SentimentSense is a YouTube sentiment analysis tool that fetches comments from a given video URL and classifies them as positive, negative, or neutral. It provides a clear visual representation of audience reactions using bar and pie charts. Additionally, users can analyze the toxicity of specific comments, offering insights into both general sentiment and individual feedback.
-
YouTube Comment Sentiment Analysis: Fetch comments from any YouTube video and classify them into positive, negative, or neutral categories.
-
Sentiment Visualization: Display the sentiment distribution using bar charts and pie charts for a clear overview of audience reactions.
-
Toxicity Analysis: Enter a specific comment to analyze its toxicity parameters such as Toxic, Severe Toxic, Obscene, Threat, Insult, and Identity Hate.
-
Versatile Insights: Understand overall audience sentiment while also providing detailed analysis for individual user comments.
This project demonstrates the potential of AI-driven emotion detection, allowing for more intuitive interaction between humans and machines.
- Python
- YouTube Data API
- FastAPI
- TensorFlow
Recording.2024-10-01.120337.mp4
- Clone the repository:
git clone https://github.com/Chauhan-Aman/SentimentSense---YouTube-Sentiment-Analyzer.git
cd SentimentSense
- Set up a virtual environment:
python -m venv venv
source venv/bin/activate.ps1 # On Windows use `venv\Scripts\activate.ps1`
- Install required dependencies:
Install required dependencies:
- Run the FastAPI server:
uvicorn main:app --reload
- Access the API: Open your browser and go to http://127.0.0.1:8000 to see the Human Emotion Detection API running.
- Frontend Interaction:
- To interact with the API from the frontend, ensure CORS settings allow access from your frontend domain.
- You can upload images through the frontend to classify emotions.
This setup will allow you to run the Human Emotion Detection system locally.