An observation-weighting method for mining actionable behavioral rules

Peng Su, Lei Wang, Dajun Zeng, Yuan Liu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)

Abstract

One of the critical challenges faced by the mainstream data mining community is to make the mined patterns or knowledge actionable. Knowledge is considered actionable if users can take direct actions based on such knowledge to their advantage. Among the most important and distinctive actionable knowledge are actionable behavioral rules that can directly and explicitly suggest specific actions to take to influence (restrain or encourage) the behavior in the users' best interest. The problem of mining such rules is a search problem in a framework of support and expected utility. The previous definition of a rule's support assumes that each instance which supports a rule has the uniform contribution to the support. However, this assumption is usually violated in practice to some extent, and thus will hinder the performance of algorithms for mining such rules. In this paper, to handle this problem, an observation-weighting model for support and corresponding mining algorithm are proposed. The experimental results strongly suggest the validity and the superiority of our approach.

Original languageEnglish (US)
Title of host publication2015 7th International Conference on Advanced Computational Intelligence, ICACI 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages120-125
Number of pages6
ISBN (Print)9781479972579
DOIs
StatePublished - Aug 10 2015
Externally publishedYes
Event7th International Conference on Advanced Computational Intelligence, ICACI 2015 - Wuyi, China
Duration: Mar 27 2015Mar 29 2015

Other

Other7th International Conference on Advanced Computational Intelligence, ICACI 2015
CountryChina
CityWuyi
Period3/27/153/29/15

Fingerprint

Data mining

Keywords

  • Electronic mail
  • Training

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Networks and Communications
  • Computer Vision and Pattern Recognition

Cite this

Su, P., Wang, L., Zeng, D., & Liu, Y. (2015). An observation-weighting method for mining actionable behavioral rules. In 2015 7th International Conference on Advanced Computational Intelligence, ICACI 2015 (pp. 120-125). [7184761] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICACI.2015.7184761

An observation-weighting method for mining actionable behavioral rules. / Su, Peng; Wang, Lei; Zeng, Dajun; Liu, Yuan.

2015 7th International Conference on Advanced Computational Intelligence, ICACI 2015. Institute of Electrical and Electronics Engineers Inc., 2015. p. 120-125 7184761.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Su, P, Wang, L, Zeng, D & Liu, Y 2015, An observation-weighting method for mining actionable behavioral rules. in 2015 7th International Conference on Advanced Computational Intelligence, ICACI 2015., 7184761, Institute of Electrical and Electronics Engineers Inc., pp. 120-125, 7th International Conference on Advanced Computational Intelligence, ICACI 2015, Wuyi, China, 3/27/15. https://doi.org/10.1109/ICACI.2015.7184761
Su P, Wang L, Zeng D, Liu Y. An observation-weighting method for mining actionable behavioral rules. In 2015 7th International Conference on Advanced Computational Intelligence, ICACI 2015. Institute of Electrical and Electronics Engineers Inc. 2015. p. 120-125. 7184761 https://doi.org/10.1109/ICACI.2015.7184761
Su, Peng ; Wang, Lei ; Zeng, Dajun ; Liu, Yuan. / An observation-weighting method for mining actionable behavioral rules. 2015 7th International Conference on Advanced Computational Intelligence, ICACI 2015. Institute of Electrical and Electronics Engineers Inc., 2015. pp. 120-125
@inproceedings{eea134ff72ff4fa88052668b0ce74feb,
title = "An observation-weighting method for mining actionable behavioral rules",
abstract = "One of the critical challenges faced by the mainstream data mining community is to make the mined patterns or knowledge actionable. Knowledge is considered actionable if users can take direct actions based on such knowledge to their advantage. Among the most important and distinctive actionable knowledge are actionable behavioral rules that can directly and explicitly suggest specific actions to take to influence (restrain or encourage) the behavior in the users' best interest. The problem of mining such rules is a search problem in a framework of support and expected utility. The previous definition of a rule's support assumes that each instance which supports a rule has the uniform contribution to the support. However, this assumption is usually violated in practice to some extent, and thus will hinder the performance of algorithms for mining such rules. In this paper, to handle this problem, an observation-weighting model for support and corresponding mining algorithm are proposed. The experimental results strongly suggest the validity and the superiority of our approach.",
keywords = "Electronic mail, Training",
author = "Peng Su and Lei Wang and Dajun Zeng and Yuan Liu",
year = "2015",
month = "8",
day = "10",
doi = "10.1109/ICACI.2015.7184761",
language = "English (US)",
isbn = "9781479972579",
pages = "120--125",
booktitle = "2015 7th International Conference on Advanced Computational Intelligence, ICACI 2015",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - An observation-weighting method for mining actionable behavioral rules

AU - Su, Peng

AU - Wang, Lei

AU - Zeng, Dajun

AU - Liu, Yuan

PY - 2015/8/10

Y1 - 2015/8/10

N2 - One of the critical challenges faced by the mainstream data mining community is to make the mined patterns or knowledge actionable. Knowledge is considered actionable if users can take direct actions based on such knowledge to their advantage. Among the most important and distinctive actionable knowledge are actionable behavioral rules that can directly and explicitly suggest specific actions to take to influence (restrain or encourage) the behavior in the users' best interest. The problem of mining such rules is a search problem in a framework of support and expected utility. The previous definition of a rule's support assumes that each instance which supports a rule has the uniform contribution to the support. However, this assumption is usually violated in practice to some extent, and thus will hinder the performance of algorithms for mining such rules. In this paper, to handle this problem, an observation-weighting model for support and corresponding mining algorithm are proposed. The experimental results strongly suggest the validity and the superiority of our approach.

AB - One of the critical challenges faced by the mainstream data mining community is to make the mined patterns or knowledge actionable. Knowledge is considered actionable if users can take direct actions based on such knowledge to their advantage. Among the most important and distinctive actionable knowledge are actionable behavioral rules that can directly and explicitly suggest specific actions to take to influence (restrain or encourage) the behavior in the users' best interest. The problem of mining such rules is a search problem in a framework of support and expected utility. The previous definition of a rule's support assumes that each instance which supports a rule has the uniform contribution to the support. However, this assumption is usually violated in practice to some extent, and thus will hinder the performance of algorithms for mining such rules. In this paper, to handle this problem, an observation-weighting model for support and corresponding mining algorithm are proposed. The experimental results strongly suggest the validity and the superiority of our approach.

KW - Electronic mail

KW - Training

UR - http://www.scopus.com/inward/record.url?scp=84954307853&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84954307853&partnerID=8YFLogxK

U2 - 10.1109/ICACI.2015.7184761

DO - 10.1109/ICACI.2015.7184761

M3 - Conference contribution

AN - SCOPUS:84954307853

SN - 9781479972579

SP - 120

EP - 125

BT - 2015 7th International Conference on Advanced Computational Intelligence, ICACI 2015

PB - Institute of Electrical and Electronics Engineers Inc.

ER -