Sentiment Analysis 🗿

Sentiment Analysis for YouTube, Reddit, and Reviews is a machine learning-based project that analyzes and classifies the sentiment of comments from YouTube videos, Reddit posts, and Product/Service reviews. The sentiments are categorized as Positive, Negative, and Neutral. This project allows users to input a YouTube video URL to analyze its comments, specify a subreddit to analyze posts and comments, and even paste product or service reviews for sentiment classification.

---

Introduction

The Sentiment Analysis for YouTube, Reddit, and Reviews project leverages machine learning algorithms to classify the sentiment of comments from YouTube videos, posts from Reddit, and general product/service reviews. The sentiment is classified into Positive, Negative, or Neutral categories.

The project is designed to provide users with a simple way to analyze opinions and sentiments from online platforms:

• YouTube: Analyze comments from any YouTube video by pasting the URL.
• Reddit: Analyze posts and comments from a specified subreddit.
• Review Analysis: Paste product or service reviews and obtain sentiment classification for each review.

The sentiment analysis uses Natural Language Processing (NLP) techniques, leveraging models such as VADER and scikit-learn to efficiently classify text data.

---

Technologies Used

• Python: Programming language for development.
• pandas: Data manipulation and analysis library.
• nltk: Toolkit for natural language processing.
• VADER: Sentiment analysis tool for social media text.
• scikit-learn: Machine learning library used for model building.
• Google API Client: For fetching YouTube comments.
• PRAW: Python Reddit API Wrapper for interacting with Reddit.
• Flask (Optional): Web interface for easy interaction.
• Matplotlib: Data visualization for sentiment distribution.

---

Features

• YouTube Comment Sentiment Analysis: Paste any YouTube video URL to analyze the sentiment of its comments.
• Reddit Post Sentiment Analysis: Enter a subreddit name to fetch and analyze posts along with sentiment classification of the comments.
• Review Sentiment Analysis: Paste product or service reviews and classify the sentiment as positive, negative, or neutral.
• Positive, Negative, Neutral Classification: Each comment, post, or review is classified into one of three sentiment categories.
• Searchable Subreddit Posts: Specify the number of posts to analyze from a given subreddit.
• Visualization: Sentiment distribution is displayed as a graph for better insight.
• Real-Time Analysis: Perform sentiment analysis in real-time for YouTube comments, Reddit posts, and reviews.

---

Installation

1. Clone the repository:

   git clone https://github.com/iamamrizejk/Sentiment-Analysis.git

2. Navigate to the project directory:

   cd Sentiment-Analysis

3. Create a virtual environment (optional but recommended):

   python -m venv myenv

4. Activate the virtual environment:

   • On Windows:

     myenv\Scripts\activate

   • On macOS/Linux:

     source myenv/bin/activate

---

How It Works

1. YouTube:

   • Fetches comments from the YouTube video URL provided by the user using the Google API Client.
   • Preprocesses the text and applies VADER and scikit-learn models for sentiment classification.
   • Classifies each comment as positive, negative, or neutral.

2. Reddit:

   • Fetches posts and comments from the specified subreddit using PRAW.
   • Preprocesses the text and performs sentiment analysis on the posts and comments.
   • Classifies the sentiment of each post's comments as positive, negative, or neutral.

3. Review Analysis:

   • Takes a list of product or service reviews as input.
   • Preprocesses the review text and applies VADER and scikit-learn models for sentiment classification.
   • Classifies each review as positive, negative, or neutral.

4. Visualization:

   • Displays sentiment analysis results using Matplotlib to plot graphs that show the distribution of sentiment across the comments, posts, or reviews.

---

Acknowledgments

• Google API Client: For providing access to YouTube comments.
• PRAW: For enabling interaction with Reddit data.
• VADER: For efficient sentiment analysis of social media and review content.
• Libraries: Thanks to scikit-learn, nltk, pandas, and matplotlib for providing robust tools for machine learning and data analysis.

---