Luke HaushalterDimensionality Reduction: Principal Component Analysis (PCA) pt. 1Many of the most famous examples of recommender systems are for huge content platforms; think YouTube, Netflix, or Amazon. And a byproduct…Sep 8, 2019Sep 8, 2019
Luke HaushalterError! Error!: Measuring Error in your recommender systemSo you’ve built your recommender system, and it thinks based on your model that User A will give Item 1 a rating of 5.0 out of a possible…Aug 26, 2019Aug 26, 2019
Luke HaushalterVariance and Covariance, pt. 2A few weeks ago I posted an article that explained the variance of a set of values.Aug 19, 2019Aug 19, 2019
Luke HaushalterBrief History of the Netflix PrizeA commonly referenced moment in recommender system history is the “Netflix Prize,” which brought the study of recommender systems to a…Aug 12, 2019Aug 12, 2019
Luke HaushalterDeep Freeze: The Cold Start Problem of Collaborative Filtering Recommender SystemsAs we mentioned in a previous story, collaborative filtering recommender systems generate recommendations for items to users based on how…Aug 5, 2019Aug 5, 2019
Luke HaushalterVariance and Covariance, pt. 1Some of the important formulas that come into play when applying recommender systems involve the concepts of variance and covariance. This…Jul 29, 2019Jul 29, 2019
Luke HaushalterGame on: Steam’s new collaborative filtering recommenderEarlier this month, gaming giant Steam added a new experimental feature to its store: a collaborative-filtering powered recommender system…Jul 22, 2019Jul 22, 2019
Luke HaushalterPearson CorrelationOne main component in building a collaborative filtering algorithm is a numerical representation of the similarity between two users. One…Jul 15, 2019Jul 15, 2019
Luke HaushalterIn our last post, we had talked about the central premise behind Collaborative Filtering, and we…A quantitative measure of your tasteJun 26, 2019Jun 26, 2019
