Using Association rules for classification from databases having class label ambiguities

A belief theoretic method

S. P. Subasingha, J. Zhang, Kamal Premaratne, Mei-Ling Shyu, Miroslav Kubat, K. K R G K Hewawasam

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

This chapter introduces a belief theoretic method for classification from databases having class label ambiguities. It uses a set of association rules extracted from such a database. It is assumed that a training data set with an adequate number of pre-classified instances, where each instance is assigned with an integer class label, is available. We use a modified association rule mining (ARM) technique to extract the interesting rules from the training data set and use a belief theoretic classifier based on the extracted rules to classify the incoming feature vectors. The ambiguity modelling capability of belief theory enables our classifier to perform better in the presence of class label ambiguities. It can also address the issue of the training data set being unbalanced or highly skewed by ensuring that an approximately equal number of rules are generated for each class. All these capabilities make our classifier ideally suited for those applications where (1) different experts may have conflicting opinions about the class label to be assigned to a specific training data instance; and (2) the majority of the training data instances are likely to represent a few classes giving rise to highly skewed databases. Therefore, the proposed classifier would be extremely useful in security monitoring and threat classification environments where conflicting expert opinions about the threat level are common and only a few training data instances would be considered to pose a heightened threat level. Several experiments are conducted to evaluate our proposed classifier. These experiments use several databases from the UCI data repository and data sets collected from the airport terminal simulation platform developed at the Distributed Decision Environments (DDE) Laboratory at the Department of Electrical and Computer Engineering, University of Miami. The experimental results show that, while the proposed classifier's performance is comparable to some existing classifiers when the databases have no class label ambiguities, it provides superior classification accuracy and better efficiency when class label ambiguities are present.

Original languageEnglish
Title of host publicationStudies in Computational Intelligence
Pages539-562
Number of pages24
Volume118
DOIs
StatePublished - Sep 15 2008

Publication series

NameStudies in Computational Intelligence
Volume118
ISSN (Print)1860949X

Fingerprint

Association rules
Labels
Classifiers
Airports
Experiments
Monitoring

ASJC Scopus subject areas

  • Artificial Intelligence

Cite this

Subasingha, S. P., Zhang, J., Premaratne, K., Shyu, M-L., Kubat, M., & Hewawasam, K. K. R. G. K. (2008). Using Association rules for classification from databases having class label ambiguities: A belief theoretic method. In Studies in Computational Intelligence (Vol. 118, pp. 539-562). (Studies in Computational Intelligence; Vol. 118). https://doi.org/10.1007/978-3-540-78488-3_32

Using Association rules for classification from databases having class label ambiguities : A belief theoretic method. / Subasingha, S. P.; Zhang, J.; Premaratne, Kamal; Shyu, Mei-Ling; Kubat, Miroslav; Hewawasam, K. K R G K.

Studies in Computational Intelligence. Vol. 118 2008. p. 539-562 (Studies in Computational Intelligence; Vol. 118).

Research output: Chapter in Book/Report/Conference proceedingChapter

Subasingha, SP, Zhang, J, Premaratne, K, Shyu, M-L, Kubat, M & Hewawasam, KKRGK 2008, Using Association rules for classification from databases having class label ambiguities: A belief theoretic method. in Studies in Computational Intelligence. vol. 118, Studies in Computational Intelligence, vol. 118, pp. 539-562. https://doi.org/10.1007/978-3-540-78488-3_32
Subasingha SP, Zhang J, Premaratne K, Shyu M-L, Kubat M, Hewawasam KKRGK. Using Association rules for classification from databases having class label ambiguities: A belief theoretic method. In Studies in Computational Intelligence. Vol. 118. 2008. p. 539-562. (Studies in Computational Intelligence). https://doi.org/10.1007/978-3-540-78488-3_32
Subasingha, S. P. ; Zhang, J. ; Premaratne, Kamal ; Shyu, Mei-Ling ; Kubat, Miroslav ; Hewawasam, K. K R G K. / Using Association rules for classification from databases having class label ambiguities : A belief theoretic method. Studies in Computational Intelligence. Vol. 118 2008. pp. 539-562 (Studies in Computational Intelligence).
@inbook{3c987065c0524ce0aa0478e499ae750a,
title = "Using Association rules for classification from databases having class label ambiguities: A belief theoretic method",
abstract = "This chapter introduces a belief theoretic method for classification from databases having class label ambiguities. It uses a set of association rules extracted from such a database. It is assumed that a training data set with an adequate number of pre-classified instances, where each instance is assigned with an integer class label, is available. We use a modified association rule mining (ARM) technique to extract the interesting rules from the training data set and use a belief theoretic classifier based on the extracted rules to classify the incoming feature vectors. The ambiguity modelling capability of belief theory enables our classifier to perform better in the presence of class label ambiguities. It can also address the issue of the training data set being unbalanced or highly skewed by ensuring that an approximately equal number of rules are generated for each class. All these capabilities make our classifier ideally suited for those applications where (1) different experts may have conflicting opinions about the class label to be assigned to a specific training data instance; and (2) the majority of the training data instances are likely to represent a few classes giving rise to highly skewed databases. Therefore, the proposed classifier would be extremely useful in security monitoring and threat classification environments where conflicting expert opinions about the threat level are common and only a few training data instances would be considered to pose a heightened threat level. Several experiments are conducted to evaluate our proposed classifier. These experiments use several databases from the UCI data repository and data sets collected from the airport terminal simulation platform developed at the Distributed Decision Environments (DDE) Laboratory at the Department of Electrical and Computer Engineering, University of Miami. The experimental results show that, while the proposed classifier's performance is comparable to some existing classifiers when the databases have no class label ambiguities, it provides superior classification accuracy and better efficiency when class label ambiguities are present.",
author = "Subasingha, {S. P.} and J. Zhang and Kamal Premaratne and Mei-Ling Shyu and Miroslav Kubat and Hewawasam, {K. K R G K}",
year = "2008",
month = "9",
day = "15",
doi = "10.1007/978-3-540-78488-3_32",
language = "English",
isbn = "9783540784876",
volume = "118",
series = "Studies in Computational Intelligence",
pages = "539--562",
booktitle = "Studies in Computational Intelligence",

}

TY - CHAP

T1 - Using Association rules for classification from databases having class label ambiguities

T2 - A belief theoretic method

AU - Subasingha, S. P.

AU - Zhang, J.

AU - Premaratne, Kamal

AU - Shyu, Mei-Ling

AU - Kubat, Miroslav

AU - Hewawasam, K. K R G K

PY - 2008/9/15

Y1 - 2008/9/15

N2 - This chapter introduces a belief theoretic method for classification from databases having class label ambiguities. It uses a set of association rules extracted from such a database. It is assumed that a training data set with an adequate number of pre-classified instances, where each instance is assigned with an integer class label, is available. We use a modified association rule mining (ARM) technique to extract the interesting rules from the training data set and use a belief theoretic classifier based on the extracted rules to classify the incoming feature vectors. The ambiguity modelling capability of belief theory enables our classifier to perform better in the presence of class label ambiguities. It can also address the issue of the training data set being unbalanced or highly skewed by ensuring that an approximately equal number of rules are generated for each class. All these capabilities make our classifier ideally suited for those applications where (1) different experts may have conflicting opinions about the class label to be assigned to a specific training data instance; and (2) the majority of the training data instances are likely to represent a few classes giving rise to highly skewed databases. Therefore, the proposed classifier would be extremely useful in security monitoring and threat classification environments where conflicting expert opinions about the threat level are common and only a few training data instances would be considered to pose a heightened threat level. Several experiments are conducted to evaluate our proposed classifier. These experiments use several databases from the UCI data repository and data sets collected from the airport terminal simulation platform developed at the Distributed Decision Environments (DDE) Laboratory at the Department of Electrical and Computer Engineering, University of Miami. The experimental results show that, while the proposed classifier's performance is comparable to some existing classifiers when the databases have no class label ambiguities, it provides superior classification accuracy and better efficiency when class label ambiguities are present.

AB - This chapter introduces a belief theoretic method for classification from databases having class label ambiguities. It uses a set of association rules extracted from such a database. It is assumed that a training data set with an adequate number of pre-classified instances, where each instance is assigned with an integer class label, is available. We use a modified association rule mining (ARM) technique to extract the interesting rules from the training data set and use a belief theoretic classifier based on the extracted rules to classify the incoming feature vectors. The ambiguity modelling capability of belief theory enables our classifier to perform better in the presence of class label ambiguities. It can also address the issue of the training data set being unbalanced or highly skewed by ensuring that an approximately equal number of rules are generated for each class. All these capabilities make our classifier ideally suited for those applications where (1) different experts may have conflicting opinions about the class label to be assigned to a specific training data instance; and (2) the majority of the training data instances are likely to represent a few classes giving rise to highly skewed databases. Therefore, the proposed classifier would be extremely useful in security monitoring and threat classification environments where conflicting expert opinions about the threat level are common and only a few training data instances would be considered to pose a heightened threat level. Several experiments are conducted to evaluate our proposed classifier. These experiments use several databases from the UCI data repository and data sets collected from the airport terminal simulation platform developed at the Distributed Decision Environments (DDE) Laboratory at the Department of Electrical and Computer Engineering, University of Miami. The experimental results show that, while the proposed classifier's performance is comparable to some existing classifiers when the databases have no class label ambiguities, it provides superior classification accuracy and better efficiency when class label ambiguities are present.

UR - http://www.scopus.com/inward/record.url?scp=51349112429&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=51349112429&partnerID=8YFLogxK

U2 - 10.1007/978-3-540-78488-3_32

DO - 10.1007/978-3-540-78488-3_32

M3 - Chapter

SN - 9783540784876

VL - 118

T3 - Studies in Computational Intelligence

SP - 539

EP - 562

BT - Studies in Computational Intelligence

ER -