← AI 🔗

If You Liked This, You'll Like That

A real similarity-scoring recommendation engine for movies, built entirely from a hardcoded tag dataset.

🔴 HardDifficulty

4-5 hoursTime

Python, JavaScriptLanguages

What you're building

You're building a genuine content-based recommendation engine, the same family of algorithm Netflix and Spotify use as one ingredient in their systems. Every movie in your dataset gets tagged with genres, tones, and themes; when a user picks a movie they liked, the app computes tag-overlap similarity (a real algorithm called the Jaccard index) against every other movie and ranks the best matches. No black box -- you can see and explain exactly why each recommendation was picked.

What you'll unlock 🔓

Similarity scoringJaccard indexMatrix-style data modeling

You'll need

Comfortable with JavaScript
Arrays, Sets, and objects
A browser

🗺️ Build roadmap

01 Build the hardcoded movie dataset with tags → 02 Build the movie-picker UI → 03 Implement the Jaccard similarity function → 04 Rank all other movies by similarity to the pick → 05 Show the top 3 recommendations with shared-tag explanations → 06 Add a similarity percentage badge → 07 Add a 'build your taste profile' mode using multiple picks

🔒

Log in to start building

Your first project is free -- create an account to unlock it.