Two-level matrix factorization for recommender systems

Abstract

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.

Movie Recommendation Systems using Sentiment Analysis and Cosine Similarity

“…The results showed an enhanced recommendation system can be used by bridging the gap between the communication knowledge , social networking sites and programs being watched over television. Also a rating inference approach to transform text reviews in form of ratings for easy integration in sentiment analysis and CF technique was proposed in [24] .…”

Recommending Curated Content Using Implicit Feedback

“…Different types of matrix factorization techniques have been used widely in hybrid recommender systems. In [7] a two-level matrix factorization (TLMF) has been proposed. TLMF computes the semantic relations between items based on a novel approach -Weighted Textual Matrix Factorization (WTMF).…”

Peer Review #3 of "Latent based temporal optimization approach for improving the performance of collaborative filtering (v0.1)"

“…Integration of temporal preferences with factorization methods to solve the sparsity issue has yielded a better performance compared to basic factorization approaches (Al-Hadi et al, 2017b;Li et al, 2016;Nilashi et al, 2019Nilashi et al, , 2014. The temporal dynamics approach (Koren, 2009) separates the time period of preferences into static digit of bins and extracts a universal weight according to the stochastic gradient descent method to reduce overfitting.…”