Geographic Segmentation via Latent Poisson Factor Model

Discovering latent structures in spatial data is of critical importance to understanding the user behavior of location-based services. In this paper, we study the problem of geographic segmentation of spatial data, which involves dividing a collection of observations into distinct geo-spatial regions and uncovering abstract correlation structures in the data. We introduce a novel, Latent Poisson Factor (LPF) model to describe spatial count data. The model describes the spatial counts as a Poisson distribution with a mean that factors over a joint item-location latent space. The latent factors are constrained with weak labels to help uncover interesting spatial dependencies. We study the LPF model on a mobile app usage data set and a news article readership data set. We empirically demonstrate its effectiveness on a variety of prediction tasks on these two data sets.

[1]  Ramesh Nallapati,et al.  Labeled LDA: A supervised topic model for credit attribution in multi-labeled corpora , 2009, EMNLP.

[2]  Jiawei Han,et al.  Geographical topic discovery and comparison , 2011, WWW.

[3]  Hans-Peter Kriegel,et al.  A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise , 1996, KDD.

[4]  Alexander J. Smola,et al.  Discovering geographical topics in the twitter stream , 2012, WWW.

[5]  Johannes Schöning,et al.  Falling asleep with Angry Birds, Facebook and Kindle: a large scale study on mobile application usage , 2011, Mobile HCI.

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

[7]  Nikos Mamoulis,et al.  Density-based place clustering in geo-social networks , 2014, SIGMOD Conference.

[8]  David J. Crandall,et al.  Beyond co-occurrence: discovering and visualizing tag relationships from geo-spatial and temporal similarities , 2012, WSDM '12.

[9]  Charu C. Aggarwal,et al.  Mining collective intelligence in diverse groups , 2013, WWW.

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

[11]  David B. Dunson,et al.  Beta-Negative Binomial Process and Poisson Factor Analysis , 2011, AISTATS.

[12]  Chengqi Zhang,et al.  Modeling Location-Based User Rating Profiles for Personalized Recommendation , 2015, ACM Trans. Knowl. Discov. Data.

[13]  Cyrus Shahabi,et al.  Knowledge discovery from users Web-page navigation , 1997, Proceedings Seventh International Workshop on Research Issues in Data Engineering. High Performance Database Management for Large-Scale Applications.

[14]  Krzysztof Janowicz,et al.  On the semantic annotation of places in location-based social networks , 2011, KDD.

[15]  Jiawei Han,et al.  Geographic Data Mining and Knowledge Discovery , 2001 .

[16]  Tao Luo,et al.  Discovery and Evaluation of Aggregate Usage Profiles for Web Personalization , 2004, Data Mining and Knowledge Discovery.

[17]  Ahmed Eldawy,et al.  LARS: A Location-Aware Recommender System , 2012, 2012 IEEE 28th International Conference on Data Engineering.

[18]  David M. Blei,et al.  Content-based recommendations with Poisson factorization , 2014, NIPS.

[19]  Umeshwar Dayal,et al.  From User Access Patterns to Dynamic Hypertext Linking , 1996, Comput. Networks.

[20]  Nadia Magnenat-Thalmann,et al.  Who, where, when and what: discover spatio-temporal topics for twitter users , 2013, KDD.

[21]  Jiawei Han,et al.  Efficient and Effective Clustering Methods for Spatial Data Mining , 1994, VLDB.

[22]  Alexandros Karatzoglou,et al.  Climbing the app wall: enabling mobile app discovery through context-aware recommendations , 2012, CIKM '12.

[23]  Ali Taylan Cemgil,et al.  Bayesian Inference for Nonnegative Matrix Factorisation Models , 2009, Comput. Intell. Neurosci..

[24]  Philip S. Yu,et al.  Review spam detection via time series pattern discovery , 2012, WWW.