Movie Recommendation

Utilizes 700,000+ movies to create a recommendation

Comment 2

💡 Inspiration 💡

With the theme of cinema in mind, we started brainstorming ideas. We had lots of ideas for games, but we eventually decided on a movie recommendation system. All of us agreed that finding new, fun movies has been difficult lately, and this program remedies that.

⚙️ What it does ⚙️

The main feature of our app is to recommend movies based on movies that the user enjoyed. The user clicks on 3 or more movies that they enjoyed from a selection of 20 or so movies, and using those 3 movies, it will parse a list of 722,000 and recommend 5 movies that are similar or recommended for people who enjoyed the 3 movies they selected. It will also display the poster for the movie, the plot,

🛠️ How we built it 🛠️

The system was built using Flask, a popular web framework for Python. Flask provided the functionality to build the web pages and API endpoints to serve recommendations. Pandas, was used to load, manipulate and analyze the movie data that powered the recommendations. Pandas enabled loading the movie dataset, filtering and preparing the data before feeding into the recommendation algorithms. Finally, the front-end was built using HTML, CSS and JavaScript. The styling was done using simple.css, a minimal CSS framework. This allowed creating a clean and responsive web interface to display movie recommendations to users.

😣 Challenges 😣

The movie dataset required significant data cleaning and preprocessing before it could be used for generating recommendations. Additionally, building an intuitive UI that allowed users to seamlessly interact with recommendations was challenging. We had to refine the UI/UX through multiple iterations.

📚 What we learned 📚

During this Hackathon, we learned how to build a sleek looking front-end within simple.css and how to sift through massive amounts of data with pandas. We also learned how to collaborate well through Vscode Liveshare and Discord.

Share this project:

Updates