Modelling consumer e-commerce adoption behaviour using neural networks: a comparison with logistic regression

Mohammad B. Naseri; Greg Elliott

Modelling consumer e-commerce adoption behaviour using neural networks: a comparison with logistic regression

Mohammad B. Naseri, Greg Elliott

Department of Management

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

12 Downloads (Pure)

Abstract

Prior studies have generally suggested that Artificial Neural Networks (ANNs) are superior to conventional statistical models in predicting consumer buying behavior. There are, however, contradicting findings which raise question over usefulness of ANNs. This paper discusses development of three neural networks for modeling consumer e-commerce behavior and compares the findings to equivalent logistic regression models. The results showed that ANNs predict e-commerce adoption slightly more accurately than logistic models but this is hardly justifiable given the added complexity. Further, ANNs seem to be highly adaptive, particularly when a small sample is coupled with a large number of nodes in hidden layers which, in turn, limits the neural networks’ generalisability.

Original language	English
Title of host publication	ANZMAC 2007
Subtitle of host publication	proceedings : 3Rs - reputation, responsibility and relevance
Editors	Maree Thyne, Kenneth R. Deans, Juergen Gnoth
Place of Publication	Dunedin, N.Z.
Publisher	University of Otago
Pages	3085-3089
Number of pages	5
ISBN (Print)	9781877156299
Publication status	Published - 2007
Event	Australian and New Zealand Marketing Academy Conference (2007) - Dunedin, New Zealand Duration: 3 Dec 2007 → 5 Dec 2007

Conference

Conference	Australian and New Zealand Marketing Academy Conference (2007)
City	Dunedin, New Zealand
Period	3/12/07 → 5/12/07

Access to Document

Author final versionAccepted author manuscript, 102 KBLicence: Unspecified

http://conferences.anzmac.org/ANZMAC2007/papers/M%20Naseri_2a.pdf

Cite this

Naseri, M. B., & Elliott, G. (2007). Modelling consumer e-commerce adoption behaviour using neural networks: a comparison with logistic regression. In M. Thyne, K. R. Deans, & J. Gnoth (Eds.), ANZMAC 2007: proceedings : 3Rs - reputation, responsibility and relevance (pp. 3085-3089). University of Otago. http://conferences.anzmac.org/ANZMAC2007/papers/M%20Naseri_2a.pdf

@inproceedings{c1555d2b020b4ddc8e60b11daadaf667,

title = "Modelling consumer e-commerce adoption behaviour using neural networks: a comparison with logistic regression",

abstract = "Prior studies have generally suggested that Artificial Neural Networks (ANNs) are superior to conventional statistical models in predicting consumer buying behavior. There are, however, contradicting findings which raise question over usefulness of ANNs. This paper discusses development of three neural networks for modeling consumer e-commerce behavior and compares the findings to equivalent logistic regression models. The results showed that ANNs predict e-commerce adoption slightly more accurately than logistic models but this is hardly justifiable given the added complexity. Further, ANNs seem to be highly adaptive, particularly when a small sample is coupled with a large number of nodes in hidden layers which, in turn, limits the neural networks{\textquoteright} generalisability.",

author = "Naseri, {Mohammad B.} and Greg Elliott",

year = "2007",

language = "English",

isbn = "9781877156299",

pages = "3085--3089",

editor = "Maree Thyne and Deans, {Kenneth R.} and Juergen Gnoth",

booktitle = "ANZMAC 2007",

publisher = "University of Otago",

note = "Australian and New Zealand Marketing Academy Conference (2007) ; Conference date: 03-12-2007 Through 05-12-2007",

}

Naseri, MB & Elliott, G 2007, Modelling consumer e-commerce adoption behaviour using neural networks: a comparison with logistic regression. in M Thyne, KR Deans & J Gnoth (eds), ANZMAC 2007: proceedings : 3Rs - reputation, responsibility and relevance. University of Otago, Dunedin, N.Z., pp. 3085-3089, Australian and New Zealand Marketing Academy Conference (2007), Dunedin, New Zealand, 3/12/07. <http://conferences.anzmac.org/ANZMAC2007/papers/M%20Naseri_2a.pdf>

Modelling consumer e-commerce adoption behaviour using neural networks : a comparison with logistic regression. / Naseri, Mohammad B.; Elliott, Greg.

ANZMAC 2007: proceedings : 3Rs - reputation, responsibility and relevance. ed. / Maree Thyne; Kenneth R. Deans; Juergen Gnoth. Dunedin, N.Z. : University of Otago, 2007. p. 3085-3089.

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

TY - GEN

T1 - Modelling consumer e-commerce adoption behaviour using neural networks

T2 - Australian and New Zealand Marketing Academy Conference (2007)

AU - Naseri, Mohammad B.

AU - Elliott, Greg

PY - 2007

Y1 - 2007

N2 - Prior studies have generally suggested that Artificial Neural Networks (ANNs) are superior to conventional statistical models in predicting consumer buying behavior. There are, however, contradicting findings which raise question over usefulness of ANNs. This paper discusses development of three neural networks for modeling consumer e-commerce behavior and compares the findings to equivalent logistic regression models. The results showed that ANNs predict e-commerce adoption slightly more accurately than logistic models but this is hardly justifiable given the added complexity. Further, ANNs seem to be highly adaptive, particularly when a small sample is coupled with a large number of nodes in hidden layers which, in turn, limits the neural networks’ generalisability.

AB - Prior studies have generally suggested that Artificial Neural Networks (ANNs) are superior to conventional statistical models in predicting consumer buying behavior. There are, however, contradicting findings which raise question over usefulness of ANNs. This paper discusses development of three neural networks for modeling consumer e-commerce behavior and compares the findings to equivalent logistic regression models. The results showed that ANNs predict e-commerce adoption slightly more accurately than logistic models but this is hardly justifiable given the added complexity. Further, ANNs seem to be highly adaptive, particularly when a small sample is coupled with a large number of nodes in hidden layers which, in turn, limits the neural networks’ generalisability.

M3 - Conference proceeding contribution

SN - 9781877156299

SP - 3085

EP - 3089

BT - ANZMAC 2007

A2 - Thyne, Maree

A2 - Deans, Kenneth R.

A2 - Gnoth, Juergen

PB - University of Otago

CY - Dunedin, N.Z.

Y2 - 3 December 2007 through 5 December 2007

ER -

Modelling consumer e-commerce adoption behaviour using neural networks: a comparison with logistic regression

Abstract

Conference

Access to Document

Fingerprint

Cite this