A machine learning approach to web personalization

Most web sites today are designed one-size-fits-all: all visitors see the exact same pages regardless of interests, previous interactions, or, frequently, even browsing client (desktop PC or wireless PDA). But one size often does not fit all. Instead of presenting the same content, the web experience should be dynamic and personalized, adapting to visitors' preferences as evinced in previous interactions. This thesis proposes a framework for personalizing the web experience. Within our PROTEUS framework, we view personalization as a two-step process of first modeling users, and then improving the site given the model. We frame this problem as a machine learning task: the goal is to predict users' web navigation given their previous behavior. We explore several means of personalization, including improving the wireless web and building personalized, dynamic portals, and concentrate on one in particular—automatically adding shortcuts to likely navigation destinations. A challenge in modeling web navigation is that training data for an entire site may be plentiful, but sparse for any individual page. This difficulty can be overcome, however, by noting that most large web sites have a rich underlying relational structure that can be exploited for generalization: pages can belong to different types (e.g., pages about laptop computers versus pages about printers), with each type described by a different set of attributes (e.g., size of display versus printing speed). We leverage this structure by developing relational Markov models (RMMs), a novel extension to Markov models. States in an RMM belong to relations and are described by variables over hierarchically structured domains. Based on these hierarchies, the RMM defines sets of related states, learns transition probabilities between these sets, and uses shrinkage to estimate transitions between individual pages. This thesis presents RMMs in detail and provides results showing that they outperform traditional Markov models for predicting web navigation by a substantial margin.

[1]  Keiji Kanazawa,et al.  A model for reasoning about persistence and causation , 1989 .

[2]  T. Joachims.  WebWatcher : A Tour Guide for the World Wide Web , 1997 .

[3]  Pedro M. Domingos,et al.  Adaptive Web Navigation for Wireless Devices , 2001, IJCAI.

[4]  Michael I. Jordan,et al.  Factorial Hidden Markov Models , 1995, Machine Learning.

[5]  Juliana Freire,et al.  WebViews: accessing personalized web content and services , 2001, WWW '01.

[6]  Carla E. Brodley,et al.  KDD-Cup 2000 organizers' report: peeling the onion , 2000, SKDD.

[7]  Ingrid Zukerman,et al.  Predicting users' requests on the WWW , 1999 .

[8]  Lise Getoor,et al.  From Instances to Classes in Probabilistic Relational Models , 2000, ICML 2000.

[9]  Michael J. Pazzani,et al.  A learning agent for wireless news access , 2000, IUI '00.

[10]  Pedro M. Domingos,et al.  Mixed initiative interfaces for learning tasks: SMARTedit talks back , 2001, IUI '01.

[11]  Qiang Yang,et al.  A prediction system for multimedia pre-fetching in Internet , 2000, ACM Multimedia.

[12]  Michael J. Pazzani,et al.  Syskill & Webert: Identifying Interesting Web Sites , 1996, AAAI/IAAI, Vol. 1.

[13]  Alfred Kobsa,et al.  User Modeling and User-Adapted Interaction , 1994, User Modeling and User-Adapted Interaction.

[14]  Alon Y. Halevy,et al.  Declarative Web Site Management with Tiramisu , 1999, WebDB.

[15]  David A. Cohn,et al.  Improving generalization with active learning , 1994, Machine Learning.

[16]  Israel Ben-Shaul,et al.  Automatically Organizing Bookmarks per Contents , 1996, Comput. Networks.

[17]  Susan T. Dumais,et al.  Bringing order to the Web: automatically categorizing search results , 2000, CHI.

[18]  Peter Pirolli,et al.  Mining Longest Repeating Subsequences to Predict World Wide Web Surfing , 1999, USENIX Symposium on Internet Technologies and Systems.

[19]  Andrew McCallum,et al.  Information Extraction with HMMs and Shrinkage , 1999 .

[20]  Juliana Freire,et al.  Personalizing the Web using site descriptions , 1999, Proceedings. Tenth International Workshop on Database and Expert Systems Applications. DEXA 99.

[21]  D. Flannanghan.  JavaScript: The definitive guide , 1999 .

[22]  Thomas G. Dietterich.  State Abstraction in MAXQ Hierarchical Reinforcement Learning , 1999, NIPS.

[23]  George Karypis,et al.  Selective Markov models for predicting Web page accesses , 2004, TOIT.

[24]  James Kelly,et al.  AutoClass: A Bayesian Classification System , 1993, ML.

[25]  Oren Etzioni,et al.  Adaptive web sites: cluster mining and conceptual clustering for index page synthesis , 2001 .

[26]  S. Muggleton.  Stochastic Logic Programs , 1996 .

[27]  Peter Haddawy,et al.  Answering Queries from Context-Sensitive Probabilistic Knowledge Bases , 1997, Theor. Comput. Sci..

[28]  Oren Etzioni,et al.  Adaptive Web Sites: an AI Challenge , 1997, IJCAI.

[29]  Michael D. Smith,et al.  Using Path Profiles to Predict HTTP Requests , 1998, Comput. Networks.

[30]  Kristian J. Hammond,et al.  Mining navigation history for recommendation , 2000, IUI '00.

[31]  De,et al.  Relational Reinforcement Learning , 2022 .

[32]  Prabhakar Raghavan,et al.  Scalable feature selection, classification and signature generation for organizing large text databases into hierarchical topic taxonomies , 1998, The VLDB Journal.

[33]  Paul P. Maglio,et al.  Intermediaries personalize information streams , 2000, CACM.

[34]  Edward Y. Chang,et al.  PowerBookmarks: a system for personalizable Web information organization, sharing, and management , 1999, SIGMOD '99.

[35]  Michael J. Pazzani,et al.  Adaptive interfaces for ubiquitous web access , 2002, CACM.

[36]  Pedro M. Domingos,et al.  Personalizing web sites for mobile users , 2001, WWW '01.

[37]  Yoram Singer,et al.  The Hierarchical Hidden Markov Model: Analysis and Applications , 1998, Machine Learning.

[38]  Eli Upfal,et al.  Web search using automatic classification , 1996, WWW 1996.

[39]  Lawrence R. Rabiner,et al.  A tutorial on Hidden Markov Models , 1986 .

[40]  Michael J. Pazzani.  Adaptive Interfaces for Ubiquitous Web Access , 2003, User Modeling.

[41]  Tom M. Mitchell,et al.  Learning to construct knowledge bases from the World Wide Web , 2000, Artif. Intell..

[42]  Paul P. Maglio,et al.  How to personalize the Web , 1997, CHI.

[43]  Jonathan Trevor,et al.  m-links: An infrastructure for very small internet devices , 2001, MobiCom '01.

[44]  Regina Barzilay,et al.  Information Fusion in the Context of Multi-Document Summarization , 1999, ACL.

[45]  Padhraic Smyth,et al.  Visualization of navigation patterns on a Web site using model-based clustering , 2000, KDD '00.

[46]  Jaideep Srivastava,et al.  Creating adaptive Web sites through usage-based clustering of URLs , 1999, Proceedings 1999 Workshop on Knowledge and Data Engineering Exchange (KDEX'99) (Cat. No.PR00453).

[47]  Monica M. C. Schraefel,et al.  Hunter gatherer: interaction support for the creation and management of within-web-page collections , 2002, WWW.

[48]  Peter Brusilovsky,et al.  From adaptive hypermedia to the adaptive web , 2002, CACM.

[49]  Dunja Mladenic,et al.  Turning {{\sc Yahoo!}}\ into an automatic Web page classifier , 1998 .

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

[51]  Pedro M. Domingos,et al.  On the Optimality of the Simple Bayesian Classifier under Zero-One Loss , 1997, Machine Learning.

[52]  Peter Brusilovsky,et al.  Adaptive Hypermedia , 2001, User Modeling and User-Adapted Interaction.

[53]  Michael J. Pazzani,et al.  Adaptive Web Site Agents , 1999, AGENTS '99.

[54]  Tomasz Imielinski,et al.  Mining association rules between sets of items in large databases , 1993, SIGMOD Conference.

[55]  Ramesh R. Sarukkai,et al.  Link prediction and path analysis using Markov chains , 2000, Comput. Networks.

[56]  Dan Suciu,et al.  Catching the boat with Strudel: experiences with a Web-site management system , 1998, SIGMOD '98.

[57]  Elizabeth D. Liddy,et al.  Advances in Automatic Text Summarization , 2001, Information Retrieval.

[58]  Irving John Good,et al.  The Estimation of Probabilities: An Essay on Modern Bayesian Methods , 1965 .

[59]  Bill N. Schilit,et al.  Digestor: Device-Independent Access to the World Wide Web , 1997, Comput. Networks.

[60]  Juliana Freire,et al.  Automating Web navigation with the WebVCR , 2000, Comput. Networks.

[61]  Oren Etzioni,et al.  Towards adaptive Web sites: Conceptual framework and case study , 1999, Artif. Intell..

[62]  Eric Horvitz,et al.  Utility-Based Abstraction and Categorization , 1993, UAI.

[63]  Lawrence R. Rabiner,et al.  A tutorial on hidden Markov models and selected applications in speech recognition , 1989, Proc. IEEE.

[64]  Oren Etzioni,et al.  Towards adaptive Web sites: Conceptual framework and case study , 2000, Artif. Intell..

[65]  Juliana Freire,et al.  Web Services and Information Delivery for Diverse Environments , 2000, TES.

[66]  Ingrid Zukerman,et al.  Trading Off Granularity against Complexity in Predictive Models for Complex Domains , 2000 .

[67]  Eric A. Brewer,et al.  Reducing WWW Latency and Bandwidth Requirements by Real-Time Distillation , 1996, Comput. Networks.

[68]  Hakon Wium Lie,et al.  Cascading Style Sheets: Designing for the Web , 1997 .

[69]  Peter Brusilovsky,et al.  Methods and techniques of adaptive hypermedia , 1996, User Modeling and User-Adapted Interaction.

[70]  Steven Glassman,et al.  A Caching Relay for the World Wide Web , 1994, Comput. Networks ISDN Syst..

[71]  Andreas Paepcke,et al.  Seeing the whole in parts: text summarization for web browsing on handheld devices , 2001, WWW '01.

[72]  Pedro M. Domingos,et al.  Tree Induction for Probability-Based Ranking , 2003, Machine Learning.

[73]  Julita Vassileva,et al.  Adaptive Hypertext and Hypermedia , 1998, Springer Netherlands.

[74]  Andreas Paepcke,et al.  Accordion summarization for end-game browsing on PDAs and cellular phones , 2001, CHI.

[75]  Eric Horvitz,et al.  Web montage: a dynamic personalized start page , 2002, WWW '02.

[76]  Dunja Mladeni.  Machine learning for better Web browsing , 2000 .

[77]  Jiang Zhu,et al.  Combining Web Usage and Content Mining for More Effective Personalization , 2000 .

[78]  Andrew McCallum,et al.  Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data , 2001, ICML.

[79]  Kaj Grønbæk,et al.  Ariadne: A Java-based Guided Tour System for the World Wide Web , 1998, Comput. Networks.

[80]  Pedro M. Domingos,et al.  Relational Markov models and their application to adaptive web navigation , 2002, KDD.

[81]  DeshpandeMukund,et al.  Selective Markov models for predicting Web page accesses , 2004 .

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

[83]  David Maxwell Chickering,et al.  Learning Bayesian Networks: The Combination of Knowledge and Statistical Data , 1994, Machine Learning.

[84]  Cheukyin Chan.  WOOD - Web-based Object-oriented Desktop , 2001, WWW Posters.

[85]  Chris Stolte,et al.  Rendering effective route maps: improving usability through generalization , 2001, SIGGRAPH.

[86]  AgrawalRakesh,et al.  Mining association rules between sets of items in large databases , 1993 .

[87]  Andreas Paepcke,et al.  Power browser: efficient Web browsing for PDAs , 2000, CHI.

[88]  Michael I. Jordan,et al.  Advances in Neural Information Processing Systems 30 , 1995 .

[89]  Alfred Kobsa,et al.  Generic User Modeling Systems , 2001, User Modeling and User-Adapted Interaction.

[90]  Michael I. Jordan,et al.  Probabilistic Independence Networks for Hidden Markov Probability Models , 1997, Neural Computation.

[91]  Henry Lieberman,et al.  Letizia: An Agent That Assists Web Browsing , 1995, IJCAI.

[92]  Tom M. Mitchell,et al.  Improving Text Classification by Shrinkage in a Hierarchy of Classes , 1998, ICML.

[93]  Michael McGill,et al.  Introduction to Modern Information Retrieval , 1983 .

[94]  Thorsten Joachims,et al.  Web Watcher: A Tour Guide for the World Wide Web , 1997, IJCAI.

[95]  Lise Getoor,et al.  Learning Probabilistic Relational Models , 1999, IJCAI.