Acknowledgements
Abstract
I Setting the Scene
1 Introduction
1.1 Motivation and Thesis
1.2 Hypothesis-Based Collaborative Filtering in a Nutshell
1.3 Thesis Statement
1.3.1 Research Hypotheses
1.3.2 Research Goals
1.4 Contributions
1.5 Organization
2 Related Work
2.1 Recommender Systems
2.1.1 Formal Framework
2.1.2 Ratings
2.2 Collaborative Filtering
2.2.1 General Framework for Collaborative Filtering
2.2.2 Cold-Start Problem
2.3 Machine Learning
II Preference Modeling
3 Conceptualization and Specification of Preferences
3.1 Formalization of Preferences
3.1.1 PartialPreferences
3.2 Partial Preference Extraction from Machine Learning Models
3.2.1 Partial Preference Extraction from Decision Tree Classifier
3.2.2 Partial Preference Extraction from Naïve Bayesian Classifier
3.3 Ontological Specification of Hypothesized Preferences
3.4 Acceptance of Hypotheses
3.5 Summary
4 Domain Ontology-Boosted Decision Tree Induction
4.1 Decision Tree Induction
4.1.1 Feature Selection
4.2 SEMTREE Extension to the Decision Tree Model
4.2.1 Basic Idea
4.2.2 Injecting Concept Features to Generalize from Features
4.2.3 Classification
4.2.4 Implementation
4.3 Acceptance of Hypotheses
4.4 Summary
III Preference Similarity
5 Hypothesized Preference Similarity
5.1 Theoretical Foundation of Hypothesized Preference Similarity
5.1.1 Hypothesized Partial Preference Similarity
5.1.2 Hypothesized Semi-Partial Preference Similarity
5.2 Hypothesized Utility-Based Preference Similarity
5.2.1 Product Set for Utility Prediction
5.2.2 Correlative Predicted Utility-Based Similarity
5.2.3 Probabilistic Predicted Utility-Based Similarity
5.2.4 Probabilistic Predicted Utility-Based Semi-Partial Similarity
5.3 Hypothesis Composition-Based Preference Similarity
5.3.1 Similarity of Hypothesized Partial Preferences
5.3.2 Similarity Computation Based on Partial Preference Similarity Matrix
5.4 Summary
IV Evaluation
6 Evaluation
6.1 Experimental Setting
6.1.1 Performance Metrics
6.2 Candidates for Comparison
6.2.1 Hypothesis-Based Collaborative Filtering Candidates
6.2.2 Baseline Collaborative Filtering Candidates
6.2.3 Baseline Content Filtering Candidates
6.3 Dataset
6.4 Results and Discussion
6.4.1 Rating Prediction Accuracy
6.4.2 Relevance Filtering Quality
6.5 Information Theoretic Reflection of Hypothesized Preferences versus Product Ratings
6.6 Acceptance of Hypotheses
6.7 Summary
7 Analysis
7.1 Method
7.1.1 Grounded Theory
7.1.2 Data Collection
7.1.3 Data Analysis
7.2 Theory Development
7.2.1 TheoryConcepts
7.2.2 Comparison of Recommendation Performance
7.3 Theory Consolidation
7.4 Theory Validation
7.4.1 Experimental Setting
7.4.2 Results and Discussion
7.5 Acceptance of Hypotheses
V Closing
8 Limitations
8.1 Conceptual Limitations
8.2 Technical Limitations
9 Conclusions
9.1 Acceptance of Hypotheses
9.2 Achievements of Research Goals and Thesis
9.3 Opportunities for Future Research
VI Appendix
A Tools
A.1 RECOMIZER
A.2 OMORE
A.2.1 Architecture
A.3 MOLookup
A.4 LiMo Database
A.4.1 Interlinking Movies across Web Pages
B Movie Ontology MO
C MovieLens Dataset
C.1 Genres of MovieLens
C.2 Sparse MovieLens Dataset
D Distribution of Recommendation Performance
E Comparison Between Properties and Recommendation Performance
F Comparison Between Recomm. Perform. regarding Cold-Start Behavior
G Publications
Bibliography