Improving the accuracy of top-N recommendation using a preference model

Jongwuk Lee, Dongwon Lee, Yeon Chang Lee, Won Seok Hwang, Sang Wook Kim

Research output: Contribution to journalArticle

38 Citations (Scopus)

Abstract

In this paper, we study the problem of retrieving a ranked list of top-N items to a target user in recommender systems. We first develop a novel preference model by distinguishing different rating patterns of users, and then apply it to existing collaborative filtering (CF) algorithms. Our preference model, which is inspired by a voting method, is well-suited for representing qualitative user preferences. In particular, it can be easily implemented with less than 100 lines of codes on top of existing CF algorithms such as user-based, item-based, and matrix-factorization-based algorithms. When our preference model is combined to three kinds of CF algorithms, experimental results demonstrate that the preference model can improve the accuracy of all existing CF algorithms such as ATOP and NDCG@25 by 3-24% and 6-98%, respectively.

Original languageEnglish (US)
Pages (from-to)290-304
Number of pages15
JournalInformation Sciences
Volume348
DOIs
StatePublished - Jun 20 2016

Fingerprint

Collaborative filtering
Collaborative Filtering
Recommendations
Model
Matrix Factorization
Recommender Systems
Recommender systems
User Preferences
Voting
Factorization
Target
Line
Experimental Results
Demonstrate

All Science Journal Classification (ASJC) codes

  • Software
  • Control and Systems Engineering
  • Theoretical Computer Science
  • Computer Science Applications
  • Information Systems and Management
  • Artificial Intelligence

Cite this

Lee, Jongwuk ; Lee, Dongwon ; Lee, Yeon Chang ; Hwang, Won Seok ; Kim, Sang Wook. / Improving the accuracy of top-N recommendation using a preference model. In: Information Sciences. 2016 ; Vol. 348. pp. 290-304.
@article{693e62e1f85c4784be9e1bfc659ea139,
title = "Improving the accuracy of top-N recommendation using a preference model",
abstract = "In this paper, we study the problem of retrieving a ranked list of top-N items to a target user in recommender systems. We first develop a novel preference model by distinguishing different rating patterns of users, and then apply it to existing collaborative filtering (CF) algorithms. Our preference model, which is inspired by a voting method, is well-suited for representing qualitative user preferences. In particular, it can be easily implemented with less than 100 lines of codes on top of existing CF algorithms such as user-based, item-based, and matrix-factorization-based algorithms. When our preference model is combined to three kinds of CF algorithms, experimental results demonstrate that the preference model can improve the accuracy of all existing CF algorithms such as ATOP and NDCG@25 by 3-24{\%} and 6-98{\%}, respectively.",
author = "Jongwuk Lee and Dongwon Lee and Lee, {Yeon Chang} and Hwang, {Won Seok} and Kim, {Sang Wook}",
year = "2016",
month = "6",
day = "20",
doi = "10.1016/j.ins.2016.02.005",
language = "English (US)",
volume = "348",
pages = "290--304",
journal = "Information Sciences",
issn = "0020-0255",
publisher = "Elsevier Inc.",

}

Improving the accuracy of top-N recommendation using a preference model. / Lee, Jongwuk; Lee, Dongwon; Lee, Yeon Chang; Hwang, Won Seok; Kim, Sang Wook.

In: Information Sciences, Vol. 348, 20.06.2016, p. 290-304.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Improving the accuracy of top-N recommendation using a preference model

AU - Lee, Jongwuk

AU - Lee, Dongwon

AU - Lee, Yeon Chang

AU - Hwang, Won Seok

AU - Kim, Sang Wook

PY - 2016/6/20

Y1 - 2016/6/20

N2 - In this paper, we study the problem of retrieving a ranked list of top-N items to a target user in recommender systems. We first develop a novel preference model by distinguishing different rating patterns of users, and then apply it to existing collaborative filtering (CF) algorithms. Our preference model, which is inspired by a voting method, is well-suited for representing qualitative user preferences. In particular, it can be easily implemented with less than 100 lines of codes on top of existing CF algorithms such as user-based, item-based, and matrix-factorization-based algorithms. When our preference model is combined to three kinds of CF algorithms, experimental results demonstrate that the preference model can improve the accuracy of all existing CF algorithms such as ATOP and NDCG@25 by 3-24% and 6-98%, respectively.

AB - In this paper, we study the problem of retrieving a ranked list of top-N items to a target user in recommender systems. We first develop a novel preference model by distinguishing different rating patterns of users, and then apply it to existing collaborative filtering (CF) algorithms. Our preference model, which is inspired by a voting method, is well-suited for representing qualitative user preferences. In particular, it can be easily implemented with less than 100 lines of codes on top of existing CF algorithms such as user-based, item-based, and matrix-factorization-based algorithms. When our preference model is combined to three kinds of CF algorithms, experimental results demonstrate that the preference model can improve the accuracy of all existing CF algorithms such as ATOP and NDCG@25 by 3-24% and 6-98%, respectively.

UR - http://www.scopus.com/inward/record.url?scp=84959512777&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84959512777&partnerID=8YFLogxK

U2 - 10.1016/j.ins.2016.02.005

DO - 10.1016/j.ins.2016.02.005

M3 - Article

AN - SCOPUS:84959512777

VL - 348

SP - 290

EP - 304

JO - Information Sciences

JF - Information Sciences

SN - 0020-0255

ER -