.0189820333 0.0331964861 0.0425109257 0.1288129608
## >50K 0.0046616942 0.0033297816 0.0340969632 0.0458177944 0.2831646244
## education
## Y Doctorate HS-grad Masters Preschool Prof-school
## <=50K 0.0041937050 0.3629982784 0.0312982828 0.0019864919 0.0060036198
## >50K 0.0372935535 0.2153702717 0.1222695791 0.0000000000 0.0540756526
## education
## Y Some-college
## <=50K 0.2357745111
## >50K 0.1779435269
##
## educationno
## Y 1 2 3 4 5
## <=50K 0.0019864919 0.0064009182 0.0121838167 0.0230433055 0.0189820333
## >50K 0.0000000000 0.0007991476 0.0015982952 0.0046616942 0.0033297816
## educationno
## Y 6 7 8 9 10
## <=50K 0.0335937845 0.0436586766 0.0153622037 0.3629982784 0.2357745111
## >50K 0.0078582845 0.0078582845 0.0038625466 0.2153702717 0.1779435269
## educationno
## Y 11 12 13 14 15
## <=50K 0.0425109257 0.0331964861 0.1288129608 0.0312982828 0.0060036198
## >50K 0.0458177944 0.0340969632 0.2831646244 0.1222695791 0.0540756526
## educationno
## Y 16
## <=50K 0.0041937050
## >50K 0.0372935535
##
## maritalstatus
## Y Divorced Married-AF-spouse Married-civ-spouse
## <=50K 0.1660707191 0.0004855869 0.3384099236
## >50K 0.0602024507 0.0013319126 0.8522908897
## maritalstatus
## Y Married-spouse-absent Never-married Separated Widowed
## <=50K 0.0149649053 0.4085551583 0.0385379420 0.0329757648
## >50K 0.0041289291 0.0625998934 0.0087906233 0.0106553010
##
## occupation
## Y Adm-clerical Armed-Forces Craft-repair Exec-managerial
## <=50K 0.1422769611 0.0003531541 0.1378183905 0.0907164614
## >50K 0.0663292488 0.0001331913 0.1209376665 0.2579914758
## occupation
## Y Farming-fishing Handlers-cleaners Machine-op-inspct
## <=50K 0.0385820863 0.0559307818 0.0759281331
## >50K 0.0153169952 0.0110548748 0.0326318594
## occupation
## Y Other-service Priv-house-serv Prof-specialty Protective-serv
## <=50K 0.1359643314 0.0062684854 0.0983092747 0.0191586103
## >50K 0.0175812467 0.0001331913 0.2412093767 0.0279701652
## occupation
## Y Sales Tech-support Transport-moving
## <=50K 0.1153931047 0.0279874630 0.0553127621
## >50K 0.1291955248 0.0370271710 0.0424880128
##
## relationship
## Y Husband Not-in-family Other-relative Own-child
## <=50K 0.299474683 0.304727851 0.037655057 0.194323048
## >50K 0.756393181 0.109616409 0.004661694 0.008524241
## relationship
## Y Unmarried Wife
## <=50K 0.132388646 0.031430716
## >50K 0.028369739 0.092434736
##
## race
## Y Amer-Indian-Eskimo Asian-Pac-Islander Black Other
## <=50K 0.011124354 0.028561338 0.108197590 0.009270295
## >50K 0.004528503 0.033031433 0.048748002 0.002797017
## race
## Y White
## <=50K 0.842846422
## >50K 0.910895045
##
## sex
## Y Female Male
## <=50K 0.3826866 0.6173134
## >50K 0.1481087 0.8518913
##
## capitalgain
## Y [,1] [,2]
## <=50K 148.9004 936.4124
## >50K 3937.6798 14386.0600
##
## capitalloss
## Y [,1] [,2]
## <=50K 53.35302 309.9476
## >50K 193.75067 592.8256
##
## hoursperweek
## Y [,1] [,2]
## <=50K 39.34856 11.95104
## >50K 45.70658 10.73699
##
## native
## Y Cambodia Canada China Columbia Cuba
## <=50K 0.0004855869 0.0031342427 0.0021189246 0.0023837902 0.0029576657
## >50K 0.0009323388 0.0047948855 0.0026638253 0.0002663825 0.0033297816
## native
## Y Dominican-Republic Ecuador El-Salvador England
## <=50K 0.0028693771 0.0010153181 0.0040171280 0.0024720788
## >50K 0.0002663825 0.0005327651 0.0011987214 0.0039957379
## native
## Y France Germany Greece Guatemala Haiti
## <=50K 0.0006621640 0.0037081181 0.0009270295 0.0026486558 0.0016774820
## >50K 0.0015982952 0.0058604156 0.0010655301 0.0003995738 0.0005327651
## native
## Y Honduras Hong Hungary India Iran
## <=50K 0.0004855869 0.0005738754 0.0004414426 0.0026486558 0.0010594623
## >50K 0.0001331913 0.0007991476 0.0003995738 0.0053276505 0.0023974427
## native
## Y Ireland Italy Jamaica Japan Laos
## <=50K 0.0008387410 0.0019423476 0.0030900984 0.0015891935 0.0006621640
## >50K 0.0006659563 0.0031965903 0.0013319126 0.0030633990 0.0002663825
## native
## Y Mexico Nicaragua Outlying-US(Guam-USVI-etc)
## <=50K 0.0254712400 0.0013684722 0.0006180197
## >50K 0.0043953117 0.0002663825 0.0000000000
## native
## Y Peru Philippines Poland Portugal Puerto-Rico
## <=50K 0.0012360394 0.0056504657 0.0019864919 0.0013243279 0.0042819936
## >50K 0.0002663825 0.0079914758 0.0014651039 0.0005327651 0.0015982952
## native
## Y Scotland South Taiwan Thailand
## <=50K 0.0003972984 0.0025162230 0.0010153181 0.0006180197
## >50K 0.0002663825 0.0018646777 0.0025306340 0.0003995738
## native
## Y Trinadad&Tobago United-States Vietnam Yugoslavia
## <=50K 0.0007063082 0.9053546992 0.0026045115 0.0004414426
## >50K 0.0002663825 0.9316728823 0.0006659563 0.0007991476
Model_pred <- predict(Model,test_sal)
mean(Model_pred==test_sal$Salary)
## [1] 0.8187251
confusionMatrix(Model_pred,test_sal$Salary)
## Confusion Matrix and Statistics
##
## Reference
## Prediction <=50K >50K
## <=50K 10549 1919
## >50K 811 1781
##
## Accuracy : 0.8187
## 95% CI : (0.8125, 0.8248)
## No Information Rate : 0.7543
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.456
## Mcnemar's Test P-Value : < 2.2e-16
##
## Sensitivity : 0.9286
## Specificity : 0.4814
## Pos Pred Value : 0.8461
## Neg Pred Value : 0.6871
## Prevalence : 0.7543
## Detection Rate : 0.7005
## Detection Prevalence : 0.8279
## Balanced Accuracy : 0.7050
##
## 'Positive' Class : <=50K
##