Dimensionality reduction via discretization

Huan Liu, Rudy Setiono

Research output: Contribution to journalArticle

27 Citations (Scopus)

Abstract

The existence of numeric data and large numbers of records in a database present a challenging task in terms of explicit concepts extraction from the raw data. The paper introduces a method that reduces data vertically and horizontally, keeps the discriminating power of the original data, and paves the way for extracting concepts. The method is based on discretization (vertical reduction) and feature selection (horizontal reduction). The experimental results show that (a) the data can be effectively reduced by the proposed method; (b) the predictive accuracy of a classifier (C4.5) can be improved after data and dimensionality reduction; and (c) the classification rules learned are simpler.

Original languageEnglish (US)
Pages (from-to)67-72
Number of pages6
JournalKnowledge-Based Systems
Volume9
Issue number1
DOIs
StatePublished - Feb 1996
Externally publishedYes

Fingerprint

Feature extraction
Classifiers
Dimensionality reduction
Discretization
Feature selection
Predictive accuracy
Data base
Classifier

Keywords

  • Dimensionality reduction
  • Discretization
  • Knowledge discovery

ASJC Scopus subject areas

  • Artificial Intelligence

Cite this

Dimensionality reduction via discretization. / Liu, Huan; Setiono, Rudy.

In: Knowledge-Based Systems, Vol. 9, No. 1, 02.1996, p. 67-72.

Research output: Contribution to journalArticle

Liu, Huan ; Setiono, Rudy. / Dimensionality reduction via discretization. In: Knowledge-Based Systems. 1996 ; Vol. 9, No. 1. pp. 67-72.
@article{7379678ccfea4ba6aaad32218073709a,
title = "Dimensionality reduction via discretization",
abstract = "The existence of numeric data and large numbers of records in a database present a challenging task in terms of explicit concepts extraction from the raw data. The paper introduces a method that reduces data vertically and horizontally, keeps the discriminating power of the original data, and paves the way for extracting concepts. The method is based on discretization (vertical reduction) and feature selection (horizontal reduction). The experimental results show that (a) the data can be effectively reduced by the proposed method; (b) the predictive accuracy of a classifier (C4.5) can be improved after data and dimensionality reduction; and (c) the classification rules learned are simpler.",
keywords = "Dimensionality reduction, Discretization, Knowledge discovery",
author = "Huan Liu and Rudy Setiono",
year = "1996",
month = "2",
doi = "10.1016/0950-7051(95)01030-0",
language = "English (US)",
volume = "9",
pages = "67--72",
journal = "Knowledge-Based Systems",
issn = "0950-7051",
publisher = "Elsevier",
number = "1",

}

TY - JOUR

T1 - Dimensionality reduction via discretization

AU - Liu, Huan

AU - Setiono, Rudy

PY - 1996/2

Y1 - 1996/2

N2 - The existence of numeric data and large numbers of records in a database present a challenging task in terms of explicit concepts extraction from the raw data. The paper introduces a method that reduces data vertically and horizontally, keeps the discriminating power of the original data, and paves the way for extracting concepts. The method is based on discretization (vertical reduction) and feature selection (horizontal reduction). The experimental results show that (a) the data can be effectively reduced by the proposed method; (b) the predictive accuracy of a classifier (C4.5) can be improved after data and dimensionality reduction; and (c) the classification rules learned are simpler.

AB - The existence of numeric data and large numbers of records in a database present a challenging task in terms of explicit concepts extraction from the raw data. The paper introduces a method that reduces data vertically and horizontally, keeps the discriminating power of the original data, and paves the way for extracting concepts. The method is based on discretization (vertical reduction) and feature selection (horizontal reduction). The experimental results show that (a) the data can be effectively reduced by the proposed method; (b) the predictive accuracy of a classifier (C4.5) can be improved after data and dimensionality reduction; and (c) the classification rules learned are simpler.

KW - Dimensionality reduction

KW - Discretization

KW - Knowledge discovery

UR - http://www.scopus.com/inward/record.url?scp=0030079092&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030079092&partnerID=8YFLogxK

U2 - 10.1016/0950-7051(95)01030-0

DO - 10.1016/0950-7051(95)01030-0

M3 - Article

AN - SCOPUS:0030079092

VL - 9

SP - 67

EP - 72

JO - Knowledge-Based Systems

JF - Knowledge-Based Systems

SN - 0950-7051

IS - 1

ER -