TagMF represents an interactive recommending approach that merges user-generated tags with latent factors derived from user ratings to increase user profile transparency and interactive control in Recommender Systems. Taking advantage of the Matrix Factorization technique widely used in Collaborative Filtering, the method learns an integrated model of tags and latent factors, thus enabling users to understand and manipulate their preference profile expressed implicitly in the (intransparent) latent factor space through explicitly presented textual tags.

Related research topics


Benedikt Loepp



  • Tim Donkers
  • Timm Kleemann


Towards Understanding Latent Factors and User Profiles by Enhancing Matrix Factorization with Tags

Tag-Enhanced Collaborative Filtering for Increasing Transparency and Interactive Control

Merging Latent Factors and Tags to Increase Interactive Control of Recommendations