Image-regulated graph topic model for cross-media topic detection

In recent years, pictures and videos have become ubiquitous on the Internet, which encourage the development of algorithm that analyze their semantic contents for detecting topics. Among them, topic modeling plays an essential role in discovering topics from document collections. However, with rich auxiliary information (such as geo-information, user-annotated tags, pictures and videos) rising up around the text, traditional topic models show their limitations to discover latent topics effectively from the cross-media data. To address this problem, we propose a novel Image-regulated Graph Topic Model (IGTM), which combines cross-media data together in the modeling process. By utilizing auxiliary relation information among images, IGTM can achieve higher quality underlying topics as image relationships could serve as weakly-supervised information for topic modeling. Experimental results over two cross-media datasets demonstrate the effectiveness of our model.

[1]  David M. Blei,et al.  Relational Topic Models for Document Networks , 2009, AISTATS.

[2]  Qingming Huang,et al.  Large scale image understanding with non-convex multi-task learning , 2014, The 2014 5th International Conference on Game Theory for Networks.

[3]  Xiaohui Yan,et al.  A biterm topic model for short texts , 2013, WWW.

[4]  Xueqi Cheng,et al.  Intent-aware query similarity , 2011, CIKM '11.

[5]  Tao Mei,et al.  Travel Recommendation via Author Topic Model Based Collaborative Filtering , 2015, MMM.

[6]  Michael I. Jordan,et al.  Latent Dirichlet Allocation , 2001, J. Mach. Learn. Res..

[7]  Qingming Huang,et al.  Learning Hierarchical Semantic Description Via Mixed-Norm Regularization for Image Understanding , 2012, IEEE Transactions on Multimedia.

[8]  Gang Hua,et al.  Semi-supervised Relational Topic Model for Weakly Annotated Image Recognition in Social Media , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[9]  Mark Steyvers,et al.  Finding scientific topics , 2004, Proceedings of the National Academy of Sciences of the United States of America.

[10]  Chong Wang,et al.  Simultaneous image classification and annotation , 2009, 2009 IEEE Conference on Computer Vision and Pattern Recognition.

[11]  Aristides Gionis,et al.  Query similarity by projecting the query-flow graph , 2010, SIGIR.

[12]  ChengXiang Zhai,et al.  Unsupervised Feature Selection for Multi-View Clustering on Text-Image Web News Data , 2014, CIKM.

[13]  Yalou Huang,et al.  Hashtag Graph Based Topic Model for Tweet Mining , 2014, 2014 IEEE International Conference on Data Mining.

[14]  Tom Minka,et al.  Expectation Propagation for approximate Bayesian inference , 2001, UAI.

[15]  Thomas Hofmann,et al.  Probabilistic Latent Semantic Indexing , 1999, SIGIR Forum.

[16]  Thomas L. Griffiths,et al.  The Author-Topic Model for Authors and Documents , 2004, UAI.

[17]  Michael I. Jordan,et al.  Modeling annotated data , 2003, SIGIR.

[18]  Qingming Huang,et al.  Partial-Duplicate Image Retrieval via Saliency-Guided Visual Matching , 2013, IEEE MultiMedia.