🎯 SentiSense – Comment Sentiment Analyzer & Insights Engine

SentiSense is a full-stack sentiment analysis web application that scrapes comments from platforms like YouTube and Reddit, analyzes them using a custom-trained BiLSTM model, and offers meaningful insights, including visual sentiment trends, clustering, and AI-powered suggestions via Gemini API.

---

🚀 Features

• 🔎 Multi-platform scraping (Reddit, YouTube)
• 💡 Custom sentiment analysis model (.h5 + tokenizer .pkl)
• 📊 Sentiment trend graphs and clustering
• 🤖 AI-generated suggestions (via Gemini API)
• 📁 CSV export of analyzed results
• 💻 Built with FastAPI + TensorFlow + HuggingFace + Pandas

---

🧠 Tech Stack

Layer	Tech Used
Frontend	React.js, axios
Backend	FastAPI, Uvicorn
Modeling	TensorFlow (Keras), Tokenizer, HuggingFace Transformers
Scraping	yt-dlp (Youtube) and praw (Reddit)
AI Assist	Gemini API (LLM)
Infra	OnRender / Localhost

---

⚙️ Installation & Usage

1) Clone the repo:

   git clone https://github.com/Kake27/sentisense
   cd sentisense

2) Create the virtual environment:

   python -m venv venv
   venv\Scripts\activate

3) Install dependencies:

   cd backend
   python -m venv venv
   venv\Scripts\activate
   pip install -r requirements.txt

4) Add .env variables:

   Get your client id, client secret by registering an app on Reddit and add them to the file
   Also add your GenAI API key 

5) Run the backend server:

   uvicorn runserver:app --reload

6) Install frontend dependencies:

   cd ../frontend
   npm install

7) Run frontend server:

   npm run dev