NEFCLASS for JAVA']JAVA - New learning algorithms

被引：18

作者：

Nauck, D ^{[1
]}

Nauck, U ^{[1
]}

Kruse, R ^{[1
]}

机构：

[1] Univ Magdeburg, Fac Comp Sci, IWS, FIN, D-39106 Magdeburg, Germany

来源：

18TH INTERNATIONAL CONFERENCE OF THE NORTH AMERICAN FUZZY INFORMATION PROCESSING SOCIETY - NAFIPS | 1999年

关键词：

D O I：

10.1109/NAFIPS.1999.781738

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Neuro-fuzzy classification approaches aim at creating fuzzy classification rules from data. Our neuro-fuzzy model NEFCLASS is able to learn fuzzy rules and fuzzy sets by simple heuristics. The aim of NEFCLASS is to quickly create interpretable fuzzy classifiers. In this paper we present NEFCLASS-J - a new version of our approach that was written in JAVA and contains some additions to the learning algorithms, like the treatment of missing values, the ability to use symbolic data, to automatically determine the size of the rule base, and a new automatic pruning strategy.

引用

页码：472 / 476

页数：5

共 14 条

[1]

BERTHOLD M, 1997, P 7 INT FUZZ SYST AS, V1, P359

[2] MAXIMUM LIKELIHOOD FROM INCOMPLETE DATA VIA EM ALGORITHM [J].

DEMPSTER, AP ;

LAIRD, NM ;

RUBIN, DB .

JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-METHODOLOGICAL, 1977, 39 (01) :1-38

[3]

Mitchell TM., 1997, MACH LEARN, V1

[4] NEFCLASS-X - a soft computing tool to build readable fuzzy classifiers [J].

Nauck, D ;

Kruse, R .

BT TECHNOLOGY JOURNAL, 1998, 16 (03) :180-190

[5] A neuro-fuzzy method to learn fuzzy classification rules from data [J].

Nauck, D ;

Kruse, R .

FUZZY SETS AND SYSTEMS, 1997, 89 (03) :277-288

[6]

Nauck D., 1997, P 7 INT FUZZ SYST AS, VIV, P50

[7]

NAUCK D, 1996, P BIENN C N AM FUZZ

[8]

NAUCK D, 1998, HDB FUZZY COMPUTATIO

[9]

NAUCK D, 1999, P 18 INT C N AM FUZZ

[10]

NAUCK U, 1999, THESIS TU BRAUNSCHWE

← 1 2 →