Design of supervised classifiers using Boolean neural networks

Srinivas Gazula, Mansur R. Kabuka

Research output: Contribution to journalArticle

26 Citations (Scopus)

Abstract

In this paper we present two supervised pattern classifiers designed using Boolean neural networks (BNN). They are 1) nearest-to-an-exemplar (NTE) and 2) Boolean k-nearest neighbor (BKNN) classifier. The emphasis during the design of these classifiers was on simplicity, robustness, and the ease of hardware implementation. The classifiers use the idea of radius of attraction (ROA) to achieve their goal. Mathematical analysis of the algorithms presented in the paper is done to prove their feasibility. Both classifiers are tested with well-known binary and continuous feature valued data sets yielding results comparable with those obtained by similar existing classifiers.

Original languageEnglish
Pages (from-to)1239-1246
Number of pages8
JournalIEEE Transactions on Pattern Analysis and Machine Intelligence
Volume17
Issue number12
DOIs
StatePublished - Dec 1 1995

Fingerprint

Boolean Networks
Classifiers
Classifier
Neural Networks
Neural networks
Hardware Implementation
Mathematical Analysis
Design
Nearest Neighbor
Simplicity
Radius
Binary
Robustness
Hardware

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Cite this

Design of supervised classifiers using Boolean neural networks. / Gazula, Srinivas; Kabuka, Mansur R.

In: IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 17, No. 12, 01.12.1995, p. 1239-1246.

Research output: Contribution to journalArticle

@article{e14204851a0e42b3b67ac4f60443882d,
title = "Design of supervised classifiers using Boolean neural networks",
abstract = "In this paper we present two supervised pattern classifiers designed using Boolean neural networks (BNN). They are 1) nearest-to-an-exemplar (NTE) and 2) Boolean k-nearest neighbor (BKNN) classifier. The emphasis during the design of these classifiers was on simplicity, robustness, and the ease of hardware implementation. The classifiers use the idea of radius of attraction (ROA) to achieve their goal. Mathematical analysis of the algorithms presented in the paper is done to prove their feasibility. Both classifiers are tested with well-known binary and continuous feature valued data sets yielding results comparable with those obtained by similar existing classifiers.",
author = "Srinivas Gazula and Kabuka, {Mansur R.}",
year = "1995",
month = "12",
day = "1",
doi = "10.1109/34.476519",
language = "English",
volume = "17",
pages = "1239--1246",
journal = "IEEE Transactions on Pattern Analysis and Machine Intelligence",
issn = "0162-8828",
publisher = "IEEE Computer Society",
number = "12",

}

TY - JOUR

T1 - Design of supervised classifiers using Boolean neural networks

AU - Gazula, Srinivas

AU - Kabuka, Mansur R.

PY - 1995/12/1

Y1 - 1995/12/1

N2 - In this paper we present two supervised pattern classifiers designed using Boolean neural networks (BNN). They are 1) nearest-to-an-exemplar (NTE) and 2) Boolean k-nearest neighbor (BKNN) classifier. The emphasis during the design of these classifiers was on simplicity, robustness, and the ease of hardware implementation. The classifiers use the idea of radius of attraction (ROA) to achieve their goal. Mathematical analysis of the algorithms presented in the paper is done to prove their feasibility. Both classifiers are tested with well-known binary and continuous feature valued data sets yielding results comparable with those obtained by similar existing classifiers.

AB - In this paper we present two supervised pattern classifiers designed using Boolean neural networks (BNN). They are 1) nearest-to-an-exemplar (NTE) and 2) Boolean k-nearest neighbor (BKNN) classifier. The emphasis during the design of these classifiers was on simplicity, robustness, and the ease of hardware implementation. The classifiers use the idea of radius of attraction (ROA) to achieve their goal. Mathematical analysis of the algorithms presented in the paper is done to prove their feasibility. Both classifiers are tested with well-known binary and continuous feature valued data sets yielding results comparable with those obtained by similar existing classifiers.

UR - http://www.scopus.com/inward/record.url?scp=0029537753&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029537753&partnerID=8YFLogxK

U2 - 10.1109/34.476519

DO - 10.1109/34.476519

M3 - Article

AN - SCOPUS:0029537753

VL - 17

SP - 1239

EP - 1246

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

SN - 0162-8828

IS - 12

ER -