Leveraging Kernel Incorporated Matrix Factorization for Smartphone Application Recommendation

The explosive growth in the number of smartphone applications (apps) available on the market poses a significant challenge to making personalized recommendations based on user preferences. The training data usually consists of sparse binary implicit feedback (i.e. user-app installation pairs), which results in ambiguities in representing the users interests due to a lack of negative examples. In this paper, we propose two kernel incorporated matrix factorization models to predict user preferences for apps by introducing the categorical information of the apps. The two models extends Probabilistic Matrix Factorization (PMF) by constraining the user and app latent features to be similar to their neighbors in the app-categorical space, and adopts Stochastic Gradient Decent (SGD)-based methods to learn the models. The experimental results show that our model outperforms the baselines, in terms of two ranking-oriented evaluation metrics.

[1]  Seungjin Choi,et al.  Hierarchical variational Bayesian matrix co-factorization , 2012, 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[2]  Martha Larson,et al.  Collaborative Filtering beyond the User-Item Matrix , 2014, ACM Comput. Surv..

[3]  Zibin Zheng,et al.  Predicting Quality of Service for Selection by Neighborhood-Based Collaborative Filtering , 2013, IEEE Transactions on Systems, Man, and Cybernetics: Systems.

[4]  G. James Blaine,et al.  Continuous Monitoring of Physiologic Variables with a Dedicated Minicomputer , 1975, Computer.

[5]  Ruslan Salakhutdinov,et al.  Bayesian probabilistic matrix factorization using Markov chain Monte Carlo , 2008, ICML '08.

[6]  Jun Wang,et al.  Unifying user-based and item-based collaborative filtering approaches by similarity fusion , 2006, SIGIR.

[7]  MengChu Zhou,et al.  An Efficient Non-Negative Matrix-Factorization-Based Approach to Collaborative Filtering for Recommender Systems , 2014, IEEE Transactions on Industrial Informatics.

[8]  Qiang Yang,et al.  One-Class Collaborative Filtering , 2008, 2008 Eighth IEEE International Conference on Data Mining.

[9]  Tat-Seng Chua,et al.  Addressing cold-start in app recommendation: latent user models constructed from twitter followers , 2013, SIGIR.

[10]  Thomas Hofmann,et al.  Latent semantic models for collaborative filtering , 2004, TOIS.

[11]  Guillermo Sapiro,et al.  Kernelized Probabilistic Matrix Factorization: Exploiting Graphs and Side Information , 2012, SDM.

[12]  Ruslan Salakhutdinov,et al.  Probabilistic Matrix Factorization , 2007, NIPS.

[13]  Tat-Seng Chua,et al.  New and improved: modeling versions to improve app recommendation , 2014, SIGIR.

[14]  Juan-Zi Li,et al.  Typicality-Based Collaborative Filtering Recommendation , 2014, IEEE Transactions on Knowledge and Data Engineering.

[15]  Max Welling,et al.  Bayesian Matrix Factorization with Side Information and Dirichlet Process Mixtures , 2010, AAAI.

[16]  Jie Zhang,et al.  Leveraging Decomposed Trust in Probabilistic Matrix Factorization for Effective Recommendation , 2014, AAAI.

[17]  Kamal Ali,et al.  GetJar mobile application recommendations with very sparse datasets , 2012, KDD.

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

[19]  Sunho Park,et al.  Hierarchical Bayesian Matrix Factorization with Side Information , 2013, IJCAI.

[20]  Jianying Hu,et al.  One-Class Matrix Completion with Low-Density Factorizations , 2010, 2010 IEEE International Conference on Data Mining.