Collaborative Filtering beyond the User-Item Matrix
@article{Shi2014CollaborativeFB, title={Collaborative Filtering beyond the User-Item Matrix}, author={Yue Shi and Martha Larson and Alan Hanjalic}, journal={ACM Computing Surveys (CSUR)}, year={2014}, volume={47}, pages={1 - 45} }
Over the past two decades, a large amount of research effort has been devoted to developing algorithms that generate recommendations. The resulting research progress has established the importance of the user-item (U-I) matrix, which encodes the individual preferences of users for items in a collection, for recommender systems. The U-I matrix provides the basis for collaborative filtering (CF) techniques, the dominant framework for recommender systems. Currently, new recommendation scenarios…
Figures and Tables from this paper
694 Citations
Matrix completion with side information for effective recommendation
- Computer Science
- 2016
A novel and general algorithmic framework based on matrix factorization that simultaneously exploits the similarity information among users and items to alleviate the data sparsity issue and specifically the cold-start problem is proposed.
A Survey of Collaborative Filtering-Based Recommender Systems: From Traditional Methods to Hybrid Methods Based on Social Networks
- Computer ScienceIEEE Access
- 2018
The recent hybrid CF-based recommendation techniques fusing social networks to solve data sparsity and high dimensionality are introduced and provide a novel point of view to improve the performance of RS, thereby presenting a useful resource in the state-of-the-art research result for future researchers.
Neighborhood-Based Collaborative Recommendations: An Introduction
- Computer Science
- 2020
This chapter describes the main components of a generic framework that may be employed for neighborhood-based collaborative recommendations and discusses a few well-known problems associated with the rating data such as sparsity, long-tail, and cold-start problem.
Learning Multiple Similarities of Users and Items in Recommender Systems
- Computer Science2017 IEEE International Conference on Data Mining (ICDM)
- 2017
The key idea of MSUI is to simultaneously capture the associations between latent factors in the U-I matrix and multiple side sources of users and items through joint nonnegative matrix factorization.
Leveraging Hybrid Recommenders with Multifaceted Implicit Feedback
- Computer ScienceJ. Univers. Comput. Sci.
- 2015
A hybrid recommender model is proposed which exploits implicit feedback from users by considering not only the latent space of factors that describes the user and item, but also the available metadata associated with content and individuals.
Illuminating Recommendation by Understanding the Explicit Item Relations
- Computer ScienceJournal of Computer Science and Technology
- 2018
This paper argues that explicitly modeling the complex relations among items under domain-specific applications is an indispensable part for enhancing the recommendations and proposes some future research directions in recommendation area.
A Literature Review of Empirical Studies of Recommendation Systems
- Computer Science
- 2015
A survey of six existing recommendation system, which recognized a lack of authority and long term research interest in the field, 78% of the authors published no more than one paper on research paper recommender systems, and there was miniature cooperation among different co-author groups.
A social recommender system using item asymmetric correlation
- Computer ScienceApplied Intelligence
- 2017
This paper explored the effect of combining the implicit relationships of the items and user-item matrix on the accuracy of recommendations and found the new Item Asymmetric Correlation (IAC) method to be an adequate option to replace the constraints.
Collaborative Heterogeneous Information Embedding for Recommender Systems
- Computer ScienceAPWeb/WAIM Workshops
- 2017
This paper proposes a novel hybrid recommendation method called Collaborative Heterogeneous Information Embedding (CHIE), which jointly performs fused representation learning for items in HIN and Probabilistic Matrix Factorization (PMF), a model-based CF, for the ratings matrix.
Matrix Factorization Techniques for Context-Aware Collaborative Filtering Recommender Systems: A Survey
- Computer ScienceComput. Inf. Sci.
- 2018
This survey paper presents a detailed literature review of Context-aware Recommender Systems and approaches to improving performance for large scale datasets and the impact of incorporating contextual information on the quality and accuracy of the recommendation.
References
SHOWING 1-10 OF 229 REFERENCES
Recommender systems: from algorithms to user experience
- Computer ScienceUser Modeling and User-Adapted Interaction
- 2011
It is argued that evaluating the user experience of a recommender requires a broader set of measures than have been commonly used, and additional measures that have proven effective are suggested.
Item-based collaborative filtering recommendation algorithms
- Computer ScienceWWW '01
- 2001
This paper analyzes item-based collaborative ltering techniques and suggests that item- based algorithms provide dramatically better performance than user-based algorithms, while at the same time providing better quality than the best available userbased algorithms.
Learning to recommend with trust and distrust relationships
- Computer ScienceRecSys '09
- 2009
A factor analysis-based optimization framework to incorporate the user trust and distrust relationships into the recommender systems and the experimental results show that the distrust relations among users are as important as the trust relations.
Learning to recommend with social trust ensemble
- Computer ScienceSIGIR
- 2009
This work proposes a novel probabilistic factor analysis framework, which naturally fuses the users' tastes and their trusted friends' favors together and coin the term Social Trust Ensemble to represent the formulation of the social trust restrictions on the recommender systems.
Item-based top-N recommendation algorithms
- Computer ScienceTOIS
- 2004
This article presents one class of model-based recommendation algorithms that first determines the similarities between the various items and then uses them to identify the set of items to be recommended, and shows that these item-based algorithms are up to two orders of magnitude faster than the traditional user-neighborhood based recommender systems and provide recommendations with comparable or better quality.
Improving Recommender Systems by Incorporating Social Contextual Information
- Computer ScienceTOIS
- 2011
This article proposes a factor analysis approach based on probabilistic matrix factorization to alleviate the data sparsity and poor prediction accuracy problems by incorporating social contextual information, such as social networks and social tags.
Context-based splitting of item ratings in collaborative filtering
- Computer ScienceRecSys '09
- 2009
This paper introduces and analyzes a novel pre-filtering technique for context-aware CF called item splitting, in which the ratings of certain items are split, according to the value of an item-dependent contextual condition.
Active Dual Collaborative Filtering with Both Item and Attribute Feedback
- Computer ScienceAAAI
- 2011
This paper designs a unified active CF framework for incorporating both item and attribute feedback based on the random walk model and shows that it can achieve much better recommendation quality as compared to traditional active CF methods that support only item feedback.
Tag-aware recommender systems by fusion of collaborative filtering algorithms
- Computer ScienceSAC '08
- 2008
A generic method is proposed that allows tags to be incorporated to standard CF algorithms, by reducing the three-dimensional correlations to three two- dimensional correlations and then applying a fusion method to re-associate these correlations.
Time weight collaborative filtering
- Computer ScienceCIKM '05
- 2005
This paper presents a novel algorithm to compute the time weights for different items in a manner that will assign a decreasing weight to old data, and uses clustering to discriminate between different kinds of items.