Ragopedia

AI-Powered Movie Discovery Platform based on Retrieval-Augmented Generation (RAG)

Delivering recommendations and letting users create, organize, and explore movie lists with ease through intelligent conversational AI.

View Demo View Poster Learn More

Project Overview

Ragopedia leverages AI technology to revolutionize how users discover and interact with movies

What is RAG?

Retrieval-Augmented Generation (RAG) combines the power of large language models with real-time information retrieval, enabling our system to provide accurate, contextual, and up-to-date movie recommendations based on a comprehensive database.

Intelligent Recommendations

AI-powered suggestions based on user preferences and movie database

Natural Language Interface

Conversational AI that understands complex movie queries

Key Statistics

44K+
Movies in Database
RAG
AI Architecture
React
Frontend Framework
Flask
Backend Framework
SQL
Database Query
Llama3.2:1b
LLM Model

Core Features

Discover what makes Ragopedia the ultimate movie discovery platform

Smart Suggestions

Ask natural questions like "Suggest me mind-bending movies like Inception" and get AI-powered recommendations.

AI-Powered • Natural Language

Advanced Movie Search

Explore detailed information on thousands of popular films with smart semantic search for relevant results.

Vector Search • Metadata Rich

Personal Movie Lists

Create and manage your own movie collections. Easily save movies, organize them by genre or mood, and revisit anytime.

User Management • Organization

Intuitive Interface

Modern, responsive React-based interface with smooth animations and user-friendly design for seamless movie discovery.

React • Responsive Design

System Architecture

How Ragopedia processes your movie queries through our RAG pipeline

1

Natural Language Query

User types movie-related question or request via React interface

2

Text-to-SQL Translation

LLaMA converts natural language input into structured SQL query

3

Database Retrieval

SQL query executed on MySQL database with semantic search via FAISS

4

AI Response Generation

LLaMA generates contextual response with movie recommendations

5

Description Embedding

Movie descriptions embedded with SentenceTransformers

6

Semantic Similarity Search

FAISS finds movies with similar meaning based on cosine similarity

7

Context Construction

Top relevant movies are formatted into input for LLaMA

8

Frontend Display

Answer shown in React with optional movie cards and actions

Technology Stack

JavaScript

JavaScript

React

React

LLaMA

LLaMA

Python

Python

Python

Flask

MySQL

MySQL

Demo

Project Team

Meet the team behind Ragopedia

Prof. Dr. İlyas Çiçekli

Supervisor

Project guidance and academic supervision

Uygar Ersoy

Developer

AI integration, RAG implementation

Mete Uysal

Developer

Full-stack development, database design

Course Information

Course: BBM480 - Software Engineering
University: Hacettepe University
Semester: Spring 2025
Department: Computer Engineering