Contents 13.4.3 Supporting Explanation, Confidence Trust 441 13. 4.4 Visualisation, Adaptivity, and Partitioned Dynamicity..443 13.4.5 Respecting Multi-cultural Usability Differences 13.5 Evaluating Critiquing: Resources, Methodologies and Criteria.. 445 13.5. 1 Resources methodologies 13.5.2 Evaluation Criteria 13.6 Conclusion/Open Challenges Opportunities References 449 14 Creating More Credible and Persuasive Recommender Systems The Influence of Source Characteristics on Recommender System Evaluations 455 Kyung-Hyan Yoo and Ulrike gretzel 14.1 Introduction 455 14.2 Recommender Systems as Social Actors .456 14.3 Source Credibility 14.3.1 Trustworthiness 14.3.2 Expertise 458 14.3.3 Influences on Source Credibility 14.4 Source Characteristics Studied in Human-Human Interactions.459 14.4.2 Likeability 14.4.3 Symbols of Authority 14.4.4 Styles of Speech 14.4.5 Physical Attractiveness 14.4.6 Humor 14.5 Source Characteristics in Human-Computer Interactions 14.6 Source Characteristics in Human-Recommender System Interactions 463 14.6.2 Input characteristics 14. 6. 3 Process characteristics 14.6.4 Output characteristics .465 14.6.5 Characteristics of embodied agents 14.7 Discussion 14.8 Implications 14.9 Directions for future research 470 References 471 15 Designing and Evaluating Explanations for 479 Nava Tintarey and Judith Masthoff 15.1 Introduction 479 15.2 Guidelines 481 15.3 Explanations in Expert Systems 15.4 Defining goals 15.4.1 Explain How the System Works: Transparency 483
Contents xv 13.4.3 Supporting Explanation, Confidence & Trust. . . . . . . . . . . 441 13.4.4 Visualisation, Adaptivity, and Partitioned Dynamicity . . . 443 13.4.5 Respecting Multi-cultural Usability Differences . . . . . . . . 445 13.5 Evaluating Critiquing: Resources, Methodologies and Criteria . . . . 445 13.5.1 Resources & Methodologies . . . . . . . . . . . . . . . . . . . . . . . . 446 13.5.2 Evaluation Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 446 13.6 Conclusion / Open Challenges & Opportunities . . . . . . . . . . . . . . . . 448 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 449 14 Creating More Credible and Persuasive Recommender Systems: Kyung-Hyan Yoo and Ulrike Gretzel 14.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 455 14.2 Recommender Systems as Social Actors . . . . . . . . . . . . . . . . . . . . . . 456 14.3 Source Credibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 457 14.3.1 Trustworthiness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 458 14.3.2 Expertise. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 458 14.3.3 Influences on Source Credibility . . . . . . . . . . . . . . . . . . . . . 458 14.4 Source Characteristics Studied in Human-Human Interactions . . . . 459 14.4.1 Similarity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 459 14.4.2 Likeability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 460 14.4.3 Symbols of Authority . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 460 14.4.4 Styles of Speech . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 461 14.4.5 Physical Attractiveness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 461 14.4.6 Humor. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 461 14.5 Source Characteristics in Human-Computer Interactions . . . . . . . . . 462 14.6 Source Characteristics in Human-Recommender System Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 463 14.6.1 Recommender system type . . . . . . . . . . . . . . . . . . . . . . . . . . 463 14.6.2 Input characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 464 14.6.3 Process characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 465 14.6.4 Output characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 465 14.6.5 Characteristics of embodied agents . . . . . . . . . . . . . . . . . . . 467 14.7 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 468 14.8 Implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 468 14.9 Directions for future research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 470 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 471 15 Designing and Evaluating Explanations for Recommender Systems 479 Nava Tintarev and Judith Masthoff 15.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 479 15.2 Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 481 15.3 Explanations in Expert Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 481 15.4 Defining Goals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 482 15.4.1 Explain How the System Works: Transparency . . . . . . . . . 483 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Influence of Source Characteristics on Recommender System Evaluations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 455
15.4.2 Allow Users to Tell the System it is Wrong: Scrutability 485 15.4.3 Increase Users'Confidence in the System: Trust 485 15.4.4 Convince Users to Try or Buy: Persuasiveness 15.4.5 Help Users Make Good Decisions: Effectiveness .488 15.4.6 Help Users Make Decisions Faster: Efficiency 15. 4.7 Make the use of the system enjoyable: Satisfaction... 491 15.5 Evaluating the Impact of Explanations on the Recommender System 15.5.1 Accuracy Metrics 493 15.5.2 Learning Rate 493 15.5.3 Coverage 15.5.4 Acceptance 494 15.6 Designing the Presentation and Interaction with Recommendations 15.6.1 Presenting Recommendations 15.6.2 Interacting with the Recommender System 15.7 Explanation Styles 5.7.1 Collaborative-Based Style Explanations 15.7.2 Content-Based Style Explanation 15.7.3 Case-Based Reasoning(CBR) Style Explanations 15.7.4 Knowledge and Utility-Based Style Explanations 15.7.5 Demographic Style Explanations 15.8 Summary and future directions References 507 16 Usability Guidelines for Product Recommenders Based on Example iquing Researc 511 Pearl Pu, Boi Faltings, Li Chen, Jiyong Zhang and Paolo via 16.1 Introduction 16.2 Preliminaries 513 16.21 Interaction model .513 16.2.2 Utility-Based Recommenders 16.2.3 The Accuracy, Confidence, Effort Framework 517 16.2. 4 Organization of this Cha 518 16. 3 Related Work 518 16.3.1 Types of Recommenders 518 16.3.2 Rating-based Systems 16.3.3 Case-based Systems 519 16.3.4 Utility-based Systems 519 16.3.5 Critiquing-based Systems 16.3.6 Other Design Guidelines 520 16.5 Stimulating Preference Expression with Examples 16.5.1 How Many Examples to Show 527 16.5.2 What Examples to Show 16.6 Preference Revision 530
xvi Contents 15.4.2 Allow Users to Tell the System it is Wrong: Scrutability 485 15.4.3 Increase Users’ Confidence in the System: Trust . . . . . . . . 485 15.4.4 Convince Users to Try or Buy: Persuasiveness . . . . . . . . . 487 15.4.5 Help Users Make Good Decisions: Effectiveness . . . . . . . 488 15.4.6 Help Users Make Decisions Faster: Efficiency . . . . . . . . . 490 15.4.7 Make the use of the system enjoyable: Satisfaction . . . . . . 491 15.5 Evaluating the Impact of Explanations on the Recommender System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 492 15.5.1 Accuracy Metrics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 493 15.5.2 Learning Rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 493 15.5.3 Coverage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 494 15.5.4 Acceptance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 494 15.6 Designing the Presentation and Interaction with Recommendations 495 15.6.1 Presenting Recommendations . . . . . . . . . . . . . . . . . . . . . . . 495 15.6.2 Interacting with the Recommender System . . . . . . . . . . . . 496 15.7 Explanation Styles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 497 15.7.1 Collaborative-Based Style Explanations . . . . . . . . . . . . . . . 500 15.7.2 Content-Based Style Explanation . . . . . . . . . . . . . . . . . . . . 501 15.7.3 Case-Based Reasoning (CBR) Style Explanations . . . . . . 503 15.7.4 Knowledge and Utility-Based Style Explanations . . . . . . . 504 15.7.5 Demographic Style Explanations. . . . . . . . . . . . . . . . . . . . . 505 15.8 Summary and future directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 505 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 507 16 Usability Guidelines for Product Recommenders Based on Example Critiquing Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 511 Pearl Pu, Boi Faltings, Li Chen, Jiyong Zhang and Paolo Viappiani 16.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 512 16.2 Preliminaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 513 16.2.1 Interaction Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 513 16.2.2 Utility-Based Recommenders . . . . . . . . . . . . . . . . . . . . . . . 515 16.2.3 The Accuracy, Confidence, Effort Framework . . . . . . . . . . 517 16.2.4 Organization of this Chapter . . . . . . . . . . . . . . . . . . . . . . . . 518 16.3 Related Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 518 16.3.1 Types of Recommenders . . . . . . . . . . . . . . . . . . . . . . . . . . . 518 16.3.2 Rating-based Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 519 16.3.3 Case-based Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 519 16.3.4 Utility-based Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 519 16.3.5 Critiquing-based Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . 520 16.3.6 Other Design Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . 520 16.4 Initial Preference Elicitation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 521 16.5 Stimulating Preference Expression with Examples . . . . . . . . . . . . . . 525 16.5.1 How Many Examples to Show . . . . . . . . . . . . . . . . . . . . . . . 527 16.5.2 What Examples to Show . . . . . . . . . . . . . . . . . . . . . . . . . . . . 527 16.6 Preference Revision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 530 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Contents 16.6.1 Preference Conflicts and Partial Satisfaction 16.6. 2 Tradeoff Assistance 532 16.7 Display Strategies 534 16.7.1 Recommending one item at a Time 167.2 Recommending k best items 535 16.7.3 Explanation Interfaces 536 16.8 A Model for Rationalizing the guidelines 537 16.9 Conclusion References 17 Map Based Visualization of Product Catalogs Martijn Kagie, Michiel van Wezel and Patrick J F. Groenen 17.1 Introduction 17.2 Methods for Map Based Visualization 17. 2.1 Self-Organizing Maps 550 17.2.2 Treemaps 17. 2.3 Multidimensional Scaling 553 17.2.4 Nonlinear Principal Components Analysis 553 17.3 Product Catalog Maps 17.3.1 Multidimensional Scaling 555 17.4 Determining Attribute Weights using Clickstream Analysis 17.4.1 Poisson Regression Model 17.4.2 Handling Missing values 560 17.4.3 Choosing Weights Using Poisson Regression 17.4.4 Stepwise Poisson Regression Model 17.5 Graphical Shopping Interface 17.6 E-Commerce Applications 17.6.1 MDS Based Product Catalog Map Using Attribute 17.6.2 NL-PCA Based Product Catalog Map 17.6.3 Graphical Shopping Interface 570 17.7 Conclusions and Outlook References 574 Part IV Recommender Systems and Communities 18 Communities, Collaboration, and recommender Systems in Personalized Web search 579 Barry Smyth, Maurice Coyle and Peter Briggs 18.1 Introduction 18.2 A Brief History of Web Search 18. 3 The Future of Web search 18.3.1 Personalized Web Search 18.3.2 Collaborative Information retrieval 18.3.3 Towards Social Search 590
Contents xvii 16.6.1 Preference Conflicts and Partial Satisfaction . . . . . . . . . . . 531 16.6.2 Tradeoff Assistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 532 16.7 Display Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 534 16.7.1 Recommending One Item at a Time . . . . . . . . . . . . . . . . . . 534 16.7.2 Recommending K best Items . . . . . . . . . . . . . . . . . . . . . . . . 535 16.7.3 Explanation Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 536 16.8 A Model for Rationalizing the Guidelines . . . . . . . . . . . . . . . . . . . . . 537 16.9 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 541 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 541 17 Map Based Visualization of Product Catalogs . . . . . . . . . . . . . . . . . . . . 547 Martijn Kagie, Michiel van Wezel and Patrick J.F. Groenen 17.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 547 17.2 Methods for Map Based Visualization . . . . . . . . . . . . . . . . . . . . . . . . 549 17.2.1 Self-Organizing Maps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 550 17.2.2 Treemaps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 551 17.2.3 Multidimensional Scaling . . . . . . . . . . . . . . . . . . . . . . . . . . . 553 17.2.4 Nonlinear Principal Components Analysis . . . . . . . . . . . . . 553 17.3 Product Catalog Maps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 554 17.3.1 Multidimensional Scaling . . . . . . . . . . . . . . . . . . . . . . . . . . . 555 17.3.2 Nonlinear Principal Components Analysis . . . . . . . . . . . . . 558 17.4 Determining Attribute Weights using Clickstream Analysis . . . . . . 559 17.4.1 Poisson Regression Model . . . . . . . . . . . . . . . . . . . . . . . . . . 560 17.4.2 Handling Missing Values . . . . . . . . . . . . . . . . . . . . . . . . . . . 560 17.4.3 Choosing Weights Using Poisson Regression . . . . . . . . . . 561 17.4.4 Stepwise Poisson Regression Model . . . . . . . . . . . . . . . . . . 562 17.5 Graphical Shopping Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 562 17.6 E-Commerce Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 563 17.6.1 MDS Based Product Catalog Map Using Attribute Weights . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 564 17.6.2 NL-PCA Based Product Catalog Map. . . . . . . . . . . . . . . . . 568 17.6.3 Graphical Shopping Interface . . . . . . . . . . . . . . . . . . . . . . . 570 17.7 Conclusions and Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 573 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 574 Part IV Recommender Systems and Communities 18 Communities, Collaboration, and Recommender Systems in Personalized Web Search . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 579 Barry Smyth, Maurice Coyle and Peter Briggs 18.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 579 18.2 A Brief History of Web Search . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 581 18.3 The Future of Web Search . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 583 18.3.1 Personalized Web Search . . . . . . . . . . . . . . . . . . . . . . . . . . . 584 18.3.2 Collaborative Information Retrieval . . . . . . . . . . . . . . . . . . 588 18.3.3 Towards Social Search . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 590
18.4 Case-Study 1-Community-Based Web Search 18.4.1 Repetition and Regularity in Search Communities 592 18.4.2 The Collaborative Web Search System 18.43 Evaluation 18.44 Discussion 18.5 Case-Study 2-Web Search Shared 598 18.5.1 The Hey Staks System 18.5.2 The Hey Staks Recomendation Engine 18.5.3 18.54 Discussion 18.6 Conclusions 607 References 19 Social Tagging Recommender Systems 615 Leandro Balby Marinho, Alexandros Nanopoulos, Lars Schmidt Thieme. Robert Jaschke Andreas Hotho. Gerd Stumme and 19.1 Introduction 616 19.2 Social Tagging Recommenders Systems 617 19.2.1 Folksonomy 618 19.2.2 The Traditional Recommender Systems Paradigm 19.2.3 Multi-mode Recommendations 19.3 Real World Social Tagging Recommender Systems What are the Challenges? 19.3.2 BibSonomy as Study Case 19.3.3 Tag Acquisition 19.4 Recommendation Algorithms for Social Tagging Systems 626 19.41 Collaborative filterin 194.2 Recommendation based on rankin 630 19.4.3 Content-Based Social Tagging RS 19.4. 4 Evaluation Protocols and Metrics 19.5 Comparison of Algorithms 639 19.6 References 20 Trust and recommendations Patricia Victor. Martine De Cock and Chris Cornelis 20.1 Introduction 645 20.2 Computational Trust 647 0.2.2 Trust Computation 20.3 Trust-Enhanced Recommender Systems 655 20.3.2 State of the Art 658 20.4 Recent Developments and Open Challenges 670
xviii Contents 18.4 Case-Study 1 - Community-Based Web Search . . . . . . . . . . . . . . . . 591 18.4.1 Repetition and Regularity in Search Communities . . . . . . 592 18.4.2 The Collaborative Web Search System . . . . . . . . . . . . . . . . 593 18.4.3 Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 596 18.4.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 598 18.5 Case-Study 2 - Web Search. Shared.. . . . . . . . . . . . . . . . . . . . . . . . . . 598 18.5.1 The HeyStaks System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 599 18.5.2 The HeyStaks Recomendation Engine . . . . . . . . . . . . . . . . 602 18.5.3 Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 604 18.5.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 607 18.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 607 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 609 19 Social Tagging Recommender Systems . . . . . . . . . . . . . . . . . . . . . . . . . . 615 Leandro Balby Marinho, Alexandros Nanopoulos, Lars SchmidtThieme, Robert Ja¨schke, Andreas Hotho, Gerd Stumme and Panagiotis Symeonidis 19.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 616 19.2 Social Tagging Recommenders Systems . . . . . . . . . . . . . . . . . . . . . . 617 19.2.1 Folksonomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 618 19.2.2 The Traditional Recommender Systems Paradigm . . . . . . 619 19.2.3 Multi-mode Recommendations . . . . . . . . . . . . . . . . . . . . . . 620 19.3 Real World Social Tagging Recommender Systems . . . . . . . . . . . . . 621 19.3.1 What are the Challenges? . . . . . . . . . . . . . . . . . . . . . . . . . . . 621 19.3.2 BibSonomy as Study Case . . . . . . . . . . . . . . . . . . . . . . . . . . 622 19.3.3 Tag Acquisition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 624 19.4 Recommendation Algorithms for Social Tagging Systems . . . . . . . 626 19.4.1 Collaborative Filtering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 626 19.4.2 Recommendation based on Ranking . . . . . . . . . . . . . . . . . . 630 19.4.3 Content-Based Social Tagging RS . . . . . . . . . . . . . . . . . . . . 634 19.4.4 Evaluation Protocols and Metrics . . . . . . . . . . . . . . . . . . . . 637 19.5 Comparison of Algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 639 19.6 Conclusions and Research Directions . . . . . . . . . . . . . . . . . . . . . . . . . 640 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 642 20 Trust and Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 645 Patricia Victor, Martine De Cock, and Chris Cornelis 20.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 645 20.2 Computational Trust . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 647 20.2.1 Trust Representation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 648 20.2.2 Trust Computation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 650 20.3 Trust-Enhanced Recommender Systems . . . . . . . . . . . . . . . . . . . . . . 655 20.3.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 656 20.3.2 State of the Art. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 658 20.3.3 Empirical Comparison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 664 20.4 Recent Developments and Open Challenges . . . . . . . . . . . . . . . . . . . 670
Contents 0.5 Conclusions 672 References 21 Group Recommender Systems: Combining Individual Models 677 Judith masthoff 1.1 Introduction 21.2 Usage Scenarios and Classification of Group Recommenders .. 679 21.2.1 Interactive Television 21.2.2 Ambient Intelligence 21.2.3 Scenarios Underlying Related Work 680 21.2. 4 A Classification of Group Recor 21.3 Aggregation Strategies 21.3.1 Overview of Aggregation Strategies 21.3.2 Aggregation Strategies Used in Related Work 683 21.3.3 Which Strategy Performs Best 21.4 Impact of Sequence Order 21.5 Modelling Affective State 21.5. 1 Modelling an Individual,s Satisfaction on its Own 689 21.5.2 Effects of the Group on an Individuals Satisfaction.. 690 21.6 Using Affective State inside Aggregation Strategies 691 21.7 Applying Group Recommendation to Individual Users 693 21.7.1 Multiple Criteria 21.7.2 Cold-Start Problem 21.7.3 Virtual Group Members 21.8 Conclusions and Challenges 697 21.8.1 Main Issues raised 697 21.8.2 Caveat: Group Modelling 698 21.8.3 Challenges References Part V Advanced algorithms 22 Aggregation of Preferences in Recommender Systems 705 Gleb Beliakov. Tomasa Calvo and simon James 22.1 Introduction 705 22.2 Types of Aggregation in Recommender Systems 706 I Aggregation of Preferences in CF 708 22.2.2 Aggregation of Features in CB and UB Recommendation 708 22.23 Profile Construction for CB. UB 22.2.4 Item and User Similarity and Neighborhood Formation. 709 22.25 Connectives in Case-Based Reasoning for rs 22.2.6 Weighted Hybrid Systems 711 22.3 Review of Aggregation Functions 712 22.3.1 Definitions and Properties 22.3.2 Aggregation Families 716 22.4 Construction of Aggregation Functions
Contents xix 20.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 672 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 672 21 Group Recommender Systems: Combining Individual Models. . . . . . 677 Judith Masthoff 21.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 677 21.2 Usage Scenarios and Classification of Group Recommenders . . . . . 679 21.2.1 Interactive Television . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 679 21.2.2 Ambient Intelligence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 679 21.2.3 Scenarios Underlying Related Work . . . . . . . . . . . . . . . . . . 680 21.2.4 A Classification of Group Recommenders . . . . . . . . . . . . . 681 21.3 Aggregation Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 682 21.3.1 Overview of Aggregation Strategies . . . . . . . . . . . . . . . . . . 682 21.3.2 Aggregation Strategies Used in Related Work . . . . . . . . . . 683 21.3.3 Which Strategy Performs Best . . . . . . . . . . . . . . . . . . . . . . . 685 21.4 Impact of Sequence Order . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 686 21.5 Modelling Affective State . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 688 21.5.1 Modelling an Individual’s Satisfaction on its Own . . . . . . 689 21.5.2 Effects of the Group on an Individual’s Satisfaction . . . . . 690 21.6 Using Affective State inside Aggregation Strategies. . . . . . . . . . . . . 691 21.7 Applying Group Recommendation to Individual Users . . . . . . . . . . 693 21.7.1 Multiple Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 693 21.7.2 Cold-Start Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 695 21.7.3 Virtual Group Members . . . . . . . . . . . . . . . . . . . . . . . . . . . . 697 21.8 Conclusions and Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 697 21.8.1 Main Issues Raised . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 697 21.8.2 Caveat: Group Modelling . . . . . . . . . . . . . . . . . . . . . . . . . . . 698 21.8.3 Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 698 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 701 Part V Advanced Algorithms 22 Aggregation of Preferences in Recommender Systems . . . . . . . . . . . . . 705 Gleb Beliakov, Tomasa Calvo and Simon James 22.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 705 22.2 Types of Aggregation in Recommender Systems . . . . . . . . . . . . . . . 706 22.2.1 Aggregation of Preferences in CF . . . . . . . . . . . . . . . . . . . . 708 22.2.2 Aggregation of Features in CB and UB Recommendation 708 22.2.3 Profile Construction for CB, UB . . . . . . . . . . . . . . . . . . . . . 709 22.2.4 Item and User Similarity and Neighborhood Formation . . 709 22.2.5 Connectives in Case-Based Reasoning for RS . . . . . . . . . . 711 22.2.6 Weighted Hybrid Systems . . . . . . . . . . . . . . . . . . . . . . . . . . 711 22.3 Review of Aggregation Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 712 22.3.1 Definitions and Properties . . . . . . . . . . . . . . . . . . . . . . . . . . 712 22.3.2 Aggregation Families . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 716 22.4 Construction of Aggregation Functions . . . . . . . . . . . . . . . . . . . . . . . 722 . . . . . . . . . . . . . . . . . . . . . . . . . . . . .