A linear model based on Kalman filter for improving neural network classification performance

Joko Siswantoro, Anton Satria Prabuwono, Azizi Abdullah, Bahari Idrus

Research output: Contribution to journalArticle

20 Citations (Scopus)

Abstract

Neural network has been applied in several classification problems such as in medical diagnosis, handwriting recognition, and product inspection, with a good classification performance. The performance of a neural network is characterized by the neural network's structure, transfer function, and learning algorithm. However, a neural network classifier tends to be weak if it uses an inappropriate structure. The neural network's structure depends on the complexity of the relationship between the input and the output. There are no exact rules that can be used to determine the neural network's structure. Therefore, studies in improving neural network classification performance without changing the neural network's structure is a challenging issue. This paper proposes a method to improve neural network classification performance by constructing a linear model based on the Kalman filter as a post processing. The linear model transforms the predicted output of the neural network to a value close to the desired output by using the linear combination of the object features and the predicted output. This simple transformation will reduce the error of neural network and improve classification performance. The Kalman filter iteration is used to estimate the parameters of the linear model. Five datasets from various domains with various characteristics, such as attribute types, the number of attributes, the number of samples, and the number of classes, were used for empirical validation. The validation results show that the linear model based on the Kalman filter can improve the performance of the original neural network.

Original languageEnglish
Pages (from-to)112-122
Number of pages11
JournalExpert Systems with Applications
Volume49
DOIs
Publication statusPublished - 1 May 2016

Fingerprint

Kalman filters
Neural networks
Learning algorithms
Transfer functions
Classifiers
Inspection

Keywords

  • Classification performance
  • Kalman filter
  • Linear model
  • Neural network

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Engineering(all)

Cite this

A linear model based on Kalman filter for improving neural network classification performance. / Siswantoro, Joko; Prabuwono, Anton Satria; Abdullah, Azizi; Idrus, Bahari.

In: Expert Systems with Applications, Vol. 49, 01.05.2016, p. 112-122.

Research output: Contribution to journalArticle

@article{737df8f1429d48ddb65cbdcd713fdb02,
title = "A linear model based on Kalman filter for improving neural network classification performance",
abstract = "Neural network has been applied in several classification problems such as in medical diagnosis, handwriting recognition, and product inspection, with a good classification performance. The performance of a neural network is characterized by the neural network's structure, transfer function, and learning algorithm. However, a neural network classifier tends to be weak if it uses an inappropriate structure. The neural network's structure depends on the complexity of the relationship between the input and the output. There are no exact rules that can be used to determine the neural network's structure. Therefore, studies in improving neural network classification performance without changing the neural network's structure is a challenging issue. This paper proposes a method to improve neural network classification performance by constructing a linear model based on the Kalman filter as a post processing. The linear model transforms the predicted output of the neural network to a value close to the desired output by using the linear combination of the object features and the predicted output. This simple transformation will reduce the error of neural network and improve classification performance. The Kalman filter iteration is used to estimate the parameters of the linear model. Five datasets from various domains with various characteristics, such as attribute types, the number of attributes, the number of samples, and the number of classes, were used for empirical validation. The validation results show that the linear model based on the Kalman filter can improve the performance of the original neural network.",
keywords = "Classification performance, Kalman filter, Linear model, Neural network",
author = "Joko Siswantoro and Prabuwono, {Anton Satria} and Azizi Abdullah and Bahari Idrus",
year = "2016",
month = "5",
day = "1",
doi = "10.1016/j.eswa.2015.12.012",
language = "English",
volume = "49",
pages = "112--122",
journal = "Expert Systems with Applications",
issn = "0957-4174",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - A linear model based on Kalman filter for improving neural network classification performance

AU - Siswantoro, Joko

AU - Prabuwono, Anton Satria

AU - Abdullah, Azizi

AU - Idrus, Bahari

PY - 2016/5/1

Y1 - 2016/5/1

N2 - Neural network has been applied in several classification problems such as in medical diagnosis, handwriting recognition, and product inspection, with a good classification performance. The performance of a neural network is characterized by the neural network's structure, transfer function, and learning algorithm. However, a neural network classifier tends to be weak if it uses an inappropriate structure. The neural network's structure depends on the complexity of the relationship between the input and the output. There are no exact rules that can be used to determine the neural network's structure. Therefore, studies in improving neural network classification performance without changing the neural network's structure is a challenging issue. This paper proposes a method to improve neural network classification performance by constructing a linear model based on the Kalman filter as a post processing. The linear model transforms the predicted output of the neural network to a value close to the desired output by using the linear combination of the object features and the predicted output. This simple transformation will reduce the error of neural network and improve classification performance. The Kalman filter iteration is used to estimate the parameters of the linear model. Five datasets from various domains with various characteristics, such as attribute types, the number of attributes, the number of samples, and the number of classes, were used for empirical validation. The validation results show that the linear model based on the Kalman filter can improve the performance of the original neural network.

AB - Neural network has been applied in several classification problems such as in medical diagnosis, handwriting recognition, and product inspection, with a good classification performance. The performance of a neural network is characterized by the neural network's structure, transfer function, and learning algorithm. However, a neural network classifier tends to be weak if it uses an inappropriate structure. The neural network's structure depends on the complexity of the relationship between the input and the output. There are no exact rules that can be used to determine the neural network's structure. Therefore, studies in improving neural network classification performance without changing the neural network's structure is a challenging issue. This paper proposes a method to improve neural network classification performance by constructing a linear model based on the Kalman filter as a post processing. The linear model transforms the predicted output of the neural network to a value close to the desired output by using the linear combination of the object features and the predicted output. This simple transformation will reduce the error of neural network and improve classification performance. The Kalman filter iteration is used to estimate the parameters of the linear model. Five datasets from various domains with various characteristics, such as attribute types, the number of attributes, the number of samples, and the number of classes, were used for empirical validation. The validation results show that the linear model based on the Kalman filter can improve the performance of the original neural network.

KW - Classification performance

KW - Kalman filter

KW - Linear model

KW - Neural network

UR - http://www.scopus.com/inward/record.url?scp=84953329545&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84953329545&partnerID=8YFLogxK

U2 - 10.1016/j.eswa.2015.12.012

DO - 10.1016/j.eswa.2015.12.012

M3 - Article

AN - SCOPUS:84953329545

VL - 49

SP - 112

EP - 122

JO - Expert Systems with Applications

JF - Expert Systems with Applications

SN - 0957-4174

ER -