Hybrid recommendation based on low-dimensional augmentation of combined feature profiles

Authors:
Andrzej Szwabe;Tadeusz Janasiewicz;Michal Ciesielczyk
Affiliations:
Institute of Control and Information Engineering, Poznan University of Technology, Poznan, Poland;Institute of Control and Information Engineering, Poznan University of Technology, Poznan, Poland;Institute of Control and Information Engineering, Poznan University of Technology, Poznan, Poland
Venue:
ICCCI'11 Proceedings of the Third international conference on Computational collective intelligence: technologies and applications - Volume Part II
Year:
2011

Citing 20
Cited 0

Using linear algebra for intelligent information retrieval

SIAM Review
Recommendation as classification: using social and content-based information in recommendation

AAAI '98/IAAI '98 Proceedings of the fifteenth national/tenth conference on Artificial intelligence/Innovative applications of artificial intelligence
Analysis of recommendation algorithms for e-commerce

Proceedings of the 2nd ACM conference on Electronic commerce
Information Retrieval

Information Retrieval
Methods and metrics for cold-start recommendations

SIGIR '02 Proceedings of the 25th annual international ACM SIGIR conference on Research and development in information retrieval
Hybrid Recommender Systems: Survey and Experiments

User Modeling and User-Adapted Interaction
E-Commerce Recommendation Applications

Data Mining and Knowledge Discovery
Content-boosted collaborative filtering for improved recommendations

Eighteenth national conference on Artificial intelligence
Evaluating collaborative filtering recommender systems

ACM Transactions on Information Systems (TOIS)
Toward the Next Generation of Recommender Systems: A Survey of the State-of-the-Art and Possible Extensions

IEEE Transactions on Knowledge and Data Engineering
CinemaScreen Recommender Agent: Combining Collaborative and Content-Based Filtering

IEEE Intelligent Systems
Collaborative recommender systems: Combining effectiveness and efficiency

Expert Systems with Applications: An International Journal
Introduction to Information Retrieval

Introduction to Information Retrieval
Feature-Weighted User Model for Recommender Systems

UM '07 Proceedings of the 11th international conference on User Modeling
Recommending new movies: even a few ratings are more valuable than metadata

Proceedings of the third ACM conference on Recommender systems
A unified approach to building hybrid recommender systems

Proceedings of the third ACM conference on Recommender systems
Hybrid web recommender systems

The adaptive web
Combining content-based and collaborative recommendations: A hybrid approach based on Bayesian networks

International Journal of Approximate Reasoning
Performance of recommender algorithms on top-n recommendation tasks

Proceedings of the fourth ACM conference on Recommender systems
Alleviating the sparsity problem of collaborative filtering using trust inferences

iTrust'05 Proceedings of the Third international conference on Trust Management

Quantified Score

Hi-index	0.00

Visualization

Abstract

We introduce a new hybrid recommendation method that is based on four data processing steps: 1) preprocessing of content features describing items, 2) preliminary dimensionality reduction applied to user/item vectors expressed in content features space (performed by means of SVD), 3) augmentation of normalized low-dimensional preliminary user/item vectors according to collaborative filtering data and leading to the reconstruction of user/item vectors (based on final item/user vectors and the original input matrix), and 4) the estimation of missing entries in the user-item ratings matrix. In the experiments presented in the paper, we focus on the most challenging case of extreme collaborative data sparsity. We show that a low-dimensional space is suitable for recommendation generation, despite collaborative data sparsity disqualifying the use of methods widely referenced in the relevant literature. In particular, we demonstrate that the proposed low-dimensional feature augmentation method is more effective than the well-known weighted feature combination method.