O'Reilly logo

Building Recommendation Engines by Suresh Kumar Gorakala

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

SVD recommenders

Similar to the item-based and user-based recommender systems explained earlier, we can also use model-based recommender implementations in Mahout, such as SVDRecommender, which uses matrix factorization methods to generate recommendations.

The steps are similar to previous implementations. Two important steps that need to be understood here are as follows:

  • The org.apache.mahout.cf.taste.impl.recommender.svd.ALSWRFactorizer class, which factorizes the user rating matrix using Alternating-Least-Squares with Weighted-λ-Regularization. The ALSWRFactorizer class constructor takes parameters such as DataModel, the number of features, the regularization parameter, and the number of iterations as inputs. This ALSWRFactorizer class instance ...

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

Start Free Trial

No credit card required