RecommendOn - Movie Recommendation System for HackOn

Features

• Content discovery and personalized Content Advisor leverages ML model powered by Natural Language Processing Model/Large Language Models (LLM), to offer finely tuned content recommendations based on user preferences and location.
• Voice-activated commands, family-friendly filters, curated playlist

Technology

• React - for frontend
• Node/ExpressJS - for backend
• Fast API to interact with ML model
• Google's Universal Sentence Encoder to generate Vector Embeddings -> tfhub.dev/google/universal-sentence-encoder/4
• Rapid API - to interact with CHAT GPT
• react-speech-recognition - to convert speech from the microphone to text and make it available to your React components
• react-speech - to convert output from Chat-GPT to audio

Running Locally

Frontend

npm install
npm start

Backend

npm install
npm start

Fast API Model

• To run the Fast API model, follow these steps:

• Install the required Python libraries.

• Run the Fast API application using uvicorn:

uvicorn model_microservice:app --host 0.0.0.0 --port 8001 --reload

This will start the Fast API model microservice, allowing it to interact with your application