Trust-based collaborative filtering: Tackling the cold start problem using regular equivalence

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

Abstract

User-based Collaborative Filtering (CF) is one of the most popular approaches to create recommender systems. This approach is based on finding the most relevant k users from whose rating history we can extract items to recommend. CF, however, suffers from data sparsity and the cold-start problem since users often rate only a small fraction of available items. One solution is to incorporate additional information into the recommendation process such as explicit trust scores that are assigned by users to others or implicit trust relationships that result from social connections between users. Such relationships typically form a very sparse trust network, which can be utilized to generate recommendations for users based on people they trust. In our work, we explore the use of regular equivalence applied to a trust network to generate a similarity matrix that is used to select the k-nearest neighbors for recommending items. We evaluate our approach on Epinions and we find that we can outperform related methods for tackling cold-start users in terms of recommendation accuracy.

LanguageEnglish
Title of host publicationRecSys 2018 - 12th ACM Conference on Recommender Systems
PublisherAssociation of Computing Machinery
Pages446-450
Number of pages5
ISBN (Electronic)9781450359016
DOIs
StatusPublished - 27 Sep 2018
Event12th ACM Conference on Recommender Systems, RecSys 2018 - Vancouver, Canada
Duration: 2 Oct 20187 Oct 2018

Conference

Conference12th ACM Conference on Recommender Systems, RecSys 2018
CountryCanada
CityVancouver
Period2/10/187/10/18

Fingerprint

Collaborative filtering
Recommender systems

Keywords

  • Cold-start
  • Collaborative Filtering
  • Katz similarity
  • Network Science
  • Recommender Systems
  • Regular Equivalence
  • Trust

ASJC Scopus subject areas

  • Hardware and Architecture
  • Software

Cite this

Duricic, T., Kowald, D., Lacic, E., & Lex, E. (2018). Trust-based collaborative filtering: Tackling the cold start problem using regular equivalence. In RecSys 2018 - 12th ACM Conference on Recommender Systems (pp. 446-450). Association of Computing Machinery. https://doi.org/10.1145/3240323.3240404

Trust-based collaborative filtering : Tackling the cold start problem using regular equivalence. / Duricic, Tomislav; Kowald, Dominik; Lacic, Emanuel; Lex, Elisabeth.

RecSys 2018 - 12th ACM Conference on Recommender Systems. Association of Computing Machinery, 2018. p. 446-450.

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

Duricic, T, Kowald, D, Lacic, E & Lex, E 2018, Trust-based collaborative filtering: Tackling the cold start problem using regular equivalence. in RecSys 2018 - 12th ACM Conference on Recommender Systems. Association of Computing Machinery, pp. 446-450, 12th ACM Conference on Recommender Systems, RecSys 2018, Vancouver, Canada, 2/10/18. https://doi.org/10.1145/3240323.3240404
Duricic T, Kowald D, Lacic E, Lex E. Trust-based collaborative filtering: Tackling the cold start problem using regular equivalence. In RecSys 2018 - 12th ACM Conference on Recommender Systems. Association of Computing Machinery. 2018. p. 446-450 https://doi.org/10.1145/3240323.3240404
Duricic, Tomislav ; Kowald, Dominik ; Lacic, Emanuel ; Lex, Elisabeth. / Trust-based collaborative filtering : Tackling the cold start problem using regular equivalence. RecSys 2018 - 12th ACM Conference on Recommender Systems. Association of Computing Machinery, 2018. pp. 446-450
@inproceedings{1220ae91139949eb93c338b7a85907de,
title = "Trust-based collaborative filtering: Tackling the cold start problem using regular equivalence",
abstract = "User-based Collaborative Filtering (CF) is one of the most popular approaches to create recommender systems. This approach is based on finding the most relevant k users from whose rating history we can extract items to recommend. CF, however, suffers from data sparsity and the cold-start problem since users often rate only a small fraction of available items. One solution is to incorporate additional information into the recommendation process such as explicit trust scores that are assigned by users to others or implicit trust relationships that result from social connections between users. Such relationships typically form a very sparse trust network, which can be utilized to generate recommendations for users based on people they trust. In our work, we explore the use of regular equivalence applied to a trust network to generate a similarity matrix that is used to select the k-nearest neighbors for recommending items. We evaluate our approach on Epinions and we find that we can outperform related methods for tackling cold-start users in terms of recommendation accuracy.",
keywords = "Cold-start, Collaborative Filtering, Katz similarity, Network Science, Recommender Systems, Regular Equivalence, Trust",
author = "Tomislav Duricic and Dominik Kowald and Emanuel Lacic and Elisabeth Lex",
year = "2018",
month = "9",
day = "27",
doi = "10.1145/3240323.3240404",
language = "English",
pages = "446--450",
booktitle = "RecSys 2018 - 12th ACM Conference on Recommender Systems",
publisher = "Association of Computing Machinery",
address = "United States",

}

TY - GEN

T1 - Trust-based collaborative filtering

T2 - Tackling the cold start problem using regular equivalence

AU - Duricic, Tomislav

AU - Kowald, Dominik

AU - Lacic, Emanuel

AU - Lex, Elisabeth

PY - 2018/9/27

Y1 - 2018/9/27

N2 - User-based Collaborative Filtering (CF) is one of the most popular approaches to create recommender systems. This approach is based on finding the most relevant k users from whose rating history we can extract items to recommend. CF, however, suffers from data sparsity and the cold-start problem since users often rate only a small fraction of available items. One solution is to incorporate additional information into the recommendation process such as explicit trust scores that are assigned by users to others or implicit trust relationships that result from social connections between users. Such relationships typically form a very sparse trust network, which can be utilized to generate recommendations for users based on people they trust. In our work, we explore the use of regular equivalence applied to a trust network to generate a similarity matrix that is used to select the k-nearest neighbors for recommending items. We evaluate our approach on Epinions and we find that we can outperform related methods for tackling cold-start users in terms of recommendation accuracy.

AB - User-based Collaborative Filtering (CF) is one of the most popular approaches to create recommender systems. This approach is based on finding the most relevant k users from whose rating history we can extract items to recommend. CF, however, suffers from data sparsity and the cold-start problem since users often rate only a small fraction of available items. One solution is to incorporate additional information into the recommendation process such as explicit trust scores that are assigned by users to others or implicit trust relationships that result from social connections between users. Such relationships typically form a very sparse trust network, which can be utilized to generate recommendations for users based on people they trust. In our work, we explore the use of regular equivalence applied to a trust network to generate a similarity matrix that is used to select the k-nearest neighbors for recommending items. We evaluate our approach on Epinions and we find that we can outperform related methods for tackling cold-start users in terms of recommendation accuracy.

KW - Cold-start

KW - Collaborative Filtering

KW - Katz similarity

KW - Network Science

KW - Recommender Systems

KW - Regular Equivalence

KW - Trust

UR - http://www.scopus.com/inward/record.url?scp=85056796714&partnerID=8YFLogxK

U2 - 10.1145/3240323.3240404

DO - 10.1145/3240323.3240404

M3 - Conference contribution

SP - 446

EP - 450

BT - RecSys 2018 - 12th ACM Conference on Recommender Systems

PB - Association of Computing Machinery

ER -