GroupLens

Konstan, Joseph A.; Miller, Bradley N.; Maltz, David A.; Herlocker, Jonathan L.; Gordon, Lee R.; Riedl, John

doi:10.1145/245108.245126

Cited by 1,909 publications

(808 citation statements)

References 5 publications

Supporting

Mentioning

761

Contrasting

Unclassified

Order By: Relevance

“…A personalized recommender system includes three parts: data collection, model analysis and the recommendation algorithm, among which the algorithm is the core part. Motivated by the significance in economy and society, various kinds of algorithms have been proposed, including collaborative filtering (CF) approaches [7][8][9][10][11][12][13][14][15], content-based analyses [16,17], tag-based algorithms [18][19][20], link prediction approach [21], hybrid algorithms [22][23][24], and so on. For a review of current progress, see Refs.…”

Section: Introductionmentioning

confidence: 99%

Information filtering via weighted heat conduction algorithm

Liu

Guo

Zhang

2011

Physica A: Statistical Mechanics and its Applications

View full text Add to dashboard Cite

In this paper, by taking into account effects of the user and object correlations on a heat conduction (HC) algorithm, a weighted heat conduction (WHC) algorithm is presented. We argue that the edge weight of the user-object bipartite network should be embedded into the HC algorithm to measure the object similarity. The numerical results indicate that both the accuracy and diversity could be improved greatly compared with the standard HC algorithm and the optimal values reached simultaneously. On the Movielens and Netflix datasets, the algorithmic accuracy, measured by the average ranking score, can be improved by 39.7% and 56.1% in the optimal case, respectively, and the diversity could reach 0.9587 and 0.9317 when the recommendation list equals to 5. Further statistical analysis indicates that, in the optimal case, the distributions of the edge weight are changed to the Poisson form, which may be the reason why HC algorithm performance could be improved. This work highlights the effect of edge weight on a personalized recommendation study, which maybe an important factor affecting personalized recommendation performance.

show abstract

Section: Introductionmentioning

confidence: 99%

Information filtering via weighted heat conduction algorithm

Liu

Guo

Zhang

2011

Physica A: Statistical Mechanics and its Applications

View full text Add to dashboard Cite

show abstract

“…For example, in "high frequency" consumer domains, where users purchase multiple variants of the same product type over time (e.g., books, music, movies), one can learn much about a user's preferences based on passive observations of purchase history using collaborative filtering techniques (Konstan et al, 1997;Hofmann and Puzicha, 1999;Marlin and Zemel, 2007).…”

Section: Product Configurationmentioning

confidence: 99%

Computational Decision Support Regret-Based Models for Optimization and Preference Elicitation

Boutilier¹

2013

Comparative Decision Making

View full text Add to dashboard Cite

“…On the other hand, an abstraction is performed that usually leads to an information loss and the adaptability of the model to changes to a profile is an issue. Some applications of collaborative filtering can be found in [21], [17], [22] and [14].…”

Section: Fig 1 Memory Based Collaborative Filteringmentioning

confidence: 99%

Application of ART2 Networks and Self-Organizing Maps to Collaborative Filtering

Graef

Schaefer

2002

Hypermedia: Openness, Structural Awareness, and Adaptivity

View full text Add to dashboard Cite

Abstract. Since the World Wide Web has become widespread, more and more applications exist that are suitable for the application of social information filtering techniques. In collaborative filtering, preferences of a user are estimated through mining data available about the whole user population, implicitly exploiting analogies between users that show similar characteristics. These preferences are then normally used to filter content or functionality of an application. Two important factors for the quality of the filtering process are the number of users and the amount of information (such as observed behaviors) available about each user. Another factor is the number of objects in the pool of the application that can be considered during the filtering process. Today in most cases memory based approaches to collaborative filtering are used. Unfortunately with O(#users * #items) those do not scale well. Therefore we implemented a model based approach using two different types of neural networks and benchmarked them against a widely used memory based approach. Especially with ART2 networks we obtained some encouraging results.

show abstract

GroupLens

Cited by 1,909 publications

References 5 publications

Information filtering via weighted heat conduction algorithm

Information filtering via weighted heat conduction algorithm

Computational Decision Support Regret-Based Models for Optimization and Preference Elicitation

Application of ART2 Networks and Self-Organizing Maps to Collaborative Filtering

Contact Info

Product

Resources

About