Recommender Systems

  1. Hybrid Recommender Systems
    1. Rationale
      1. Combining Strengths of Different Approaches
        1. Addressing Limitations of Single Methods
          1. Improving Recommendation Quality
          2. Hybridization Methods
            1. Weighted Hybridization
              1. Linear Combination of Scores
                1. Dynamic Weight Assignment
                2. Switching Hybridization
                  1. Rule-Based Switching
                    1. Context-Dependent Switching
                    2. Cascade Hybridization
                      1. Sequential Filtering
                        1. Hierarchical Recommendation
                        2. Feature Combination
                          1. Merging Feature Sets
                            1. Unified Feature Space
                            2. Meta-level Hybridization
                              1. Using One Model's Output as Input to Another
                                1. Stacking Approaches
                              2. Common Hybrid Models
                                1. Content-Boosted Collaborative Filtering
                                  1. Collaborative Filtering with Content-Based Features
                                    1. Demographic and Knowledge-Based Hybrids
                                    2. Design Considerations
                                      1. Balancing Different Recommendation Strategies
                                        1. Computational Complexity
                                          1. System Architecture

                                        Previous

                                        4. Collaborative Filtering

                                        Go to top

                                        Next

                                        6. Advanced Recommender Models