decision_tree_Naive_bayes.Rmd

---
title: "new_decison_tree"
author: "Karan"
date: '2022-10-27'
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```

#Load Libraries

```{r}
library(tidyverse)
library(text2vec)
library(rpart)
library(plyr)
library(caret)
library(MLmetrics)
library(naivebayes)
```

## R Markdown

This is an R Markdown document. Markdown is a simple formatting syntax for authoring HTML, PDF, and MS Word documents. For more details on using R Markdown see <http://rmarkdown.rstudio.com>.

When you click the **Knit** button a document will be generated that includes both content as well as the output of any embedded R code chunks within the document. You can embed an R code chunk like this:
#load cleaned_data
```{r cars}
df <- read.csv('/Users/crazyk/Cleaned_data.csv')
df
```

## Including Plots

You can also embed plots, for example:

```{r pressure, echo=FALSE}
names(df)
```

#rename columns
```{r pressure, echo=FALSE}
colnames(df)[which(names(df) == "Hit.Sentence")] <- "Sentence"
```

```{r}
count(df$target)
```
#Map target to numeric values
```{r}
df$target <- mapvalues(df$target,from = c("pro","anti"),to = c(1,0))
```

#Take only a subset of data and create a dataframe
```{r}
df_anti <- filter(df,target==0) %>% slice(1:2000)
df_pro <- filter(df,target==1) %>% slice(1:2000)

new_df <- rbind(df_anti,df_pro)
new_df <- new_df[,c("Sentence","target")]
```

#Split into test and train
```{r}
train.index <- createDataPartition(new_df$target, p = .8, list = FALSE)
train <- new_df[ train.index,]
test  <- new_df[-train.index,]
```

```{r}
dim(train)
```

```{r}
#train tfidf vectorization
N = 3200
it = itoken(train$Sentence[1:N], preprocess_function = tolower,
tokenizer = word_tokenizer)
v = create_vocabulary(it)
#remove very common and uncommon words
pruned_vocab = prune_vocabulary(v, term_count_min = 10,
doc_proportion_max = 0.5, doc_proportion_min = 0.001)
vectorizer = vocab_vectorizer(v)
it = itoken(train$Sentence[1:N], preprocess_function = tolower,
tokenizer = word_tokenizer)
dtm = create_dtm(it, vectorizer)
# get tf-idf matrix from bag-of-words matrix
model_tfidf= TfIdf$new()
dtm_tfidf = model_tfidf$fit_transform(dtm)
df_train_idf <- as.data.frame(as.matrix(dtm_tfidf))
df_train_idf$target_val <- train$target 
```



```{r}
#transform test 
N = 800
it = itoken(test$Sentence[1:N], preprocess_function = tolower,
tokenizer = word_tokenizer)
dtm = create_dtm(it, vectorizer)
# get tf-idf matrix from bag-of-words matrix
dtm_tfidf = model_tfidf$transform(dtm)
df_test_idf <- as.data.frame(as.matrix(dtm_tfidf))
df_test_idf$target_val <- test$target 
```

```{r}
#Decision tree model
DT <- rpart(df_train_idf$target_val ~ ., data = df_train_idf, method="class")
summary(DT)
```

```{r}
#Decision tree prediction
DT_Prediction <- predict(DT, df_test_idf, type="class")
```


```{r}
#confusion matrix
table(DT_Prediction,df_test_idf$target_val)
```
```{r}
#score
F1_Score(y_pred = DT_Prediction, y_true = df_test_idf$target_val)
```
```{r}
#naive bayes model
model <- naive_bayes(df_train_idf$target_val ~ ., data = df_train_idf,usekernel = F) 
```


```{r}
#predict
p1 <- predict(model, df_test_idf)
```
```{r}
table(p1,df_test_idf$target_val)
```
```{r}
F1_Score(y_pred = p1,y_true = df_test_idf$target_val)
```
```{r}
df_test_idf
```