An efficient mechanism for compensating vague pattern identification in support of a multi-criteria recommendation system

Yi Chung Chen, Chiang Lee

Research output: Contribution to journalArticle

Abstract

A vague pattern usually makes the result of pattern identification specious. Most existing identification algorithms try to upgrade their identification accuracies by improving the clearness of the vague pattern. However, this improvement can be limited due to the poor quality of the pattern itself. Hence, the identification result can still be untrustworthy and thus a user needs to repeat the algorithm to find another possible answer, which can be quite time-consuming. In this paper, we propose a novel pattern recommendation mechanism which is able to obtain multiple highly possible answers from a large datapool based on a given vague pattern. By using the identification algorithm only one time, a user can select a correct identification answer from these candidates given by the recommendation system. Three strategies are proposed in this paper. Experiments are performed to demonstrate the effectiveness and efficiency of the proposed mechanism.

Original languageEnglish
Pages (from-to)1633-1653
Number of pages21
JournalJournal of Information Science and Engineering
Volume34
Issue number6
DOIs
Publication statusPublished - 2018 Jan 1

Fingerprint

Recommender systems
candidacy
Experiments
efficiency
experiment

All Science Journal Classification (ASJC) codes

  • Software
  • Human-Computer Interaction
  • Hardware and Architecture
  • Library and Information Sciences
  • Computational Theory and Mathematics

Cite this

@article{950423f1878545c4a41d1857b2229530,
title = "An efficient mechanism for compensating vague pattern identification in support of a multi-criteria recommendation system",
abstract = "A vague pattern usually makes the result of pattern identification specious. Most existing identification algorithms try to upgrade their identification accuracies by improving the clearness of the vague pattern. However, this improvement can be limited due to the poor quality of the pattern itself. Hence, the identification result can still be untrustworthy and thus a user needs to repeat the algorithm to find another possible answer, which can be quite time-consuming. In this paper, we propose a novel pattern recommendation mechanism which is able to obtain multiple highly possible answers from a large datapool based on a given vague pattern. By using the identification algorithm only one time, a user can select a correct identification answer from these candidates given by the recommendation system. Three strategies are proposed in this paper. Experiments are performed to demonstrate the effectiveness and efficiency of the proposed mechanism.",
author = "Chen, {Yi Chung} and Chiang Lee",
year = "2018",
month = "1",
day = "1",
doi = "10.6688/JISE.201811_34(6).0015",
language = "English",
volume = "34",
pages = "1633--1653",
journal = "Journal of Information Science and Engineering",
issn = "1016-2364",
publisher = "Institute of Information Science",
number = "6",

}

TY - JOUR

T1 - An efficient mechanism for compensating vague pattern identification in support of a multi-criteria recommendation system

AU - Chen, Yi Chung

AU - Lee, Chiang

PY - 2018/1/1

Y1 - 2018/1/1

N2 - A vague pattern usually makes the result of pattern identification specious. Most existing identification algorithms try to upgrade their identification accuracies by improving the clearness of the vague pattern. However, this improvement can be limited due to the poor quality of the pattern itself. Hence, the identification result can still be untrustworthy and thus a user needs to repeat the algorithm to find another possible answer, which can be quite time-consuming. In this paper, we propose a novel pattern recommendation mechanism which is able to obtain multiple highly possible answers from a large datapool based on a given vague pattern. By using the identification algorithm only one time, a user can select a correct identification answer from these candidates given by the recommendation system. Three strategies are proposed in this paper. Experiments are performed to demonstrate the effectiveness and efficiency of the proposed mechanism.

AB - A vague pattern usually makes the result of pattern identification specious. Most existing identification algorithms try to upgrade their identification accuracies by improving the clearness of the vague pattern. However, this improvement can be limited due to the poor quality of the pattern itself. Hence, the identification result can still be untrustworthy and thus a user needs to repeat the algorithm to find another possible answer, which can be quite time-consuming. In this paper, we propose a novel pattern recommendation mechanism which is able to obtain multiple highly possible answers from a large datapool based on a given vague pattern. By using the identification algorithm only one time, a user can select a correct identification answer from these candidates given by the recommendation system. Three strategies are proposed in this paper. Experiments are performed to demonstrate the effectiveness and efficiency of the proposed mechanism.

UR - http://www.scopus.com/inward/record.url?scp=85056741729&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85056741729&partnerID=8YFLogxK

U2 - 10.6688/JISE.201811_34(6).0015

DO - 10.6688/JISE.201811_34(6).0015

M3 - Article

AN - SCOPUS:85056741729

VL - 34

SP - 1633

EP - 1653

JO - Journal of Information Science and Engineering

JF - Journal of Information Science and Engineering

SN - 1016-2364

IS - 6

ER -