The collaborative filtering algorithm with time weight based on MapReduce

Hongyi Su; Xianfei Lin; Bo Yan; Hong Zheng

doi:10.1007/978-3-319-22047-5_31

The collaborative filtering algorithm with time weight based on MapReduce

Hongyi Su^*, Xianfei Lin, Bo Yan, Hong Zheng

^*Corresponding author for this work

School of Computer Science and Technology

Beijing Institute of Technology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

10 Citations (Scopus)

Abstract

As one of the most successful recommendation algorithm, collaborative filtering algorithm still faces many challenges, such as accuracy, extensibility, and sparsity. In the algorithm, ratings produced in different period are treated equally, so changes of users’ interests have been ignored. This paper considers the influence of time factor on users’ interests, and presents a new algorithm that involves time decay factor in the collaborative filtering algorithm, the new algorithm makes a more accurate recommendation by reducing the weight of old data. Deploying the collaborative filtering algorithm with time weight on parallel computing frame of MapReduce also achieves the extensibility of algorithm and improves the processing performance of large data sets.

Original language	English
Title of host publication	Big Data Computing and Communications - 1st International Conference, BigCom 2015, Proceedings
Editors	Shlomo Argamon, Xiang Yang Li, Hui Xiong, JianZhong Li, Yu Wang
Publisher	Springer Verlag
Pages	386-395
Number of pages	10
ISBN (Print)	9783319220468
DOIs	https://doi.org/10.1007/978-3-319-22047-5_31
Publication status	Published - 2015
Event	1st International Conference on Big Data Computing and Communications, BigCom 2015 - Taiyuan, China Duration: 1 Aug 2015 → 3 Aug 2015

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	9196
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	1st International Conference on Big Data Computing and Communications, BigCom 2015
Country/Territory	China
City	Taiyuan
Period	1/08/15 → 3/08/15

Keywords

Collaborative filtering
Distributed application
Time weight

Access to Document

10.1007/978-3-319-22047-5_31

Cite this

Su, H., Lin, X., Yan, B., & Zheng, H. (2015). The collaborative filtering algorithm with time weight based on MapReduce. In S. Argamon, X. Y. Li, H. Xiong, J. Li, & Y. Wang (Eds.), Big Data Computing and Communications - 1st International Conference, BigCom 2015, Proceedings (pp. 386-395). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9196). Springer Verlag. https://doi.org/10.1007/978-3-319-22047-5_31

Su, Hongyi ; Lin, Xianfei ; Yan, Bo et al. / The collaborative filtering algorithm with time weight based on MapReduce. Big Data Computing and Communications - 1st International Conference, BigCom 2015, Proceedings. editor / Shlomo Argamon ; Xiang Yang Li ; Hui Xiong ; JianZhong Li ; Yu Wang. Springer Verlag, 2015. pp. 386-395 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{b5b951f090674272a6041fdd47051fb0,

title = "The collaborative filtering algorithm with time weight based on MapReduce",

abstract = "As one of the most successful recommendation algorithm, collaborative filtering algorithm still faces many challenges, such as accuracy, extensibility, and sparsity. In the algorithm, ratings produced in different period are treated equally, so changes of users{\textquoteright} interests have been ignored. This paper considers the influence of time factor on users{\textquoteright} interests, and presents a new algorithm that involves time decay factor in the collaborative filtering algorithm, the new algorithm makes a more accurate recommendation by reducing the weight of old data. Deploying the collaborative filtering algorithm with time weight on parallel computing frame of MapReduce also achieves the extensibility of algorithm and improves the processing performance of large data sets.",

keywords = "Collaborative filtering, Distributed application, Time weight",

author = "Hongyi Su and Xianfei Lin and Bo Yan and Hong Zheng",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing Switzerland 2015.; 1st International Conference on Big Data Computing and Communications, BigCom 2015 ; Conference date: 01-08-2015 Through 03-08-2015",

year = "2015",

doi = "10.1007/978-3-319-22047-5_31",

language = "English",

isbn = "9783319220468",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "386--395",

editor = "Shlomo Argamon and Li, {Xiang Yang} and Hui Xiong and JianZhong Li and Yu Wang",

booktitle = "Big Data Computing and Communications - 1st International Conference, BigCom 2015, Proceedings",

address = "Germany",

}

Su, H, Lin, X, Yan, B & Zheng, H 2015, The collaborative filtering algorithm with time weight based on MapReduce. in S Argamon, XY Li, H Xiong, J Li & Y Wang (eds), Big Data Computing and Communications - 1st International Conference, BigCom 2015, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9196, Springer Verlag, pp. 386-395, 1st International Conference on Big Data Computing and Communications, BigCom 2015, Taiyuan, China, 1/08/15. https://doi.org/10.1007/978-3-319-22047-5_31

The collaborative filtering algorithm with time weight based on MapReduce. / Su, Hongyi; Lin, Xianfei; Yan, Bo et al.
Big Data Computing and Communications - 1st International Conference, BigCom 2015, Proceedings. ed. / Shlomo Argamon; Xiang Yang Li; Hui Xiong; JianZhong Li; Yu Wang. Springer Verlag, 2015. p. 386-395 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9196).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - The collaborative filtering algorithm with time weight based on MapReduce

AU - Su, Hongyi

AU - Lin, Xianfei

AU - Yan, Bo

AU - Zheng, Hong

N1 - Publisher Copyright: © Springer International Publishing Switzerland 2015.

PY - 2015

Y1 - 2015

N2 - As one of the most successful recommendation algorithm, collaborative filtering algorithm still faces many challenges, such as accuracy, extensibility, and sparsity. In the algorithm, ratings produced in different period are treated equally, so changes of users’ interests have been ignored. This paper considers the influence of time factor on users’ interests, and presents a new algorithm that involves time decay factor in the collaborative filtering algorithm, the new algorithm makes a more accurate recommendation by reducing the weight of old data. Deploying the collaborative filtering algorithm with time weight on parallel computing frame of MapReduce also achieves the extensibility of algorithm and improves the processing performance of large data sets.

AB - As one of the most successful recommendation algorithm, collaborative filtering algorithm still faces many challenges, such as accuracy, extensibility, and sparsity. In the algorithm, ratings produced in different period are treated equally, so changes of users’ interests have been ignored. This paper considers the influence of time factor on users’ interests, and presents a new algorithm that involves time decay factor in the collaborative filtering algorithm, the new algorithm makes a more accurate recommendation by reducing the weight of old data. Deploying the collaborative filtering algorithm with time weight on parallel computing frame of MapReduce also achieves the extensibility of algorithm and improves the processing performance of large data sets.

KW - Collaborative filtering

KW - Distributed application

KW - Time weight

UR - http://www.scopus.com/inward/record.url?scp=84947295270&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-22047-5_31

DO - 10.1007/978-3-319-22047-5_31

M3 - Conference contribution

AN - SCOPUS:84947295270

SN - 9783319220468

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 386

EP - 395

BT - Big Data Computing and Communications - 1st International Conference, BigCom 2015, Proceedings

A2 - Argamon, Shlomo

A2 - Li, Xiang Yang

A2 - Xiong, Hui

A2 - Li, JianZhong

A2 - Wang, Yu

PB - Springer Verlag

T2 - 1st International Conference on Big Data Computing and Communications, BigCom 2015

Y2 - 1 August 2015 through 3 August 2015

ER -

Su H, Lin X, Yan B , Zheng H. The collaborative filtering algorithm with time weight based on MapReduce. In Argamon S, Li XY, Xiong H, Li J, Wang Y, editors, Big Data Computing and Communications - 1st International Conference, BigCom 2015, Proceedings. Springer Verlag. 2015. p. 386-395. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-22047-5_31

The collaborative filtering algorithm with time weight based on MapReduce

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this