Recommender_systems_handbook.pdf - Francesco Ricci Lior Rokach Bracha Shapira Paul B Kantor Editors Recommender Systems Handbook 123 Editors Francesco

Recommender_systems_handbook.pdf - Francesco Ricci Lior...

This preview shows page 1 out of 845 pages.

You've reached the end of your free preview.

Want to read all 845 pages?

Unformatted text preview: Francesco Ricci · Lior Rokach · Bracha Shapira · Paul B. Kantor Editors Recommender Systems Handbook 123 Editors Francesco Ricci Free University of Bozen-Bolzano Faculty of Computer Science Piazza Domenicani 3 39100 Bolzano Italy [email protected] Bracha Shapira Ben-Gurion University of the Negev Dept. Information Systems Engineering Beer-Sheva Israel [email protected] Lior Rokach Ben-Gurion University of the Negev Dept. Information Systems Engineering 84105 Beer-Sheva Israel [email protected] Paul B. Kantor Rutgers University School of Communication, Information & Library Studies Huntington Street 4 08901-1071 New Brunswick New Jersey SCILS Bldg. USA [email protected] ISBN 978-0-387-85819-7 e-ISBN 978-0-387-85820-3 DOI 10.1007/978-0-387-85820-3 Springer New York Dordrecht Heidelberg London Library of Congress Control Number: 2010937590 c Springer Science+Business Media, LLC 2011  All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. Printed on acid-free paper Springer is part of Springer Science+Business Media ( ) Preface Recommender Systems are software tools and techniques providing suggestions for items to be of use to a user. The suggestions provided are aimed at supporting their users in various decision-making processes, such as what items to buy, what music to listen, or what news to read. Recommender systems have proven to be valuable means for online users to cope with the information overload and have become one of the most powerful and popular tools in electronic commerce. Correspondingly, various techniques for recommendation generation have been proposed and during the last decade, many of them have also been successfully deployed in commercial environments. Development of recommender systems is a multi-disciplinary effort which involves experts from various fields such as Artificial intelligence, Human Computer Interaction, Information Technology, Data Mining, Statistics, Adaptive User Interfaces, Decision Support Systems, Marketing, or Consumer Behavior. Recommender Systems Handbook: A Complete Guide for Research Scientists and Practitioners aims to impose a degree of order upon this diversity by presenting a coherent and unified repository of recommender systems’ major concepts, theories, methodologies, trends, challenges and applications. This is the first comprehensive book which is dedicated entirely to the field of recommender systems and covers several aspects of the major techniques. Its informative, factual pages will provide researchers, students and practitioners in industry with a comprehensive, yet concise and convenient reference source to recommender systems. The book describes in detail the classical methods, as well as extensions and novel approaches that were recently introduced. The book consists of five parts: techniques, applications and evaluation of recommender systems, interacting with recommender systems, recommender systems and communities, and advanced algorithms. The first part presents the most popular and fundamental techniques used nowadays for building recommender systems, such as collaborative filtering, content-based filtering, data mining methods and context-aware methods. The second part starts by surveying techniques and approaches that have been used to evaluate the quality of the recommendations. Then deals with the practical aspects of designing recommender systems, it describes design and implementation consideration, setting guidelines for the selection of the vii viii Preface more suitable algorithms. The section continues considering aspects that may affect the design and finally, it discusses methods, challenges and measures to be applied for the evaluation of the developed systems. The third part includes papers dealing with a number of issues related to the presentation, browsing, explanation and visualization of the recommendations, and techniques that make the recommendation process more structured and conversational. The fourth part is fully dedicated to a rather new topic, which is however rooted in the core idea of a collaborative recommender, i.e., exploiting user generated content of various types to build new types and more credible recommendations. Finally the last section collects a few papers on some advanced topics, such as the exploitation of active learning principles to guide the acquisition of new knowledge, techniques suitable for making a recommender system robust against attacks of malicious users, and recommender systems that aggregate multiple types of user feedbacks and preferences to build more reliable recommendations. We would like to thank all authors for their valuable contributions. We would like to express gratitude for all reviewers that generously gave comments on drafts or counsel otherwise. We would like to express our special thanks to Susan LagerstromFife and staff members of Springer for their kind cooperation throughout the production of this book. Finally, we wish this handbook will contribute to the growth of this subject, we wish to the novices a fruitful learning path, and to those more experts a compelling application of the ideas discussed in this handbook and a fruitful development of this challenging research area. May 2010 Francesco Ricci Lior Rokach Bracha Shapira Paul B. Kantor Contents 1 Introduction to Recommender Systems Handbook . . . . . . . . . . . . . . . . Francesco Ricci, Lior Rokach and Bracha Shapira 1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 Recommender Systems Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3 Data and Knowledge Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.4 Recommendation Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.5 Application and Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.6 Recommender Systems and Human Computer Interaction . . . . . . . 1.6.1 Trust, Explanations and Persuasiveness . . . . . . . . . . . . . . . 1.6.2 Conversational Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.6.3 Visualization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.7 Recommender Systems as a Multi-Disciplinary Field . . . . . . . . . . . 1.8 Emerging Topics and Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.8.1 Emerging Topics Discussed in the Handbook . . . . . . . . . . 1.8.2 Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 4 7 10 14 17 18 19 21 21 23 23 26 29 Part I Basic Techniques 2 Data Mining Methods for Recommender Systems . . . . . . . . . . . . . . . . Xavier Amatriain, Alejandro Jaimes, Nuria Oliver, and Josep M. Pujol 2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Data Preprocessing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.1 Similarity Measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.2 Sampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.3 Reducing Dimensionality . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.4 Denoising . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.1 Nearest Neighbors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.2 Decision Trees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.3 Ruled-based Classifiers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 39 40 41 42 44 47 48 48 50 51 ix x Contents 2.3.4 Bayesian Classifiers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.5 Artificial Neural Networks . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.6 Support Vector Machines . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.7 Ensembles of Classifiers . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.8 Evaluating Classifiers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Cluster Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.1 k-Means . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.2 Alternatives to k-means . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5 Association Rule Mining . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 54 56 58 59 61 62 63 64 66 67 3 Content-based Recommender Systems: State of the Art and Trends . 73 Pasquale Lops, Marco de Gemmis and Giovanni Semeraro 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 3.2 Basics of Content-based Recommender Systems . . . . . . . . . . . . . . . 75 3.2.1 A High Level Architecture of Content-based Systems . . . 75 3.2.2 Advantages and Drawbacks of Content-based Filtering . . 78 3.3 State of the Art of Content-based Recommender Systems . . . . . . . . 79 3.3.1 Item Representation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 3.3.2 Methods for Learning User Profiles . . . . . . . . . . . . . . . . . . 90 3.4 Trends and Future Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 3.4.1 The Role of User Generated Content in the Recommendation Process . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 3.4.2 Beyond Over-specializion: Serendipity . . . . . . . . . . . . . . . . 96 3.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 4 A Comprehensive Survey of Neighborhood-based Recommendation Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 Christian Desrosiers and George Karypis 4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 4.1.1 Formal Definition of the Problem . . . . . . . . . . . . . . . . . . . . 108 4.1.2 Overview of Recommendation Approaches . . . . . . . . . . . . 110 4.1.3 Advantages of Neighborhood Approaches . . . . . . . . . . . . . 112 4.1.4 Objectives and Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 4.2 Neighborhood-based Recommendation . . . . . . . . . . . . . . . . . . . . . . . 114 4.2.1 User-based Rating Prediction . . . . . . . . . . . . . . . . . . . . . . . . 115 4.2.2 User-based Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . 116 4.2.3 Regression VS Classification . . . . . . . . . . . . . . . . . . . . . . . . 117 4.2.4 Item-based Recommendation . . . . . . . . . . . . . . . . . . . . . . . . 117 4.2.5 User-based VS Item-based Recommendation . . . . . . . . . . 118 4.3 Components of Neighborhood Methods . . . . . . . . . . . . . . . . . . . . . . . 120 4.3.1 Rating Normalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 4.3.2 Similarity Weight Computation . . . . . . . . . . . . . . . . . . . . . . 124 4.3.3 Neighborhood Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 Contents xi 4.4 Advanced Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131 4.4.1 Dimensionality Reduction Methods . . . . . . . . . . . . . . . . . . 132 4.4.2 Graph-based Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 4.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140 5 Advances in Collaborative Filtering . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 Yehuda Koren and Robert Bell 5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 5.2 Preliminaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147 5.2.1 Baseline predictors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148 5.2.2 The Netflix data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 5.2.3 Implicit feedback . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150 5.3 Matrix factorization models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 5.3.1 SVD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 5.3.2 SVD++ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 5.3.3 Time-aware factor model . . . . . . . . . . . . . . . . . . . . . . . . . . . 154 5.3.4 Comparison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159 5.3.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160 5.4 Neighborhood models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 5.4.1 Similarity measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162 5.4.2 Similarity-based interpolation . . . . . . . . . . . . . . . . . . . . . . . 163 5.4.3 Jointly derived interpolation weights . . . . . . . . . . . . . . . . . 165 5.4.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168 5.5 Enriching neighborhood models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168 5.5.1 A global neighborhood model . . . . . . . . . . . . . . . . . . . . . . . 169 5.5.2 A factorized neighborhood model . . . . . . . . . . . . . . . . . . . . 173 5.5.3 Temporal dynamics at neighborhood models . . . . . . . . . . . 180 5.5.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182 5.6 Between neighborhood and factorization . . . . . . . . . . . . . . . . . . . . . . 182 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184 6 Developing Constraint-based Recommenders . . . . . . . . . . . . . . . . . . . . 187 Alexander Felfernig, Gerhard Friedrich, Dietmar Jannach and Markus Zanker 6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187 6.2 Development of Recommender Knowledge Bases . . . . . . . . . . . . . . 191 6.3 User Guidance in Recommendation Processes . . . . . . . . . . . . . . . . . 194 6.4 Calculating Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203 6.5 Experiences from Projects and Case Studies . . . . . . . . . . . . . . . . . . . 205 6.6 Future Research Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207 6.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212 xii 7 Contents Context-Aware Recommender Systems . . . . . . . . . . . . . . . . . . . . . . . . . 217 Gediminas Adomavicius and Alexander Tuzhilin 7.1 Introduction and Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218 7.2 Context in Recommender Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . 219 7.2.1 What is Context? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219 7.2.2 Modeling Contextual Information in Recommender Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223 7.2.3 Obtaining Contextual Information . . . . . . . . . . . . . . . . . . . . 228 7.3 Paradigms for Incorporating Context in Recommender Systems . . 230 7.3.1 Contextual Pre-Filtering . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233 7.3.2 Contextual Post-Filtering . . . . . . . . . . . . . . . . . . . . . . . . . . . 237 7.3.3 Contextual Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238 7.4 Combining Multiple Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243 7.4.1 Case Study of Combining Multiple Pre-Filters: Algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244 7.4.2 Case Study of Combining Multiple Pre-Filters: Experimental Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245 7.5 Additional Issues in Context-Aware Recommender Systems . . . . . 247 7.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250 Part II Applications and Evaluation of RSs 8 Evaluating Recommendation Systems . . . . . . . . . . . . . . . . . . . . . . . . . . 257 Guy Shani and Asela Gunawardana 8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258 8.2 Experimental Settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260 8.2.1 Offline Experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261 8.2.2 User Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263 8.2.3 Online Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266 8.2.4 Drawing Reliable Conclusions . . . . . . . . . . . . . . . . . . . . . . . 267 8.3 Recommendation System Properties . . . . . . . . . . . . . . . . . . . . . . . . . 271 8.3.1 User Preference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272 8.3.2 Prediction Accuracy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273 8.3.3 Coverage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281 8.3.4 Confidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283 8.3.5 Trust . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285 8.3.6 Novelty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285 8.3.7 Serendipity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286 8.3.8 Diversity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288 8.3.9 Utility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289 8.3.10 Risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290 8.3.11 Robustness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290 8.3.12 Privacy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291 8.3.13 Adaptivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292 Contents xiii 8.3....
View Full Document

  • Fall '05
  • Elizabeth Thomas
  • Cold start, Recommender Systems, Recommender system, Implicit data collection

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture