Machine Learning

, Volume 106, Issue 6, pp 771–798

Stream-based semi-supervised learning for recommender systems

Article

DOI: 10.1007/s10994-016-5614-4

Cite this article as:
Matuszyk, P. & Spiliopoulou, M. Mach Learn (2017) 106: 771. doi:10.1007/s10994-016-5614-4
  • 290 Downloads

Abstract

To alleviate the problem of data sparsity inherent to recommender systems, we propose a semi-supervised framework for stream-based recommendations. Our framework uses abundant unlabelled information to improve the quality of recommendations. We extend a state-of-the-art matrix factorization algorithm by the ability to add new dimensions to the matrix at runtime and implement two approaches to semi-supervised learning: co-training and self-learning. We introduce a new evaluation protocol including statistical testing and parameter optimization. We then evaluate our framework on five real-world datasets in a stream setting. On all of the datasets our method achieves statistically significant improvements in the quality of recommendations.

Keywords

Recommender systems Semi-supervised learning Matrix factorization Collaborative filtering Stream mining 

Copyright information

© The Author(s) 2017

Authors and Affiliations

  1. 1.Otto-von-Guericke-University MagdeburgMagdeburgGermany

Personalised recommendations