Predicting response in mobile advertising with hierarchical importance-aware factorization machine

Mobile advertising has recently seen dramatic growth, fueled by the global proliferation of mobile phones and devices. The task of predicting ad response is thus crucial for maximizing business revenue. However, ad response data change dynamically over time, and are subject to cold-start situations in which limited history hinders reliable prediction. There is also a need for a robust regression estimation for high prediction accuracy, and good ranking to distinguish the impacts of different ads. To this end, we develop a Hierarchical Importance-aware Factorization Machine (HIFM), which provides an effective generic latent factor framework that incorporates importance weights and hierarchical learning. Comprehensive empirical studies on a real-world mobile advertising dataset show that HIFM outperforms the contemporary temporal latent factor models. The results also demonstrate the efficacy of the HIFM's importance-aware and hierarchical learning in improving the overall prediction and prediction in cold-start scenarios, respectively.

[1]  Ambuj Tewari,et al.  Stochastic methods for l1 regularized loss minimization , 2009, ICML '09.

[2]  Markus Reichstein,et al.  Gap Filling in the Plant Kingdom - Trait Prediction Using Hierarchical Probabilistic Matrix Factorization , 2012, ICML.

[3]  Thomas Bäck,et al.  Evolutionary algorithms in theory and practice - evolution strategies, evolutionary programming, genetic algorithms , 1996 .

[4]  Sachin Garg,et al.  Response prediction using collaborative filtering with hierarchies and side-information , 2011, KDD.

[5]  Rajiv Khanna,et al.  Estimating rates of rare events with multiple hierarchies through scalable log-linear models , 2010, KDD '10.

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

[7]  Kush R. Varshney,et al.  Dynamic matrix factorization: A state space approach , 2011, 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[8]  Tommi S. Jaakkola,et al.  Maximum-Margin Matrix Factorization , 2004, NIPS.

[9]  Qiang Yang,et al.  Contextual Collaborative Filtering via Hierarchical Matrix Factorization , 2012, SDM.

[10]  Joaquin Quiñonero Candela,et al.  Web-Scale Bayesian Click-Through rate Prediction for Sponsored Search Advertising in Microsoft's Bing Search Engine , 2010, ICML.

[11]  Inderjit S. Dhillon,et al.  A spatio-temporal approach to collaborative filtering , 2009, RecSys '09.

[12]  Tom Fawcett,et al.  ROC Graphs: Notes and Practical Considerations for Researchers , 2007 .

[13]  F. Wilcoxon.  Individual Comparisons by Ranking Methods , 1945 .

[14]  Steffen Rendle.  Scaling Factorization Machines to Relational Data , 2013, Proc. VLDB Endow..

[15]  Steffen Rendle,et al.  Factorization Machines , 2010, 2010 IEEE International Conference on Data Mining.

[16]  Qiang Yang,et al.  Personalized click model through collaborative filtering , 2012, WSDM '12.

[17]  Steffen Rendle,et al.  Factorization Machines with libFM , 2012, TIST.

[18]  Matthew Richardson,et al.  Predicting clicks: estimating the click-through rate for new ads , 2007, WWW '07.

[19]  P. Comon,et al.  Tensor decompositions, alternating least squares and other tales , 2009 .

[20]  D. Sculley,et al.  Combined regression and ranking , 2010, KDD.

[21]  Deepak Agarwal,et al.  Spatio-temporal models for estimating click-through rate , 2009, WWW '09.

[22]  Ashish Goel,et al.  Truthful auctions for pricing search keywords , 2006, EC '06.