/Movie-Recommender-System

This ML model recommends movies that may align with the user's preferences based on TF-IDF matrix.

Primary LanguageJupyter Notebook

Movie-Recommender-System

Overview

  • When a user enters a movie title into the input box, this recommendation system swiftly generates suggestions for other movies that may align with the user's preferences.
  • This repository contains all the codes and resources used to build and utilize the recommendation system.

Key Features

  • Data Exploration: Comprehensive analysis to understand the MovieLens 25M dataset's structure and distribution.
  • Search Engine: Building a search engine to find a specific movie title in the dataset.
  • Recommendation Engine: Creating a recommendation engine to suggest specific movies based on user preferences and movie ratings.

Steps

  • Reading in movie data with pandas.
  • Cleaning movie titles with regex.
  • Creating a TF-IDF matrix. (Time Frequency-Inverse Document Frequency)
  • Creating a search function.
  • Building an interactive search box with Jupyter.
  • Reading in movie ratings data.
  • Finding users who liked the same movie.
  • Finding how much all users like movies.
  • Creating a recommendation score.
  • Building a recommendation function.
  • Creating an interactive recommendation widget.

Code

You can find the code for this project here:

Technologies/Tools

  • Jupyter Notebook / Google Colab
  • Python 3.10.12
  • Python packages
    • Pandas - pip install pandas
    • Numpy - pip install numpy
    • Scikit-learn - pip install scikit-learn
    • Regex - pip install regex

Python Jupyter Notebook Pandas NumPy scikit-learn regex

Data

You can download the dataset files used in this project here: