A movie recommendation engine demonstration website built with Python and ReactJs

Built as a project for Microsoft Engage 2022 (Problem Statement 3 - Algorithm)

What is it?

Movie-Recommender is a website to demonstrate how recommendation engine works in real world applications. Here I have applied only content based filtering for the recommendation engine. Also I have scraped the movie dataset from imdb. This recommendation engine accuracy will improve more if we put more relevant data in dataset.

Main Features

• When visiting a movie details, the similar movies are recommended on the basis of title, genre, director, casts similarities
• The search function is also implemented to find any movies using search algorithms (only you can search within the 1000 movies which are present in the dataset)
• In homepage, few category movies are recommended

Technologies Used

• Python and Flask (for backend)
• ReactJs (for frontend)

Dependencies

• Backend

  Backend Dependencies are provided in /requirements.txt

  • flask
  • flask_cors
  • pandas
  • numpy
  • sklearn
• Frontend

  Frontend Dependencies are provided in /frontend/package.json

Setup Guide To Run In Localhost

• Firstly clone the github repo using git clone https://github.com/SayanDeveloper/movie-recommender.git
• Then for backend setup, open terminal in the root directory of this repo,
• Type in terminal pip3 install -r requirements.txt (You should have python3 and pip installed in your machine)
• Then python app.py
• Now your backend server is running
• For frontend, open another terminal and go to /frontend directory
• Type npm install to install all dependencies of frontend (npm and node is required in your machine to run this)
• Then npm start to start react server
• I have pushed the code for production server so using my production backend server domain for api fetching
• To use your localhost as api endpoint, replace url constant in useEffect of the files DetailsPage.js, HomePage.js, SearchResult.js

Happy Coding :-)