Hack-Nocturne-2025 - AI Devrel System

"Empowering Companies with AI-Driven Social Media Intelligence for DevRel"

• Companies struggle to understand public sentiment and trending topics across social media platforms, impacting decision-making and brand management.
• We built multiple models per company like DistilBERT, BERTopic, BERT-based Clustering, GPT-4, NER Models to classify social media posts. The data was scraped, cleaned, and fine-tuned for multi-label categorization, ensuring efficient and accurate classification.
• The solution automates social media analysis, providing real-time insights to help companies prioritize features, address complaints, and track competition.
• Unlike generic sentiment analysis, our company-specific models ensure precise categorization, offering deeper insights and real-time tracking for better decision-making.

Frontend

This folder contains the frontend code for the Hack-Nocturne-2025 project. It is built using modern web development tools and frameworks.

Folder Structure

• index.html: The main HTML file for the application.
• ProtectedRoute.jsx: A React component for handling protected routes.
• vite.config.js: Configuration file for Vite.
• src/: Contains the source code for the frontend application.

Prerequisites

• Node.js (v16 or higher)
• npm or yarn

Tech Stack

• React
• Vite
• Tailwind CSS

Setup

1. Navigate to the frontend folder:

   cd frontend

2. Install the dependencies:

   npm install

3. Start the application:

   npm run dev

Backend

This folder contains the backend code for the Hack-Nocturne-2025 project. It is built using Node.js and Express.

Folder Structure

• app.js: The main application file for the backend.
• routes/: Contains route definitions for the application.
• controller/: Contains the logic for handling requests.
• model/: Contains the data models for the application.
• middleware/: Contains middleware functions for request handling.
• db/: Contains database connection and configuration files.

Prerequisites

• Node.js (v16 or higher)
• npm

Tech Stack

• Node.js
• Express
• MongoDB
• Mongoose
• JWT (JSON Web Tokens)
• Cors (for cross-origin resource sharing)

Setup

1. Navigate to the backend folder:

   cd backend

2. Install the dependencies:

   npm install

3. Start the application:

   node index.js

   This will start the backend server, and you can access the API at http://localhost:5000.

Machine Learning

This folder contains the machine learning code and models for Hack-Nocturne-2025. It leverages advanced NLP and ML techniques to provide insightful analytics and predictions.

ML Model: DistilBERT, BERTopic, BERT-based Clustering, GPT-4, NER Models

ML Tech Stack: PyTorch, Matplotlib, Seaborn, NumPy, Scikit-Learn, spaCy, HuggingFace, Flask, NLTK

Contributing

We welcome contributions to the Hack-Nocturne-2025 project! Please follow these steps to contribute:

1. Fork the repository.
2. Create a new branch for your feature or bug fix.
3. Make your changes and commit them.
4. Push your branch to your forked repository.
5. Create a pull request to the main repository.

Contributors

• Inchara J
• Zrahay
• Harsh K G
• Himanshu

License

This project is licensed under the MIT License. See the LICENSE file for more details.