Two-level matrix factorization for recommender systems

Fangfang LI, Guandong XU, Longbing CAO

Research output: Contribution to journalArticlespeer-review

20 Citations (Scopus)


Many existing recommendation methods such as matrix factorization (MF) mainly rely on user–item rating matrix, which sometimes is not informative enough, often suffering from the cold-start problem. To solve this challenge, complementary textual relations between items are incorporated into recommender systems (RS) in this paper. Specifically, we first apply a novel weighted textual matrix factorization (WTMF) approach to compute the semantic similarities between items, then integrate the inferred item semantic relations into MF and propose a two-level matrix factorization (TLMF) model for RS. Experimental results on two open data sets not only demonstrate the superiority of TLMF model over benchmark methods, but also show the effectiveness of TLMF for solving the cold-start problem. Copyright © 2015 The Natural Computing Applications Forum.

Original languageEnglish
Pages (from-to)2267-2278
JournalNeural Computing and Applications
Early online dateSept 2015
Publication statusPublished - Nov 2016


Li, F., Xu, G., & Cao, L. (2016). Two-level matrix factorization for recommender systems. Neural Computing and Applications, 27, 2267-2278.


  • Recommender system
  • Matrix factorization
  • Latent factor model
  • Textual semantic relation


Dive into the research topics of 'Two-level matrix factorization for recommender systems'. Together they form a unique fingerprint.