Process identification through modular neural networks and rule extraction (extended abstract)

B.J. van der Zwaag; Cornelis H. Slump; L. Spaanenburg

Process identification through modular neural networks and rule extraction (extended abstract)

B.J. van der Zwaag
, Cornelis H. Slump
, L. Spaanenburg

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic

91 Downloads (Pure)

Abstract

Monolithic neural networks may be trained from measured data to establish knowledge about the process. Unfortunately, this knowledge is not guaranteed to be found and – if at all – hard to extract. Modular neural networks are better suited for this purpose. Domain-ordered by topology, rule extraction is performed module by module. This has all the benefits of a divide-and-conquer method and opens the way to structured design. This paper discusses a next step in this direction by illustrating the potential of base functions to design the neural model.

Original language	Undefined
Title of host publication	Proceedings of the Fourteenth Belgium/Netherlands Conference on Artificial Intelligence (BNAIC'02)
Editors	Hendrik Blockeel, Marc Denecker
Place of Publication	Leuven, Belgium
Pages	507-508
Number of pages	2
Publication status	Published - Oct 2002
Event	14th Belgium-Dutch Conference on Artificial Intelligence, BNAIC 2002 - Leuven, Belgium Duration: 21 Oct 2002 → 22 Oct 2002 Conference number: 14

Conference

Conference	14th Belgium-Dutch Conference on Artificial Intelligence, BNAIC 2002
Abbreviated title	BNAIC
Country/Territory	Belgium
City	Leuven
Period	21/10/02 → 22/10/02

Keywords

EWI-1773
METIS-207392
IR-43784

Access to Document

Cite this

@inproceedings{ebc659270d88430f86cc6742151faf3b,

title = "Process identification through modular neural networks and rule extraction (extended abstract)",

abstract = "Monolithic neural networks may be trained from measured data to establish knowledge about the process. Unfortunately, this knowledge is not guaranteed to be found and {\textendash} if at all {\textendash} hard to extract. Modular neural networks are better suited for this purpose. Domain-ordered by topology, rule extraction is performed module by module. This has all the benefits of a divide-and-conquer method and opens the way to structured design. This paper discusses a next step in this direction by illustrating the potential of base functions to design the neural model.",

keywords = "EWI-1773, METIS-207392, IR-43784",

author = "\{van der Zwaag\}, B.J. and Slump, \{Cornelis H.\} and L. Spaanenburg",

note = "056.02 ; 14th Belgium-Dutch Conference on Artificial Intelligence, BNAIC 2002 ; Conference date: 21-10-2002 Through 22-10-2002",

year = "2002",

month = oct,

language = "Undefined",

pages = "507--508",

editor = "Hendrik Blockeel and Marc Denecker",

booktitle = "Proceedings of the Fourteenth Belgium/Netherlands Conference on Artificial Intelligence (BNAIC'02)",

}

van der Zwaag, BJ, Slump, CH & Spaanenburg, L 2002, Process identification through modular neural networks and rule extraction (extended abstract). in H Blockeel & M Denecker (eds), Proceedings of the Fourteenth Belgium/Netherlands Conference on Artificial Intelligence (BNAIC'02). Leuven, Belgium, pp. 507-508, 14th Belgium-Dutch Conference on Artificial Intelligence, BNAIC 2002, Leuven, Belgium, 21/10/02.

Process identification through modular neural networks and rule extraction (extended abstract). / van der Zwaag, B.J.; Slump, Cornelis H.; Spaanenburg, L.
Proceedings of the Fourteenth Belgium/Netherlands Conference on Artificial Intelligence (BNAIC'02). ed. / Hendrik Blockeel; Marc Denecker. Leuven, Belgium, 2002. p. 507-508.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic

TY - GEN

T1 - Process identification through modular neural networks and rule extraction (extended abstract)

AU - van der Zwaag, B.J.

AU - Slump, Cornelis H.

AU - Spaanenburg, L.

N1 - Conference code: 14

PY - 2002/10

Y1 - 2002/10

N2 - Monolithic neural networks may be trained from measured data to establish knowledge about the process. Unfortunately, this knowledge is not guaranteed to be found and – if at all – hard to extract. Modular neural networks are better suited for this purpose. Domain-ordered by topology, rule extraction is performed module by module. This has all the benefits of a divide-and-conquer method and opens the way to structured design. This paper discusses a next step in this direction by illustrating the potential of base functions to design the neural model.

AB - Monolithic neural networks may be trained from measured data to establish knowledge about the process. Unfortunately, this knowledge is not guaranteed to be found and – if at all – hard to extract. Modular neural networks are better suited for this purpose. Domain-ordered by topology, rule extraction is performed module by module. This has all the benefits of a divide-and-conquer method and opens the way to structured design. This paper discusses a next step in this direction by illustrating the potential of base functions to design the neural model.

KW - EWI-1773

KW - METIS-207392

KW - IR-43784

M3 - Conference contribution

SP - 507

EP - 508

BT - Proceedings of the Fourteenth Belgium/Netherlands Conference on Artificial Intelligence (BNAIC'02)

A2 - Blockeel, Hendrik

A2 - Denecker, Marc

CY - Leuven, Belgium

T2 - 14th Belgium-Dutch Conference on Artificial Intelligence, BNAIC 2002

Y2 - 21 October 2002 through 22 October 2002

ER -