Combining analytic models with neural networks

T. Jan

doi:10.1109/ISSPIT.2003.1341193

Combining analytic models with neural networks

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

Abstract

In this paper, an ensemble of models is introduced which combines a linear parametric model and a nonlinear non-parametric model such as artificial neural network (ANN). This model aims to embody the desirable characteristics of linear parametric model such as stable generalization capability while retaining the data-based learning and prediction capacity of ANNs. The proposed model is applied for short term time series prediction and the results show that the proposed model achieves good generalization (prediction) performance utilizing the nonparametric ANN model component while achieving much improved stability utilizing the linear model component. The experiment compares the proposed model to other ANN models and linear models for generalization.

Original language	English
Title of host publication	Proceedings of the 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	605-608
Number of pages	4
ISBN (Electronic)	0780382927, 9780780382923
DOIs	https://doi.org/10.1109/ISSPIT.2003.1341193
Publication status	Published - 2003
Externally published	Yes
Event	3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003 - Darmstadt, Germany Duration: 14 Dec 2003 → 17 Dec 2003

Publication series

Name	Proceedings of the 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003

Conference

Conference	3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003
Country/Territory	Germany
City	Darmstadt
Period	14/12/03 → 17/12/03

Keywords

Artificial neural networks
Computational modeling
Computer networks
Ear
Multilayer perceptrons
Neural networks
Parametric statistics
Predictive models
Stability
Training data

Access to Document

10.1109/ISSPIT.2003.1341193

Cite this

Jan, T. (2003). Combining analytic models with neural networks. In Proceedings of the 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003 (pp. 605-608). Article 1341193 (Proceedings of the 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ISSPIT.2003.1341193

@inproceedings{bff946af5341499cb2bc49693e39b2a8,

title = "Combining analytic models with neural networks",

abstract = "In this paper, an ensemble of models is introduced which combines a linear parametric model and a nonlinear non-parametric model such as artificial neural network (ANN). This model aims to embody the desirable characteristics of linear parametric model such as stable generalization capability while retaining the data-based learning and prediction capacity of ANNs. The proposed model is applied for short term time series prediction and the results show that the proposed model achieves good generalization (prediction) performance utilizing the nonparametric ANN model component while achieving much improved stability utilizing the linear model component. The experiment compares the proposed model to other ANN models and linear models for generalization.",

keywords = "Artificial neural networks, Computational modeling, Computer networks, Ear, Multilayer perceptrons, Neural networks, Parametric statistics, Predictive models, Stability, Training data",

author = "T. Jan",

note = "Publisher Copyright: {\textcopyright} 2003 IEEE.; 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003 ; Conference date: 14-12-2003 Through 17-12-2003",

year = "2003",

doi = "10.1109/ISSPIT.2003.1341193",

language = "English",

series = "Proceedings of the 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "605--608",

booktitle = "Proceedings of the 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003",

}

Jan, T 2003, Combining analytic models with neural networks. in Proceedings of the 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003., 1341193, Proceedings of the 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003, Institute of Electrical and Electronics Engineers Inc., pp. 605-608, 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003, Darmstadt, Germany, 14/12/03. https://doi.org/10.1109/ISSPIT.2003.1341193

Combining analytic models with neural networks. / Jan, T.
Proceedings of the 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003. Institute of Electrical and Electronics Engineers Inc., 2003. p. 605-608 1341193 (Proceedings of the 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003).

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

TY - GEN

T1 - Combining analytic models with neural networks

AU - Jan, T.

PY - 2003

Y1 - 2003

N2 - In this paper, an ensemble of models is introduced which combines a linear parametric model and a nonlinear non-parametric model such as artificial neural network (ANN). This model aims to embody the desirable characteristics of linear parametric model such as stable generalization capability while retaining the data-based learning and prediction capacity of ANNs. The proposed model is applied for short term time series prediction and the results show that the proposed model achieves good generalization (prediction) performance utilizing the nonparametric ANN model component while achieving much improved stability utilizing the linear model component. The experiment compares the proposed model to other ANN models and linear models for generalization.

AB - In this paper, an ensemble of models is introduced which combines a linear parametric model and a nonlinear non-parametric model such as artificial neural network (ANN). This model aims to embody the desirable characteristics of linear parametric model such as stable generalization capability while retaining the data-based learning and prediction capacity of ANNs. The proposed model is applied for short term time series prediction and the results show that the proposed model achieves good generalization (prediction) performance utilizing the nonparametric ANN model component while achieving much improved stability utilizing the linear model component. The experiment compares the proposed model to other ANN models and linear models for generalization.

KW - Artificial neural networks

KW - Computational modeling

KW - Computer networks

KW - Ear

KW - Multilayer perceptrons

KW - Neural networks

KW - Parametric statistics

KW - Predictive models

KW - Stability

KW - Training data

UR - http://www.scopus.com/inward/record.url?scp=84948146339&partnerID=8YFLogxK

U2 - 10.1109/ISSPIT.2003.1341193

DO - 10.1109/ISSPIT.2003.1341193

M3 - Conference contribution

AN - SCOPUS:84948146339

T3 - Proceedings of the 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003

SP - 605

EP - 608

BT - Proceedings of the 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003

Y2 - 14 December 2003 through 17 December 2003

ER -

Jan T. Combining analytic models with neural networks. In Proceedings of the 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003. Institute of Electrical and Electronics Engineers Inc. 2003. p. 605-608. 1341193. (Proceedings of the 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003). doi: 10.1109/ISSPIT.2003.1341193

Combining analytic models with neural networks

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this