FindMovix

Project Overview

FindMovix is an AI-powered movie recommendation application that suggests films based on user preferences and viewing history. The platform utilizes machine learning algorithms to analyze user behavior and provide highly personalized content recommendations.

This project was developed as part of my exploration into practical applications of artificial intelligence in consumer-facing products. The goal was to create a system that could deliver increasingly accurate recommendations as users interact with the platform.

Project Details

Client: Personal Project
Timeline: November 2024 - February 2025
Team Size: 1 members
My Role: Lead Developer & AI Integration & Backend Develooper & Db Develooper

Technologies Used

React

ASP.NET Core

SQL Server

Open Ai

TMDB API

Git/GitHub

The Problem

Traditional movie recommendation systems often suffer from several limitations:

Generic recommendations based on broad categories
Limited understanding of user preferences beyond explicit ratings
Inability to adapt to changing user tastes over time
Poor discovery of niche content that users might enjoy

Users often find themselves scrolling through endless options without finding content that truly matches their preferences, leading to decision fatigue and a poor user experience.

The Solution

FindMovix addresses these issues through a multi-faceted approach:

Hybrid recommendation system combining collaborative filtering and content-based analysis
Implicit preference tracking through viewing behavior and engagement metrics
Continuous learning algorithm that adapts to evolving user preferences
Contextual recommendations based on time, mood, and social factors

By leveraging machine learning and AI, FindMovix creates a personalized movie discovery experience that improves over time, helping users find content they'll truly enjoy.

Key Features

AI-Powered Recommendations

Machine learning algorithms analyze user behavior, preferences, and viewing history to provide increasingly accurate movie suggestions.

Personalized User Profiles

Detailed user profiles that capture explicit preferences and implicit behavior patterns to refine recommendation accuracy.

Comprehensive Movie Database

Integration with TMDB API provides access to a vast library of movies with detailed metadata, reviews, and ratings.

Mood-Based Suggestions

Users can receive recommendations based on their current mood, offering contextually relevant content for different situations.

Social Recommendations

Friend network integration allows users to see what their connections are watching and enjoying.

Viewing Analytics

Insights into personal viewing habits, genres, and trends to help users understand their preferences.

Implementation Details

Technical Architecture

FindMovix follows a modern client-server architecture with clear separation of concerns:

Frontend

React Native for UI components
Redux for state management
Axios for API requests
Styled Components for styling

Backend

ASP.NET Core Web API
Entity Framework Core
Open Ai for Ai learning
JWT authentication

Infrastructure

SQL Server database
CI/CD with GitHub Actions

Database Design

The database schema was designed to efficiently store user data, movie information, and the relationships between them:

Entity	Description	Key Attributes
Users	User account information	ID, Username, Email, Password Hash, Profile Data
Movies	Movie metadata from TMDB	ID, Title, Overview, Release Date, Genres, Cast
UserPreferences	Explicit user preferences	UserID, Preferred Genres, Directors, Actors
Ratings	User ratings for movies	UserID, MovieID, Rating, Timestamp
ViewingHistory	Record of user movie views	UserID, MovieID, ViewDate, CompletionPercentage

Project Links

Live Demo

Experience the application in action

GitHub Repository

View the source code and documentation

Mobil App

Designed for your movie discovery experience on the App Store

Previous Project: Lawyer Portfolio Back to All Projects Next Project: SonKoç

FindMovix Project