Analyzing divisia rules extracted from a feedforward neural network

Vincent A. Schmidt; Jane M. Binner

Analyzing divisia rules extracted from a feedforward neural network

Vincent A. Schmidt^*, Jane M. Binner

^*Corresponding author for this work

Economics, Finance and Entrepreneurship

Research output: Chapter in Book/Published conference output › Conference publication

Abstract

This paper introduces a mechanism for generating a series of rules that characterize the money-price relationship, defined as the relationship between the rate of growth of the money supply and inflation. Divisia component data is used to train a selection of candidate feedforward neural networks. The selected network is mined for rules, expressed in human-readable and machine-executable form. The rule and network accuracy are compared, and expert commentary is made on the readability and reliability of the extracted rule set. The ultimate goal of this research is to produce rules that meaningfully and accurately describe inflation in terms of the Divisia component dataset.

Original language	English
Title of host publication	Proceedings of the 2006 International Conference on Artificial Intelligence, ICAI'06
Pages	127-133
Number of pages	7
Volume	1
Publication status	Published - 1 Dec 2006
Event	2006 International Conference on Artificial Intelligence, ICAI'06 - Las Vegas, NV, United Kingdom Duration: 26 Jun 2006 → 29 Jun 2006

Conference

Conference	2006 International Conference on Artificial Intelligence, ICAI'06
Country/Territory	United Kingdom
City	Las Vegas, NV
Period	26/06/06 → 29/06/06

Keywords

Data mining
Divisia
Inflation
Neural network
Rule generation

Cite this

@inproceedings{498765a692174ba49bd94f1e57252c1f,

title = "Analyzing divisia rules extracted from a feedforward neural network",

abstract = "This paper introduces a mechanism for generating a series of rules that characterize the money-price relationship, defined as the relationship between the rate of growth of the money supply and inflation. Divisia component data is used to train a selection of candidate feedforward neural networks. The selected network is mined for rules, expressed in human-readable and machine-executable form. The rule and network accuracy are compared, and expert commentary is made on the readability and reliability of the extracted rule set. The ultimate goal of this research is to produce rules that meaningfully and accurately describe inflation in terms of the Divisia component dataset.",

keywords = "Data mining, Divisia, Inflation, Neural network, Rule generation",

author = "Schmidt, {Vincent A.} and Binner, {Jane M.}",

year = "2006",

month = dec,

day = "1",

language = "English",

isbn = "9781932415988",

volume = "1",

pages = "127--133",

booktitle = "Proceedings of the 2006 International Conference on Artificial Intelligence, ICAI'06",

note = "2006 International Conference on Artificial Intelligence, ICAI'06 ; Conference date: 26-06-2006 Through 29-06-2006",

}

TY - GEN

T1 - Analyzing divisia rules extracted from a feedforward neural network

AU - Schmidt, Vincent A.

AU - Binner, Jane M.

PY - 2006/12/1

Y1 - 2006/12/1

N2 - This paper introduces a mechanism for generating a series of rules that characterize the money-price relationship, defined as the relationship between the rate of growth of the money supply and inflation. Divisia component data is used to train a selection of candidate feedforward neural networks. The selected network is mined for rules, expressed in human-readable and machine-executable form. The rule and network accuracy are compared, and expert commentary is made on the readability and reliability of the extracted rule set. The ultimate goal of this research is to produce rules that meaningfully and accurately describe inflation in terms of the Divisia component dataset.

AB - This paper introduces a mechanism for generating a series of rules that characterize the money-price relationship, defined as the relationship between the rate of growth of the money supply and inflation. Divisia component data is used to train a selection of candidate feedforward neural networks. The selected network is mined for rules, expressed in human-readable and machine-executable form. The rule and network accuracy are compared, and expert commentary is made on the readability and reliability of the extracted rule set. The ultimate goal of this research is to produce rules that meaningfully and accurately describe inflation in terms of the Divisia component dataset.

KW - Data mining

KW - Divisia

KW - Inflation

KW - Neural network

KW - Rule generation

UR - http://www.scopus.com/inward/record.url?scp=84866073203&partnerID=8YFLogxK

M3 - Conference publication

AN - SCOPUS:84866073203

SN - 9781932415988

VL - 1

SP - 127

EP - 133

BT - Proceedings of the 2006 International Conference on Artificial Intelligence, ICAI'06

T2 - 2006 International Conference on Artificial Intelligence, ICAI'06

Y2 - 26 June 2006 through 29 June 2006

ER -

Analyzing divisia rules extracted from a feedforward neural network

Abstract

Conference

Keywords

Other files and links

Fingerprint

Cite this