Data pre-processing

# Sentiment Analysis on Text Data
# This script processes a dataset of text data and calculates sentiment scores based on positive and negative word dictionaries.
# Positive and negative word dictionaries are used to determine sentiment (positive, negative, or neutral).

# Step 1: Load necessary libraries
library(stringr)  # For string manipulation
library(tm)       # For text mining and text preprocessing
library(plyr)     # For applying functions over lists
library(stringr)  # For string operations

# Step 2: Read in the dataset
# Replace the file path with the correct path to your dataset
Dataset <- read.csv('C:/Users/shreyash/Desktop/New folder/scoresportscars.csv')

# Convert the text column into a factor (categorical data) for easier processing
Dataset$text <- as.factor(Dataset$text)

# Step 3: Load positive and negative word dictionaries
# Replace the file paths with the actual locations of your dictionaries
pos.words <- scan('C:/Users/BHARAT/Desktop/DataSet/Dataset 2017/pos_words.txt', what='character', comment.char=';') # Positive word list
neg.words <- scan('C:/Users/BHARAT/Desktop/DataSet/Dataset 2017/neg_words.txt', what='character', comment.char=';') # Negative word list

# Optional: Adding custom words to the dictionaries
# pos.words <- c(pos.words, 'upgrade')   # Add your custom positive words
# neg.words <- c(neg.words, 'wtf', 'wait', 'waiting', 'epicfail')  # Add custom negative words

# Step 4: Function to calculate sentiment scores
# This function processes sentences, calculates sentiment score based on positive and negative word dictionaries

score.sentiment <- function(sentences, pos.words, neg.words, .progress='none') {
  
  # Create a list to store scores
  scores <- laply(sentences, function(sentence, pos.words, neg.words) {
    
    # Text cleaning: Removing retweets, URLs, punctuation, control characters, and digits
    sentence = gsub("(RT|via)((?:\\b\\W*@\\w+)+)", "", sentence)  # Remove RTs and mentions
    sentence <- gsub('[[:punct:]]', "", sentence)  # Remove punctuation
    sentence <- gsub('[[:cntrl:]]', "", sentence)  # Remove control characters
    sentence <- gsub('\d+', "", sentence)  # Remove digits
    
    # Convert text to lowercase
    sentence <- tolower(sentence)
    
    # Split the sentence into words
    word.list <- str_split(sentence, '\\s+')  # Split by spaces
    words <- unlist(word.list)  # Flatten the list
    
    # Match words to positive and negative dictionaries
    pos.matches <- match(words, pos.words)  # Find positive words
    neg.matches <- match(words, neg.words)  # Find negative words
    
    # Convert matches to boolean values (TRUE = match found, FALSE = no match)
    pos.matches <- !is.na(pos.matches)  # TRUE for positive word match
    neg.matches <- !is.na(neg.matches)  # TRUE for negative word match
    
    # Calculate score: positive score - negative score
    score <- sum(pos.matches) - sum(neg.matches)
    
    # Return the score
    return(score)
  }, pos.words, neg.words, .progress = .progress)
  
  # Create a data frame with scores and corresponding text
  scores.df <- data.frame(score = scores, text = sentences)
  
  # Return the data frame
  return(scores.df)
}

# Step 5: Apply sentiment analysis to the dataset
# Apply the sentiment function on the 'text' column of the dataset
scores <- score.sentiment(Dataset$text, pos.words, neg.words, .progress='text')

# Step 6: Save the sentiment scores to a new CSV file
# The sentiment scores are saved to a new CSV file
write.csv(scores, file = 'C:/Users/shreyash/Desktop/New folder/scoresportscars.csv', row.names = TRUE)


sentiment-analysis-real-time/
│
├── scripts/
│   └── sentiment_analysis.R         # Main R script for sentiment analysis
│
├── data/
│   ├── scoresportscars.csv          # Input data for sentiment analysis
│   └── pos_words.txt               # Positive word dictionary
│   └── neg_words.txt               # Negative word dictionary
│
├── README.md                       # Project documentation (this file)