A top-down and greedy method for discretization of continuous attributes

Chien I. Lee, Cheng Jung Tsai, Ya Ru Yang, Wei Pang Yang

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

4 Citations (Scopus)

Abstract

Experiments show that CAIM discretization algorithm is superior to all the other top-down discretization algorithms. However, CAIM algorithm does not take the data distribution into account. The discretization formula used in CAIM also gives a high factor to the numbers of generated intervals. The two disadvantages make CAIM may generate irrational discrete results in some cases and further leads to the decrease of predictive accuracy of a classifier. In this paper we propose the Class-Attribute Contingency Coefficient discretization algorithm. The experimental results showed that compared with CAIM, our method can generate a better discretization scheme to bring on the improvement of accuracy of classification. With regard to the number of generated rules and execution time of a classifier, CACC and CAIM achieve comparable results.

Original languageEnglish
Title of host publicationProceedings - Fourth International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2007
Pages472-476
Number of pages5
DOIs
Publication statusPublished - 2007 Dec 1
Event4th International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2007 - Haikou, China
Duration: 2007 Aug 242007 Aug 27

Publication series

NameProceedings - Fourth International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2007
Volume1

Other

Other4th International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2007
CountryChina
CityHaikou
Period07-08-2407-08-27

Fingerprint

Discretization
Attribute
Classifiers
Classifier
Discretization Scheme
Data Distribution
Execution Time
Decrease
Interval
Experimental Results
Coefficient
Experiment
Experiments

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Software
  • Applied Mathematics
  • Theoretical Computer Science

Cite this

Lee, C. I., Tsai, C. J., Yang, Y. R., & Yang, W. P. (2007). A top-down and greedy method for discretization of continuous attributes. In Proceedings - Fourth International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2007 (pp. 472-476). [4405970] (Proceedings - Fourth International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2007; Vol. 1). https://doi.org/10.1109/FSKD.2007.129
Lee, Chien I. ; Tsai, Cheng Jung ; Yang, Ya Ru ; Yang, Wei Pang. / A top-down and greedy method for discretization of continuous attributes. Proceedings - Fourth International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2007. 2007. pp. 472-476 (Proceedings - Fourth International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2007).
@inproceedings{6880a9eccc85488fa2f987a50d3b665b,
title = "A top-down and greedy method for discretization of continuous attributes",
abstract = "Experiments show that CAIM discretization algorithm is superior to all the other top-down discretization algorithms. However, CAIM algorithm does not take the data distribution into account. The discretization formula used in CAIM also gives a high factor to the numbers of generated intervals. The two disadvantages make CAIM may generate irrational discrete results in some cases and further leads to the decrease of predictive accuracy of a classifier. In this paper we propose the Class-Attribute Contingency Coefficient discretization algorithm. The experimental results showed that compared with CAIM, our method can generate a better discretization scheme to bring on the improvement of accuracy of classification. With regard to the number of generated rules and execution time of a classifier, CACC and CAIM achieve comparable results.",
author = "Lee, {Chien I.} and Tsai, {Cheng Jung} and Yang, {Ya Ru} and Yang, {Wei Pang}",
year = "2007",
month = "12",
day = "1",
doi = "10.1109/FSKD.2007.129",
language = "English",
isbn = "0769528740",
series = "Proceedings - Fourth International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2007",
pages = "472--476",
booktitle = "Proceedings - Fourth International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2007",

}

Lee, CI, Tsai, CJ, Yang, YR & Yang, WP 2007, A top-down and greedy method for discretization of continuous attributes. in Proceedings - Fourth International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2007., 4405970, Proceedings - Fourth International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2007, vol. 1, pp. 472-476, 4th International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2007, Haikou, China, 07-08-24. https://doi.org/10.1109/FSKD.2007.129

A top-down and greedy method for discretization of continuous attributes. / Lee, Chien I.; Tsai, Cheng Jung; Yang, Ya Ru; Yang, Wei Pang.

Proceedings - Fourth International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2007. 2007. p. 472-476 4405970 (Proceedings - Fourth International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2007; Vol. 1).

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

TY - GEN

T1 - A top-down and greedy method for discretization of continuous attributes

AU - Lee, Chien I.

AU - Tsai, Cheng Jung

AU - Yang, Ya Ru

AU - Yang, Wei Pang

PY - 2007/12/1

Y1 - 2007/12/1

N2 - Experiments show that CAIM discretization algorithm is superior to all the other top-down discretization algorithms. However, CAIM algorithm does not take the data distribution into account. The discretization formula used in CAIM also gives a high factor to the numbers of generated intervals. The two disadvantages make CAIM may generate irrational discrete results in some cases and further leads to the decrease of predictive accuracy of a classifier. In this paper we propose the Class-Attribute Contingency Coefficient discretization algorithm. The experimental results showed that compared with CAIM, our method can generate a better discretization scheme to bring on the improvement of accuracy of classification. With regard to the number of generated rules and execution time of a classifier, CACC and CAIM achieve comparable results.

AB - Experiments show that CAIM discretization algorithm is superior to all the other top-down discretization algorithms. However, CAIM algorithm does not take the data distribution into account. The discretization formula used in CAIM also gives a high factor to the numbers of generated intervals. The two disadvantages make CAIM may generate irrational discrete results in some cases and further leads to the decrease of predictive accuracy of a classifier. In this paper we propose the Class-Attribute Contingency Coefficient discretization algorithm. The experimental results showed that compared with CAIM, our method can generate a better discretization scheme to bring on the improvement of accuracy of classification. With regard to the number of generated rules and execution time of a classifier, CACC and CAIM achieve comparable results.

UR - http://www.scopus.com/inward/record.url?scp=44049083520&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=44049083520&partnerID=8YFLogxK

U2 - 10.1109/FSKD.2007.129

DO - 10.1109/FSKD.2007.129

M3 - Conference contribution

AN - SCOPUS:44049083520

SN - 0769528740

SN - 9780769528748

T3 - Proceedings - Fourth International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2007

SP - 472

EP - 476

BT - Proceedings - Fourth International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2007

ER -

Lee CI, Tsai CJ, Yang YR, Yang WP. A top-down and greedy method for discretization of continuous attributes. In Proceedings - Fourth International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2007. 2007. p. 472-476. 4405970. (Proceedings - Fourth International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2007). https://doi.org/10.1109/FSKD.2007.129