Data-Set-Analysis-Using-NaiveBayesClassifier/afterPreprocessing.R at main · HChandeepa/Data-Set-Analysis-Using-NaiveBayesClassifier · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
library(e1071)
library(caret)
library(caTools)

# Read the dataset
ds <- read.csv("F:\\Year 2 Sem 1\\R programming\\Datasets\\student_portuguese_clean.csv")
ds


#convert strings to  numeric-encoding
str(ds$school)
ds$school<-as.numeric(factor(ds$school))
ds$school
str(ds$sex)

ds$sex<-as.numeric(factor(ds$sex))
ds$sex


ds$address_type <- factor(ds$address_type, levels = c("Urban", "Suburban", "Rural"))


ds$address_type <- as.numeric(ds$address_type)

str(ds$address_type)
ds$address_type<-as.numeric(ds$address_type)
ds$address_type
str(ds$address_type)
str(ds$family_size)
ds$family_size<-factor(ds$family_size,levels = c("Greater than 3","Less than or equal to 3"))
ds$family_size<-as.numeric(ds$family_size)
ds$family_size
str(ds$parent_status)
ds$parent_status<-factor(ds$parent_status,levels = c("Apart","Living together"))
ds$parent_status<-as.numeric(ds$parent_status)
ds$parent_status
str(ds$mother_education)

ds$mother_education <- factor(ds$mother_education, levels = c("higher education", "primary education (4th grade)", "secondary education", "5th to 9th grade"))
ds$mother_education <- as.numeric(ds$mother_education)
ds$mother_education

ds$father_education<-factor(ds$father_education,levels = c("higher education", "primary education (4th grade)", "secondary education", "5th to 9th grade","none"))
ds$father_education<-as.numeric(ds$father_education)
ds$father_education

unique(ds$school_choice_reason)
ds$school_choice_reason<-factor(ds$school_choice_reason,levels = c("course","other","home","reputation"))
ds$school_choice_reason<-as.numeric(ds$school_choice_reason)

unique(ds$study_time)
ds$study_time<-factor(ds$study_time,levels = c("2 to 5 hours","5 to 10 hours","<2 hours",">10 hours"))
ds$study_time<-as.numeric(ds$study_time)
ds$study_time


ds$school_support<-as.numeric(ds$school_support=="yes")
ds$school_support

unique(ds$extra_paid_classes)
ds$extra_paid_classes<-as.numeric(ds$extra_paid_classes=="yes")
ds$extra_paid_classes


ds$activities<-as.numeric(ds$activities=="yes")
ds$activities

ds$nursery_school<-as.numeric(ds$nursery_school=="yes")
ds$nursery_school

ds$higher_ed<-as.numeric(ds$higher_ed=="yes")
ds$higher_ed

ds$internet_access<-as.numeric(ds$internet_access=="yes")
ds$internet_access


unique(ds$guardian)
ds$guardian<-factor(ds$guardian,levels = c("mother", "father", "other" ))
ds$guardian<-as.numeric(ds$guardian)
ds$guardian

unique(ds$travel_time)
ds$travel_time<-factor(ds$travel_time,levels = c("15 to 30 min.","<15 min.","30 min. to 1 hour",">1 hour"))
ds$travel_time<-as.numeric(ds$travel_time)
ds$travel_time


#handling missing values
#sum(is.na(ds))

missing_cols <- colnames(ds)[colSums(is.na(ds)) >0]
missing_cols

# Impute missing values with column means
for (col in missing_cols) {
  ds[[col]][is.na(ds[[col]])] <- mean(ds[[col]], na.rm = TRUE)
}

#remove rows that with missing values
#na.omit(ds)

# Perform data splitting
split_ratio <- sample.split(ds, SplitRatio = 0.75)
training_dataset <- subset(ds, split_ratio == TRUE)
testing_dataset <- subset(ds, split_ratio == FALSE)


# Ensure 'final_grade' is a factor in both datasets with the same levels
training_dataset$final_grade <- as.factor(training_dataset$final_grade)
training_dataset$
testing_dataset$final_grade <- as.factor(testing_dataset$final_grade)

# Create training and testing data frames with the selected columns
training_ds <- training_dataset[, 1:34]
testing_ds <- testing_dataset[, 1:34]

# Train the Naive Bayes model
set.seed(400)
#model <- naiveBayes(final_grade ~ ., data = training_dataset)
# Train the Naive Bayes model using the correct data frame
# Train the Naive Bayes model using the correct data frame
model <- naiveBayes(final_grade ~ ., data = training_ds)

# Make predictions on the testing dataset
predicted_results <- predict(model, newdata = testing_ds)

# Convert predicted_results to a factor with the same levels as training_ds$final_grade
predicted_results <- factor(predicted_results, levels = levels(training_ds$final_grade))

# Convert testing_dataset$final_grade to a factor with the same levels as predicted_results
testing_dataset$final_grade <- factor(testing_dataset$final_grade, levels = levels(predicted_results))

# Create confusion matrix
confusionMatrix(predicted_results, testing_dataset$final_grade)