Learning business rules with association rule classifiers

Tomáš Kliegr, Jaroslav Kuchař, Davide Sottara, Stanislav Vojíř

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

12 Citations (Scopus)

Abstract

The main obstacles for a straightforward use of association rules as candidate business rules are the excessive number of rules discovered even on small datasets, and the fact that contradicting rules are generated. This paper shows that Association Rule Classification algorithms, such as CBA, solve both these problems, and provides a practical guide on using discovered rules in the Drools BRMS and on setting the ARC parameters. Experiments performed with modified CBA on several UCI datasets indicate that data coverage rule pruning keeps the number of rules manageable, while not adversely impacting the accuracy. The best results in terms of overall accuracy are obtained using minimum support and confidence thresholds. Disjunction between attribute values seem to provide a desirable balance between accuracy and rule count, while negated literals have not been found beneficial.

Original languageEnglish (US)
Title of host publicationRules on the Web
Subtitle of host publicationFrom Theory to Applications - 8th International Symposium, RuleML 2014, Co-located with the 21st European Conference on Artificial Intelligence, ECAI 2014, Proceedings
PublisherSpringer Verlag
Pages236-250
Number of pages15
ISBN (Print)9783319098692
DOIs
StatePublished - Jan 1 2014
Event8th International Web Rule Symposium, RuleML 2014, Co-located with the 21st European Conference on Artificial Intelligence, ECAI 2014 - Prague, Czech Republic
Duration: Aug 18 2014Aug 20 2014

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume8620 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other8th International Web Rule Symposium, RuleML 2014, Co-located with the 21st European Conference on Artificial Intelligence, ECAI 2014
CountryCzech Republic
CityPrague
Period8/18/148/20/14

Fingerprint

Business Rules
Association rules
Association Rules
Classifiers
Classifier
Industry
Experiments
Classification Algorithm
Pruning
Confidence
Learning
Count
Coverage
Attribute
Experiment

Keywords

  • association rules
  • business rules
  • Drools
  • rule pruning

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Kliegr, T., Kuchař, J., Sottara, D., & Vojíř, S. (2014). Learning business rules with association rule classifiers. In Rules on the Web: From Theory to Applications - 8th International Symposium, RuleML 2014, Co-located with the 21st European Conference on Artificial Intelligence, ECAI 2014, Proceedings (pp. 236-250). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8620 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-09870-8_18

Learning business rules with association rule classifiers. / Kliegr, Tomáš; Kuchař, Jaroslav; Sottara, Davide; Vojíř, Stanislav.

Rules on the Web: From Theory to Applications - 8th International Symposium, RuleML 2014, Co-located with the 21st European Conference on Artificial Intelligence, ECAI 2014, Proceedings. Springer Verlag, 2014. p. 236-250 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8620 LNCS).

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

Kliegr, T, Kuchař, J, Sottara, D & Vojíř, S 2014, Learning business rules with association rule classifiers. in Rules on the Web: From Theory to Applications - 8th International Symposium, RuleML 2014, Co-located with the 21st European Conference on Artificial Intelligence, ECAI 2014, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 8620 LNCS, Springer Verlag, pp. 236-250, 8th International Web Rule Symposium, RuleML 2014, Co-located with the 21st European Conference on Artificial Intelligence, ECAI 2014, Prague, Czech Republic, 8/18/14. https://doi.org/10.1007/978-3-319-09870-8_18
Kliegr T, Kuchař J, Sottara D, Vojíř S. Learning business rules with association rule classifiers. In Rules on the Web: From Theory to Applications - 8th International Symposium, RuleML 2014, Co-located with the 21st European Conference on Artificial Intelligence, ECAI 2014, Proceedings. Springer Verlag. 2014. p. 236-250. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-09870-8_18
Kliegr, Tomáš ; Kuchař, Jaroslav ; Sottara, Davide ; Vojíř, Stanislav. / Learning business rules with association rule classifiers. Rules on the Web: From Theory to Applications - 8th International Symposium, RuleML 2014, Co-located with the 21st European Conference on Artificial Intelligence, ECAI 2014, Proceedings. Springer Verlag, 2014. pp. 236-250 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{b3e1291c677d44469010bc1cda268196,
title = "Learning business rules with association rule classifiers",
abstract = "The main obstacles for a straightforward use of association rules as candidate business rules are the excessive number of rules discovered even on small datasets, and the fact that contradicting rules are generated. This paper shows that Association Rule Classification algorithms, such as CBA, solve both these problems, and provides a practical guide on using discovered rules in the Drools BRMS and on setting the ARC parameters. Experiments performed with modified CBA on several UCI datasets indicate that data coverage rule pruning keeps the number of rules manageable, while not adversely impacting the accuracy. The best results in terms of overall accuracy are obtained using minimum support and confidence thresholds. Disjunction between attribute values seem to provide a desirable balance between accuracy and rule count, while negated literals have not been found beneficial.",
keywords = "association rules, business rules, Drools, rule pruning",
author = "Tom{\'a}š Kliegr and Jaroslav Kuchař and Davide Sottara and Stanislav Voj{\'i}ř",
year = "2014",
month = "1",
day = "1",
doi = "10.1007/978-3-319-09870-8_18",
language = "English (US)",
isbn = "9783319098692",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "236--250",
booktitle = "Rules on the Web",

}

TY - GEN

T1 - Learning business rules with association rule classifiers

AU - Kliegr, Tomáš

AU - Kuchař, Jaroslav

AU - Sottara, Davide

AU - Vojíř, Stanislav

PY - 2014/1/1

Y1 - 2014/1/1

N2 - The main obstacles for a straightforward use of association rules as candidate business rules are the excessive number of rules discovered even on small datasets, and the fact that contradicting rules are generated. This paper shows that Association Rule Classification algorithms, such as CBA, solve both these problems, and provides a practical guide on using discovered rules in the Drools BRMS and on setting the ARC parameters. Experiments performed with modified CBA on several UCI datasets indicate that data coverage rule pruning keeps the number of rules manageable, while not adversely impacting the accuracy. The best results in terms of overall accuracy are obtained using minimum support and confidence thresholds. Disjunction between attribute values seem to provide a desirable balance between accuracy and rule count, while negated literals have not been found beneficial.

AB - The main obstacles for a straightforward use of association rules as candidate business rules are the excessive number of rules discovered even on small datasets, and the fact that contradicting rules are generated. This paper shows that Association Rule Classification algorithms, such as CBA, solve both these problems, and provides a practical guide on using discovered rules in the Drools BRMS and on setting the ARC parameters. Experiments performed with modified CBA on several UCI datasets indicate that data coverage rule pruning keeps the number of rules manageable, while not adversely impacting the accuracy. The best results in terms of overall accuracy are obtained using minimum support and confidence thresholds. Disjunction between attribute values seem to provide a desirable balance between accuracy and rule count, while negated literals have not been found beneficial.

KW - association rules

KW - business rules

KW - Drools

KW - rule pruning

UR - http://www.scopus.com/inward/record.url?scp=84905260896&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84905260896&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-09870-8_18

DO - 10.1007/978-3-319-09870-8_18

M3 - Conference contribution

SN - 9783319098692

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

SP - 236

EP - 250

BT - Rules on the Web

PB - Springer Verlag

ER -