From the course: Building Recommender Systems with Machine Learning and AI
Join today to access over 22,600 courses taught by industry experts.
Singular value decomposition (SVD) - Python Tutorial
From the course: Building Recommender Systems with Machine Learning and AI
Singular value decomposition (SVD)
- [Instructor] So, just like we can run PCA on a four-dimensional Iris dataset, we can also run it on our multi-dimensional movie rating dataset, where every dimension is a movie. We'll call this ratings matrix that has users as rows R. Just like it did with our Iris dataset, PCA can boil this down to a much smaller number of dimensions that best describe the variance in the data. And often, the dimensions it finds correspond to features humans have learned to associate with movies as well; for example, how action-y is a movie, how romantic is it, how funny is it. Whatever it is about movies that causes individuals to rate them differently, PCA will find those latent features and extract them from the data. PCA doesn't know what they mean, but it finds them nonetheless. So, we could ask PCA to distill things down to, say, three dimensions in this example, and it would boil our ratings down to three latent features it identified. PCA won't know what to call them, but let's say they end…
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.
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.