Collaborative Filtering

Introduction

Numerous ML algorithms have been targeted for LA but collaborative filtering (CF) is one of the most promising methods that will fulfill the objectives in learning.

The goal of CF is to recommend unfamiliar items to a user based on existing ratings of those items by others. For example, let's assume two users (user-1 and user-2) rate n items similarly will eventually rate other items similarly.

CF for LA

Implementations of CF in LA have been reported to efficiently analyze concurrently both learners and questions; however there has been less focus on comparing many algorithms and identify the parameters that work best for improving prediction performance and easy interpretation.

CF methods use a set of preferences for items by users to predict other items.

Two categories of CF:


For more details, please contact me here.
Date of last modification: 2021