Extracting rules from DRASiW's "mental images"

Paulo Coutinho; Hugo C. C. Carneiro; Danilo S. Carvalho; Felipe M. G. França

Extracting rules from DRASiW's "mental images"

Paulo Coutinho
, Hugo C. C. Carneiro
, Danilo S. Carvalho
, Felipe M. G. França

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

Abstract

DRASiW is an extension of the WiSARD weightless neural model that provides the ability of producing examples/prototypes, called “mental images”, from learnt categories. This work introduces a novel way of performing rule extraction by applying the WiSARD/DRASiW RAM-based neural model upon a well-known machine learning benchmark. A functional exploration is offered in order to demonstrate how the new rule extraction mechanism behaves under different system configurations. Experimental results suggest that the rules conformance to data increases proportionally to the corresponding classifier accuracy. Furthermore, comparison with C4.5 decision tree algorithm shows that the DRASiW-based technique produces more compact sets of rules.

Original language	English
Title of host publication	ESANN 2014 - Proceedings
Subtitle of host publication	22st European Symposium on Artificial Neural Networks, Computational Intelligence And Machine Learning, Bruges, April 23-24-25, 2014
Place of Publication	Bruges
Publisher	European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning
Pages	529-534
Number of pages	6
ISBN (Electronic)	9782874190964
ISBN (Print)	9782874190957
Publication status	Published - 17 Mar 2014
Event	22nd European Symposium on Artificial Neural Networks - Bruges, Belgium Duration: 23 Apr 2014 → 25 Apr 2014

Conference

Conference	22nd European Symposium on Artificial Neural Networks
Abbreviated title	ESANN 2014
Country/Territory	Belgium
City	Bruges
Period	23/04/14 → 25/04/14

Access to Document

https://www.esann.org/sites/default/files/proceedings/legacy/es2014-167.pdf

Cite this

Coutinho, P., Carneiro, H. C. C., Carvalho, D. S., & França, F. M. G. (2014). Extracting rules from DRASiW's "mental images". In ESANN 2014 - Proceedings: 22st European Symposium on Artificial Neural Networks, Computational Intelligence And Machine Learning, Bruges, April 23-24-25, 2014 (pp. 529-534). Article 167 European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. https://www.esann.org/sites/default/files/proceedings/legacy/es2014-167.pdf

Coutinho, Paulo ; Carneiro, Hugo C. C. ; Carvalho, Danilo S. et al. / Extracting rules from DRASiW's "mental images". ESANN 2014 - Proceedings: 22st European Symposium on Artificial Neural Networks, Computational Intelligence And Machine Learning, Bruges, April 23-24-25, 2014. Bruges : European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, 2014. pp. 529-534

@inproceedings{7efded8e5aab4d99bf789340b54d5138,

title = "Extracting rules from DRASiW's {"}mental images{"}",

abstract = "DRASiW is an extension of the WiSARD weightless neural model that provides the ability of producing examples/prototypes, called {\textquotedblleft}mental images{\textquotedblright}, from learnt categories. This work introduces a novel way of performing rule extraction by applying the WiSARD/DRASiW RAM-based neural model upon a well-known machine learning benchmark. A functional exploration is offered in order to demonstrate how the new rule extraction mechanism behaves under different system configurations. Experimental results suggest that the rules conformance to data increases proportionally to the corresponding classifier accuracy. Furthermore, comparison with C4.5 decision tree algorithm shows that the DRASiW-based technique produces more compact sets of rules.",

author = "Paulo Coutinho and Carneiro, \{Hugo C. C.\} and Carvalho, \{Danilo S.\} and Fran{\c c}a, \{Felipe M. G.\}",

year = "2014",

month = mar,

day = "17",

language = "English",

isbn = "9782874190957",

pages = "529--534",

booktitle = "ESANN 2014 - Proceedings",

publisher = "European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning",

address = "Belgium",

note = "22nd European Symposium on Artificial Neural Networks, ESANN 2014 ; Conference date: 23-04-2014 Through 25-04-2014",

}

Coutinho, P, Carneiro, HCC, Carvalho, DS & França, FMG 2014, Extracting rules from DRASiW's "mental images". in ESANN 2014 - Proceedings: 22st European Symposium on Artificial Neural Networks, Computational Intelligence And Machine Learning, Bruges, April 23-24-25, 2014., 167, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, Bruges, pp. 529-534, 22nd European Symposium on Artificial Neural Networks, Bruges, Belgium, 23/04/14. <https://www.esann.org/sites/default/files/proceedings/legacy/es2014-167.pdf>

Extracting rules from DRASiW's "mental images". / Coutinho, Paulo; Carneiro, Hugo C. C.; Carvalho, Danilo S. et al.
ESANN 2014 - Proceedings: 22st European Symposium on Artificial Neural Networks, Computational Intelligence And Machine Learning, Bruges, April 23-24-25, 2014. Bruges: European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, 2014. p. 529-534 167.

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Extracting rules from DRASiW's "mental images"

AU - Coutinho, Paulo

AU - Carneiro, Hugo C. C.

AU - Carvalho, Danilo S.

AU - França, Felipe M. G.

PY - 2014/3/17

Y1 - 2014/3/17

N2 - DRASiW is an extension of the WiSARD weightless neural model that provides the ability of producing examples/prototypes, called “mental images”, from learnt categories. This work introduces a novel way of performing rule extraction by applying the WiSARD/DRASiW RAM-based neural model upon a well-known machine learning benchmark. A functional exploration is offered in order to demonstrate how the new rule extraction mechanism behaves under different system configurations. Experimental results suggest that the rules conformance to data increases proportionally to the corresponding classifier accuracy. Furthermore, comparison with C4.5 decision tree algorithm shows that the DRASiW-based technique produces more compact sets of rules.

AB - DRASiW is an extension of the WiSARD weightless neural model that provides the ability of producing examples/prototypes, called “mental images”, from learnt categories. This work introduces a novel way of performing rule extraction by applying the WiSARD/DRASiW RAM-based neural model upon a well-known machine learning benchmark. A functional exploration is offered in order to demonstrate how the new rule extraction mechanism behaves under different system configurations. Experimental results suggest that the rules conformance to data increases proportionally to the corresponding classifier accuracy. Furthermore, comparison with C4.5 decision tree algorithm shows that the DRASiW-based technique produces more compact sets of rules.

M3 - Conference contribution

SN - 9782874190957

SP - 529

EP - 534

BT - ESANN 2014 - Proceedings

PB - European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning

CY - Bruges

T2 - 22nd European Symposium on Artificial Neural Networks

Y2 - 23 April 2014 through 25 April 2014

ER -

Coutinho P, Carneiro HCC, Carvalho DS, França FMG. Extracting rules from DRASiW's "mental images". In ESANN 2014 - Proceedings: 22st European Symposium on Artificial Neural Networks, Computational Intelligence And Machine Learning, Bruges, April 23-24-25, 2014. Bruges: European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. 2014. p. 529-534. 167

Extracting rules from DRASiW's "mental images"

Abstract

Conference

Access to Document

Fingerprint

Cite this