Extraction of comprehensible logical rules from neural networks: application of TREPAN in bio and cheminformatics

Brian Hudson; David Whitley; A. Browne; M. Ford

Extraction of comprehensible logical rules from neural networks: application of TREPAN in bio and cheminformatics

Brian Hudson, David Whitley, A. Browne, M. Ford

School of Pharmacy & Biomedical Sciences

Research output: Contribution to journal › Article › peer-review

41 Downloads (Pure)

Abstract

TREPAN is an algorithm for the extraction of comprehensible rules from trained neural networks. The method has been applied successfully to biological sequence (bioinformatics) problems. It has now been extended to handle chemoinformatics (QSAR) datasets. The method has been shown to have advantages over traditional symbolic rule induction methods such as C5. Results obtained for bioinformatics and chemoinformatics problems using the TREPAN algorithm are presented.

Original language	English
Pages (from-to)	557-561
Number of pages	5
Journal	Croatica Chemica Acta
Volume	78
Issue number	4
Publication status	Published - 2005

Access to Document

CCA_78_2005_557_561_hudson_1.pdfFinal published version, 147 KB

Cite this

@article{d659977756d1410f919651b35e84a471,

title = "Extraction of comprehensible logical rules from neural networks: application of TREPAN in bio and cheminformatics",

abstract = "TREPAN is an algorithm for the extraction of comprehensible rules from trained neural networks. The method has been applied successfully to biological sequence (bioinformatics) problems. It has now been extended to handle chemoinformatics (QSAR) datasets. The method has been shown to have advantages over traditional symbolic rule induction methods such as C5. Results obtained for bioinformatics and chemoinformatics problems using the TREPAN algorithm are presented.",

author = "Brian Hudson and David Whitley and A. Browne and M. Ford",

year = "2005",

language = "English",

volume = "78",

pages = "557--561",

journal = "Croatica Chemica Acta",

issn = "0011-1643",

publisher = "Croatian Chemical Society",

number = "4",

}

TY - JOUR

T1 - Extraction of comprehensible logical rules from neural networks: application of TREPAN in bio and cheminformatics

AU - Hudson, Brian

AU - Whitley, David

AU - Browne, A.

AU - Ford, M.

PY - 2005

Y1 - 2005

N2 - TREPAN is an algorithm for the extraction of comprehensible rules from trained neural networks. The method has been applied successfully to biological sequence (bioinformatics) problems. It has now been extended to handle chemoinformatics (QSAR) datasets. The method has been shown to have advantages over traditional symbolic rule induction methods such as C5. Results obtained for bioinformatics and chemoinformatics problems using the TREPAN algorithm are presented.

AB - TREPAN is an algorithm for the extraction of comprehensible rules from trained neural networks. The method has been applied successfully to biological sequence (bioinformatics) problems. It has now been extended to handle chemoinformatics (QSAR) datasets. The method has been shown to have advantages over traditional symbolic rule induction methods such as C5. Results obtained for bioinformatics and chemoinformatics problems using the TREPAN algorithm are presented.

M3 - Article

SN - 0011-1643

VL - 78

SP - 557

EP - 561

JO - Croatica Chemica Acta

JF - Croatica Chemica Acta

IS - 4

ER -

Extraction of comprehensible logical rules from neural networks: application of TREPAN in bio and cheminformatics

Abstract

Access to Document

Fingerprint

Cite this