Constructing an Interaction Behavior Model for Web Image Search

User interaction behavior is a valuable source of implicit relevance feedback. In Web image search a different type of search result presentation is used than in general Web search, which leads to different interaction mechanisms and user behavior. For example, image search results are self-contained, so that users do not need to click the results to view the landing page as in general Web search, which generates sparse click data. Also, two-dimensional result placement instead of a linear result list makes browsing behaviors more complex. Thus, it is hard to apply standard user behavior models (e.g., click models) developed for general Web search to Web image search. In this paper, we conduct a comprehensive image search user behavior analysis using data from a lab-based user study as well as data from a commercial search log. We then propose a novel interaction behavior model, called grid-based user browsing model (GUBM), whose design is motivated by observations from our data analysis. GUBM can both capture users' interaction behavior, including cursor hovering, and alleviate position bias. The advantages of GUBM are two-fold: (1) It is based on an unsupervised learning method and does not need manually annotated data for training. (2) It is based on user interaction features on search engine result pages (SERPs) and is easily transferable to other scenarios that have a grid-based interface such as video search engines. We conduct extensive experiments to test the performance of our model using a large-scale commercial image search log. Experimental results show that in terms of behavior prediction (perplexity), and topical relevance and image quality (normalized discounted cumulative gain (NDCG)), GUBM outperforms state-of-the-art baseline models as well as the original ranking. We make the implementation of GUBM and related datasets publicly available for future studies.

[1]  Rossano Schifanella,et al.  A Large-Scale Study of User Image Search Behavior on the Web , 2015, CHI.

[2]  Chong-Wah Ngo,et al.  Click-through-based cross-view learning for image search , 2014, SIGIR.

[3]  Xian-Sheng Hua,et al.  The role of attractiveness in web image search , 2011, ACM Multimedia.

[4]  Pu-Jen Cheng,et al.  Learning user reformulation behavior for query auto-completion , 2014, SIGIR.

[5]  M. de Rijke,et al.  Online Learning to Rank for Information Retrieval: SIGIR 2016 Tutorial , 2016, SIGIR.

[6]  Bracha Shapira,et al.  Study of the usefulness of known and new implicit indicators and their optimal combination for accurate inference of users interests , 2006, SAC.

[7]  Yiqun Liu,et al.  Incorporating vertical results into search click models , 2013, SIGIR.

[8]  Hsiao-Tieh Pu,et al.  A comparative analysis of web image and textual queries , 2005, Online Inf. Rev..

[9]  Chao Liu,et al.  Efficient multiple-click models in web search , 2009, WSDM '09.

[10]  Yue Wang,et al.  Beyond Ranking: Optimizing Whole-Page Presentation , 2016, WSDM.

[11]  M. de Rijke,et al.  Click Models for Web Search , 2015, Click Models for Web Search.

[12]  Filip Radlinski,et al.  Online Evaluation for Information Retrieval , 2016, Found. Trends Inf. Retr..

[13]  Yiqun Liu,et al.  Incorporating revisiting behaviors into click models , 2012, WSDM '12.

[14]  Maarten de Rijke,et al.  A Context-aware Time Model for Web Search , 2016, SIGIR.

[15]  Ryen W. White Interactions with Search Systems , 2016 .

[16]  Susan T. Dumais,et al.  Improving Web Search Ranking by Incorporating User Behavior Information , 2019, SIGIR Forum.

[17]  Sara Shatford,et al.  Analyzing the Subject of a Picture: A Theoretical Approach , 1986 .

[18]  Yiqun Liu,et al.  Influence of Vertical Result in Web Search Examination , 2015, SIGIR.

[19]  Yiqun Liu,et al.  Why People Search for Images using Web Search Engines , 2017, WSDM.

[20]  Ryen W. White,et al.  Improving searcher models using mouse cursor activity , 2012, SIGIR '12.

[21]  Enhong Chen,et al.  Context-aware query suggestion by mining click-through and session data , 2008, KDD.

[22]  Qiang Yang,et al.  Beyond ten blue links: enabling user click modeling in federated web search , 2012, WSDM '12.

[23]  Yiqun Liu,et al.  Incorporating Non-sequential Behavior into Click Models , 2015, SIGIR.

[24]  Amanda Spink,et al.  A study and comparison of multimedia Web searching: 1997-2006 , 2009, J. Assoc. Inf. Sci. Technol..

[25]  Thorsten Joachims,et al.  Optimizing search engines using clickthrough data , 2002, KDD.

[26]  Xiaolong Li,et al.  Inferring search behaviors using partially observable Markov (POM) model , 2010, WSDM '10.

[27]  Vidit Jain,et al.  Learning to re-rank: query-dependent image re-ranking using click data , 2011, WWW.

[28]  Y. Rui,et al.  Learning to Rank Using User Clicks and Visual Features for Image Retrieval , 2015, IEEE Transactions on Cybernetics.

[29]  Jaana Kekäläinen,et al.  Cumulated gain-based evaluation of IR techniques , 2002, TOIS.

[30]  J. Fleiss Measuring nominal scale agreement among many raters. , 1971 .

[31]  Hongbo Deng,et al.  A two-dimensional click model for query auto-completion , 2014, SIGIR.

[32]  Olivier Chapelle,et al.  A dynamic bayesian network click model for web search ranking , 2009, WWW '09.

[33]  M. de Rijke,et al.  A Click Sequence Model for Web Search , 2018, SIGIR.

[34]  Benjamin Piwowarski,et al.  A user browsing model to predict search engine click data from past observations. , 2008, SIGIR '08.

[35]  Rossano Schifanella,et al.  Leveraging User Interaction Signals for Web Image Search , 2016, SIGIR.

[36]  Katja Hofmann,et al.  Reusing historical interaction data for faster online learning to rank for IR , 2013, DIR.

[37]  Nick Craswell,et al.  An experimental comparison of click position-bias models , 2008, WSDM '08.

[38]  Yiqun Liu,et al.  Time-Aware Click Model , 2016, ACM Trans. Inf. Syst..

[39]  M. de Rijke,et al.  An Introduction to Click Models for Web Search: SIGIR 2015 Tutorial , 2015, SIGIR.

[40]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[41]  Wei Wu,et al.  Learning query and document similarities from click-through bipartite graph with metadata , 2013, WSDM.

[42]  Meng Wang,et al.  Investigating Examination Behavior of Image Search Users , 2017, SIGIR.

[43]  M. de Rijke,et al.  Ranking for Relevance and Display Preferences in Complex Presentation Layouts , 2018, SIGIR.

[44]  Eugene Agichtein,et al.  Exploring mouse movements for inferring query intent , 2008, SIGIR '08.