Hybrid Learning Model with Barzilai-Borwein Optimization for Context-aware Recommendations

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

3 Citations (Scopus)
18 Downloads (Pure)

Abstract

We propose an improved learning model for non-negative matrix factorization in the context-aware recommendation. We extend the collective non-negative matrix factorization through hybrid regularization method by combining multiplicative update rules with Barzilai-Borwein optimization. This provides new improved way of learning factorized matrices. We combine ratings, content features, and contextual information in three different 2-dimensional matrices. We study the performance of the proposed method on recommending top-N items. The method was empirically tested on 4 datasets, including movies, music, and mobile apps, showing an improvement in comparison with other state-of-the-art for top-N recommendations, and time convergence to the stationary point for larger datasets.
Original languageEnglish
Title of host publicationProceedings of the Thirtieth-First Florida Artificial Intelligence Research Society Conference
Number of pages6
PublisherAAAI Publications
Publication date21 May 2018
Pages456-461
Publication statusPublished - 21 May 2018
EventInternational Florida Artificial Intelligence Research Society Conference - Crowne Plaza Melbourne Oceanfront, Melbourne, United States
Duration: 21 May 201823 May 2018
Conference number: 31
https://sites.google.com/site/flairs31conference/

Conference

ConferenceInternational Florida Artificial Intelligence Research Society Conference
Number31
LocationCrowne Plaza Melbourne Oceanfront
CountryUnited States
CityMelbourne
Period21/05/201823/05/2018
Internet address

Fingerprint

Factorization
Computer music
Application programs

Cite this

Da Costa, F. S., & Dolog, P. (2018). Hybrid Learning Model with Barzilai-Borwein Optimization for Context-aware Recommendations. In Proceedings of the Thirtieth-First Florida Artificial Intelligence Research Society Conference (pp. 456-461). AAAI Publications.
Da Costa, Felipe Soares ; Dolog, Peter. / Hybrid Learning Model with Barzilai-Borwein Optimization for Context-aware Recommendations. Proceedings of the Thirtieth-First Florida Artificial Intelligence Research Society Conference. AAAI Publications, 2018. pp. 456-461
@inproceedings{82745512b91445c69c6674143b491e71,
title = "Hybrid Learning Model with Barzilai-Borwein Optimization for Context-aware Recommendations",
abstract = "We propose an improved learning model for non-negative matrix factorization in the context-aware recommendation. We extend the collective non-negative matrix factorization through hybrid regularization method by combining multiplicative update rules with Barzilai-Borwein optimization. This provides new improved way of learning factorized matrices. We combine ratings, content features, and contextual information in three different 2-dimensional matrices. We study the performance of the proposed method on recommending top-N items. The method was empirically tested on 4 datasets, including movies, music, and mobile apps, showing an improvement in comparison with other state-of-the-art for top-N recommendations, and time convergence to the stationary point for larger datasets.",
author = "{Da Costa}, {Felipe Soares} and Peter Dolog",
year = "2018",
month = "5",
day = "21",
language = "English",
pages = "456--461",
booktitle = "Proceedings of the Thirtieth-First Florida Artificial Intelligence Research Society Conference",
publisher = "AAAI Publications",

}

Da Costa, FS & Dolog, P 2018, Hybrid Learning Model with Barzilai-Borwein Optimization for Context-aware Recommendations. in Proceedings of the Thirtieth-First Florida Artificial Intelligence Research Society Conference. AAAI Publications, pp. 456-461, International Florida Artificial Intelligence Research Society Conference , Melbourne, United States, 21/05/2018.

Hybrid Learning Model with Barzilai-Borwein Optimization for Context-aware Recommendations. / Da Costa, Felipe Soares; Dolog, Peter.

Proceedings of the Thirtieth-First Florida Artificial Intelligence Research Society Conference. AAAI Publications, 2018. p. 456-461.

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

TY - GEN

T1 - Hybrid Learning Model with Barzilai-Borwein Optimization for Context-aware Recommendations

AU - Da Costa, Felipe Soares

AU - Dolog, Peter

PY - 2018/5/21

Y1 - 2018/5/21

N2 - We propose an improved learning model for non-negative matrix factorization in the context-aware recommendation. We extend the collective non-negative matrix factorization through hybrid regularization method by combining multiplicative update rules with Barzilai-Borwein optimization. This provides new improved way of learning factorized matrices. We combine ratings, content features, and contextual information in three different 2-dimensional matrices. We study the performance of the proposed method on recommending top-N items. The method was empirically tested on 4 datasets, including movies, music, and mobile apps, showing an improvement in comparison with other state-of-the-art for top-N recommendations, and time convergence to the stationary point for larger datasets.

AB - We propose an improved learning model for non-negative matrix factorization in the context-aware recommendation. We extend the collective non-negative matrix factorization through hybrid regularization method by combining multiplicative update rules with Barzilai-Borwein optimization. This provides new improved way of learning factorized matrices. We combine ratings, content features, and contextual information in three different 2-dimensional matrices. We study the performance of the proposed method on recommending top-N items. The method was empirically tested on 4 datasets, including movies, music, and mobile apps, showing an improvement in comparison with other state-of-the-art for top-N recommendations, and time convergence to the stationary point for larger datasets.

M3 - Article in proceeding

SP - 456

EP - 461

BT - Proceedings of the Thirtieth-First Florida Artificial Intelligence Research Society Conference

PB - AAAI Publications

ER -

Da Costa FS, Dolog P. Hybrid Learning Model with Barzilai-Borwein Optimization for Context-aware Recommendations. In Proceedings of the Thirtieth-First Florida Artificial Intelligence Research Society Conference. AAAI Publications. 2018. p. 456-461