A Collaborative Filtering Algorithm of Weighted Information Entropy and User Attributes

Xu Wang*

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Citations (Scopus)

Abstract

Collaborative filtering algorithm is one of the most successful technologies in recommendation system, and similarity calculation is the core. To solve the problem that the traditional similarity calculation method is not accurate in the case of sparse data, a cooperative filtering recommendation algorithm combining weighted entropy and user attribute is proposed in this paper. This algorithm applies the information entropy theory in the information theory to the similarity calculation, and takes into account the influence of the user interest and user difference on the similarity calculation. The simulation results show that the algorithm has a smaller MAE value than the recommended algorithm based on Pearson correlation coefficient and the cosine similarity, thus it improves the recommended quality.

Original languageEnglish
Title of host publicationProceedings - 9th International Conference on Intelligent Human-Machine Systems and Cybernetics, IHMSC 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages422-425
Number of pages4
ISBN (Electronic)9781538630228
DOIs
Publication statusPublished - 20 Sept 2017
Externally publishedYes
Event9th International Conference on Intelligent Human-Machine Systems and Cybernetics, IHMSC 2017 - Hangzhou, Zhejiang, China
Duration: 26 Aug 201727 Aug 2017

Publication series

NameProceedings - 9th International Conference on Intelligent Human-Machine Systems and Cybernetics, IHMSC 2017
Volume1

Conference

Conference9th International Conference on Intelligent Human-Machine Systems and Cybernetics, IHMSC 2017
Country/TerritoryChina
CityHangzhou, Zhejiang
Period26/08/1727/08/17

Fingerprint

Dive into the research topics of 'A Collaborative Filtering Algorithm of Weighted Information Entropy and User Attributes'. Together they form a unique fingerprint.

Cite this