Netflix Movie Recommenderv1.0
✦ Project Overview
A content-based recommendation system that suggests movies based on similarity. It analyzes metadata such as genres, cast, crew, and keywords using cosine similarity to provide personalized movie suggestions.
✦ Key Features
- ♥Content-based filtering using Cosine Similarity
- ♥Metadata analysis including genres, cast, and keywords
- ♥Personalized movie recommendations
- ♥Efficient NLP processing of textual data
✦ Methodology
Utilizes Natural Language Processing (NLP) for content analysis:
01.
Bag of Words Construction
Aggregated movie metadata (description, genre, actors) into a single textual 'tag' for each movie.
02.
Vectorization
Applied CountVectorizer to transform text tags into high-dimensional vectors, creating a mathematical representation of movie content.
03.
Similarity Matrix
Computed the Cosine Similarity matrix effectively measuring the angular distance between every pair of movie vectors to find the closest matches.