Personalization Techniques And Recommender Systems (Series in Machine Perception and Artificial Intelligence ???) (Series in Machine Perception and Artificial ... Perception and Artifical Intelligence)

✍ Scribed by Gulden Uchyigit, Gulden Uchyigit, Matthew Y Ma

• Publisher: World Scientific Publishing Company
• Year: 2008
• Tongue: English
• Leaves: 334
• Series: Series in Machine Perception and Artificial Intelligence ??? Series in Machine Perception and Artificial ... Perception and Artifical Intelligence
• Category: Library

✦ Synopsis

The phenomenal growth of the Internet has resulted in huge amounts of online information, a situation that is overwhelming to the end users. To overcome this problem, personalization technologies have been extensively employed.
The book is the first of its kind, representing research efforts in the diversity of personalization and recommendation techniques. These include user modeling, content, collaborative, hybrid and knowledge-based recommender systems. It presents theoretic research in the context of various applications from mobile information access, marketing and sales and web services, to library and personalized TV recommendation systems.
This volume will serve as a basis to researchers who wish to learn more in the field of recommender systems, and also to those intending to deploy advanced personalization techniques in their systems.
Contents:
User Modeling and Profiling: Personalization-Privacy Tradeoffs in Adaptive Information Access (B Smyth); A Deep Evaluation of Two Cognitive User Models for Personalized Search (F Gasparetti & A Micarelli); Unobtrusive User Modeling for Adaptive Hypermedia (H J Holz et al.); User Modelling Sharing for Adaptive e-Learning and Intelligent Help (K Kabassi et al.); Collaborative Filtering: Experimental Analysis of Multiattribute Utility Collaborative Filtering on a Synthetic Data Set (N Manouselis & C Costopoulou); Efficient Collaborative Filtering in Content-Addressable Spaces (S Berkovsky et al.); Identifying and Analyzing User Model Information from Collaborative Filtering Datasets (J Griffith et al.); Content-Based Systems, Hybrid Systems and Machine Learning Methods: Personalization Strategies and Semantic Reasoning: Working in Tandem in Advanced Recommender Systems (Y Blanco-Fernández et al.); Content Classification and Recommendation Techniques for Viewing Electronic Programming Guide on a Portable Device (J Zhu et al.); User Acceptance of Knowledge-Based Recommenders (A Felfernig et al.); Using Restricted Random Walks for Library Recommendations and Knowledge Space Exploration (M Franke & A Geyer-Schulz); An Experimental Study of Feature Selection Methods for Text Classification (G Uchyigit & K Clark).

✦ Table of Contents

Contents......Page 10
Preface......Page 6
User Modeling and Profiling......Page 12
1.1. Introduction......Page 14
1.2.1. The challenges of mobile information access......Page 16
1.2.1.1. Mobile internet devices......Page 17
1.2.1.2. Browsing versus search on the mobile internet......Page 18
1.2.2. The click-distance problem......Page 19
1.2.3. Personalized navigation......Page 20
1.2.3.1. Pro ling the user......Page 21
1.2.3.2. Personalizing the portal......Page 22
1.2.4. Evaluation......Page 23
1.2.4.1. Click-distance reduction......Page 24
1.2.4.2. Navigation time versus content time......Page 25
1.3.1. The challenges of web search......Page 27
1.3.2. Exploiting repetition and regularity in community- based web search......Page 29
1.3.3. A case-based approach to personalizing web search......Page 30
1.3.4.1. Successful sessions......Page 33
1.3.4.2. Selection positions......Page 35
1.4. Personalization-Privacy: Striking a Balance......Page 36
References......Page 38
BIOGRAPHY......Page 42
2.1. Introduction......Page 44
2.2. Related Work......Page 47
2.3. SAM-based User Modeling Approach......Page 48
2.3.1. SAM: search of associative memory......Page 49
2.3.2. The user modeling approach......Page 50
2.3.2.1. LTS and STS......Page 52
2.3.2.2. Sampling and Recovery......Page 53
2.3.2.3. Learning......Page 54
2.3.2.4. Interaction with Information Sources......Page 55
2.3.3. HAL-based User Modeling Approach......Page 56
2.4.1. Evaluating User Models in Browsing Activities......Page 58
2.4.2. Corpus-based evaluation......Page 59
2.4.3. Precision vs. Number of Topics......Page 61
2.4.4. Precision vs. Extracted Cues......Page 63
2.4.6. Precision vs. Number of Recovery Attempts......Page 64
2.5. Conclusions......Page 65
References......Page 66
BIOGRAPHIES......Page 71
3.1.1. User modeling in adaptive hypermedia......Page 72
3.1.3. Our solution: unobtrusive user modeling......Page 74
3.2. Approach......Page 75
3.2.1. Classi er-independent feature selection......Page 76
3.2.2. Inference design......Page 77
3.3.1. ACUT......Page 79
3.3.3. Feature design......Page 81
3.3.5. Self-organizing maps......Page 82
3.3.6. Revising the features......Page 85
3.4. Discussion......Page 89
References......Page 92
BIOGRAPHIES......Page 95
4.1. Introduction......Page 96
4.2.2. Systems for Intelligent Help in le manipulation and e-mailing......Page 99
4.2.3. Error Diagnosis in three systems of different domains......Page 100
4.3. Common attributes for evaluating alternative actions......Page 101
4.4. Example of a user interacting with three di erent sys- tems......Page 103
4.5. User Modelling based on Web Services......Page 106
4.5.1. UM-Server's Architecture......Page 107
4.5.2. UM-Server's Operation......Page 109
4.6. Multi-Attribute Decision Making on the Server side......Page 111
4.7. Conclusions......Page 113
Appendix A. Multi-Attribute Decision Making......Page 114
References......Page 116
BIOGRAPHIES......Page 118
Collaborative Filtering......Page 120
5.1. Introduction......Page 122
5.2. MAUT Collaborative Filtering......Page 124
5.3. MAUT Algorithms for Collaborative Filtering......Page 127
5.3.1.2. Similarity per evaluation (PG) algorithm......Page 128
5.3.1.3. Similarity per partial utility (PU) algorithm......Page 130
5.3.2. Nonpersonalized algorithms......Page 131
5.4. Case Study and Experimental Analysis......Page 132
5.4.1. Experimental setting......Page 133
5.4.2. Results......Page 135
5.5. Discussion......Page 139
References......Page 141
BIOGRAPHIES......Page 145
6.1. Introduction......Page 146
6.2. Collaborative Filtering......Page 148
6.2.1. Reducing the computational effort required by the CF......Page 149
6.3. Content-Addressable Data Management......Page 150
6.4.1. Mapping user pro les to content-addressable space......Page 153
6.4.2. Heuristic nearest-neighbors search......Page 154
6.4.3. Heuristic completions of user pro les......Page 157
6.5. Experimental Results......Page 158
6.5.1. Scalability of the search......Page 159
6.5.2. Accuracy of the search......Page 161
6.5.3. Inherent clustering......Page 165
6.5.4. Completion heuristics......Page 167
6.6.1. Conclusions......Page 170
6.6.2. Future research......Page 171
References......Page 172
BIOGRAPHIES......Page 175
7.1. Introduction......Page 176
7.2. Related Work......Page 178
7.2.2. Graph-based approaches for recommendation......Page 179
7.2.3. Collaborative ltering as a social network......Page 180
7.3. Methodology......Page 181
7.3.1. Collaborative ltering approach......Page 182
7.3.2. Graph-based representations of the collaborative l- tering space......Page 184
7.3.3. User features......Page 187
7.4.1. User model features......Page 189
7.4.2. Spreading activation......Page 190
7.5.1. User model features......Page 191
7.5.2. Spreading activation......Page 195
References......Page 196
BIOGRAPHIES......Page 199
Content-based Systems, Hybrid Systems and Machine Learn- ing Methods......Page 200
8.1. Introduction......Page 202
8.2. Related Work......Page 204
8.3.1. The TV ontology......Page 207
8.3.2. The User Pro les......Page 209
8.3.2.2. Level of interest of the users......Page 210
8.3.3.1. Content-based phase......Page 212
8.3.3.2. Collaborative ltering phase......Page 214
8.4. An Example......Page 216
8.4.1. A hybrid recommendation by AVATAR......Page 217
8.4.1.1. The content-based strategy in AVATAR......Page 218
8.4.1.2. The collaborative strategy in AVATAR......Page 219
8.5.1. Test algorithms......Page 220
8.5.1.1. Approach based on association rules (Asso-Rules)......Page 221
8.5.1.2. Item-based collaborative ltering approach (Item-CF)......Page 222
8.5.1.3. Semantically enhanced item-based collaborative ltering (Sem-ItemCF)......Page 223
8.5.2. Test data......Page 224
8.5.4. Assessment of experimental results......Page 225
8.6. Final Discussion......Page 228
References......Page 229
BIOGRAPHIES......Page 232
9.1. Introduction......Page 234
9.2. Related Work......Page 236
9.3.1. Overview......Page 239
9.3.2. EPG recommender system......Page 240
9.4.1. Classification problem and design choice......Page 241
9.4.2. Maximum entropy model......Page 243
9.4.4. Domain identification......Page 245
9.4.5. Content classi er for recommendation......Page 246
9.5.2. Experimental database and protocol......Page 247
9.5.3. Evaluation of ME classifier for domain identification......Page 248
9.5.4. Evaluation of content recommendation......Page 250
9.5.5. Preliminary evaluation of domain based recommen- dation......Page 253
9.6. Conclusion......Page 254
References......Page 256
BIOGRAPHIES......Page 258
10.1. Introduction......Page 260
10.2.2. Recommender knowledge base......Page 262
10.2.3. Recommender process de nition......Page 266
10.2.4. Calculating recommendations......Page 268
10.3.1. Example application......Page 271
10.3.2. Experiences from projects......Page 272
10.3.3. Empirical ndings regarding user acceptance......Page 274
10.4. Related Work......Page 281
10.5. Summary and Future Work......Page 282
References......Page 283
BIOGRAPHIES......Page 286
11.1. Motivation......Page 288
11.2. Cluster Algorithms and Recommender Systems......Page 289
11.3. Restricted Random Walk Clustering......Page 291
11.3.1. Library usage data and similarity graphs......Page 292
11.3.2. The restricted random walk stage......Page 293
11.3.3. The cluster construction stage......Page 296
11.4. Giving Recommendations......Page 299
11.5. Results......Page 303
11.6. Updates of the Recommendation Lists......Page 307
References......Page 308
BIOGRAPHIES......Page 312
12.1. Introduction......Page 314
12.3. Text Representation......Page 316
12.4. Feature Selection......Page 317
12.5. Feature Selection for Textual Domains......Page 318
12.6. Feature Scoring Metrics......Page 319
12.6.1. Chi-Squared statistic......Page 320
12.6.4. Mutual information......Page 321
12.6.6. Odds ratio......Page 322
12.6.8. Fisher criterion......Page 323
12.7. Experimental Setting......Page 324
12.9. Results......Page 326
12.10. Summary and Conclusions......Page 327
References......Page 329
BIOGRAPHIES......Page 331
Subject Index......Page 332