A Novel Recommendation Algorithm Incorporating Temporal Dynamics, Reviews and Item Correlation

Recommender systems (RS) exploit user ratings on items and side information to make personalized recommendations. In order to recommend the right products to users, RS must accurately model the implicit preferences of each user and the properties of each product. In reality, both user preferences and item properties are changing dynamically over time, so treating the historical decisions of a user or the received comments of an item as static is inappropriate. Besides, the review text accompanied with a rating score can help us to understand why a user likes or dislikes an item, so temporal dynamics and text information in reviews are important side information for recommender systems. Moreover, compared with the large number of available items, the number of items a user can buy is very limited, which is called the sparsity problem. In order to solve this problem, utilizing item correlation provides a promising solution. Although famous methods like TimeSVD++, TopicMF and CoFactor partially take temporal dynamics, reviews and correlation into consideration, none of them combine these information together for accurate recommendation. Therefore, in this paper we propose a novel combined model called TmRevCo which is based on matrix factorization. Our model combines the dynamic user factor of TimeSVD++ with the hidden topic of each review text mined by the topic model of TopicMF through a new transformation function. Meanwhile, to support our five-scoring datasets, we use a more appropriate item correlation measure in CoFactor and associate the item factors of CoFactor with that of matrix factorization. Our model comprehensively combines the temporal dynamics, review information and item correlation simultaneously. Experimental results on three real-world datasets show that our proposed model leads to significant improvement compared with the baseline methods. key words: matrix factorization, time-aware recommender systems, topic model, item correlation

[1]  Martin Ester,et al.  On the design of LDA models for aspect-based opinion mining , 2012, CIKM.

[2]  Amélie Marian,et al.  Beyond the Stars: Improving Rating Predictions using Review Text Content , 2009, WebDB.

[3]  Mihai Surdeanu,et al.  The Stanford CoreNLP Natural Language Processing Toolkit , 2014, ACL.

[4]  Gerhard Weikum,et al.  Continuous Experience-aware Language Model , 2016, KDD.

[5]  Jure Leskovec,et al.  From amateurs to connoisseurs: modeling the evolution of user expertise through online reviews , 2013, WWW.

[6]  Jimeng Sun,et al.  Temporal recommendation on graphs via long- and short-term preference fusion , 2010, KDD.

[7]  Jure Leskovec,et al.  Hidden factors and hidden topics: understanding rating dimensions with review text , 2013, RecSys.

[8]  Yehuda Koren,et al.  Yahoo! music recommendations: modeling music ratings with temporal dynamics and item taxonomy , 2011, RecSys '11.

[9]  Bin Xu,et al.  Content Recommendation System Based on Private Dynamic User Profile , 2007, 2007 International Conference on Machine Learning and Cybernetics.

[10]  David M. Blei,et al.  Factorization Meets the Item Embedding: Regularizing Matrix Factorization with Item Co-occurrence , 2016, RecSys.

[11]  Sheng Wang,et al.  Supervised Topic Model with Consideration of User and Item , 2013, AAAI.

[12]  Wei Chu,et al.  Personalized recommendation on dynamic content using predictive bilinear models , 2009, WWW '09.

[13]  Guokun Lai,et al.  Daily-Aware Personalized Recommendation based on Feature-Level Time Series Analysis , 2015, WWW.

[14]  Yehuda Koren,et al.  Collaborative filtering with temporal dynamics , 2009, KDD.

[15]  Zhiting Hu,et al.  Dynamic User Modeling in Social Media Systems , 2015, TOIS.

[16]  Gerhard Weikum,et al.  Experience-Aware Item Recommendation in Evolving Review Communities , 2015, 2015 IEEE International Conference on Data Mining.

[17]  Yehuda Koren,et al.  Matrix Factorization Techniques for Recommender Systems , 2009, Computer.

[18]  Jie Zhang,et al.  TopicMF: Simultaneously Exploiting Ratings and Reviews for Recommendation , 2014, AAAI.