From the course: Building a Recommendation System with Python Machine Learning and AI
Unlock the full course today
Join today to access over 23,200 courses taught by industry experts.
Model-based collaborative filtering systems
From the course: Building a Recommendation System with Python Machine Learning and AI
Model-based collaborative filtering systems
- [Instructor] Turning now to model-based collaborative filtering systems. With these systems, you build a model from user ratings and then make recommendations based on that model. This offers a speed and scalability that's not available when you first refer back to the entire dataset to make a prediction. In the demo for the segment, you're going to see truncated singular value decomposition. You're also going to see something called a utility matrix. Utility matrix is also known as user item matrix. These matrices contain values for each user, each item, and the rating each user gave to each item. Another thing to note is that utility matrices are sparse, because every user does not review every item. Actually, only a few users provide reviews for a few items, so in these matrices, you're likely to see mostly null values. Before explaining the truncated version, let me explain to you regular singular value decomposition, or SVD. SVD is a linear algebra method you can use to…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.