Materials discovery via topologically-correct display of reduced-dimension data

被引:4
作者
Pao, YH [1 ]
Meng, Z
LeClair, S
Igelnik, B
机构
[1] Case Western Reserve Univ, Cleveland, OH 44106 USA
[2] AI WARE Inc, Cleveland, OH 44106 USA
[3] USAF, Res Lab, Mat & Mfg Directorate, WPAFB, OH USA
关键词
dimension reduction; materials design; materials property prediction; neural networks;
D O I
10.1016/S0925-8388(98)00608-2
中图分类号
O64 [物理化学(理论化学)、化学物理学];
学科分类号
070304 ; 081704 ;
摘要
Authors describe and demonstrate a 'ratio-conserving' mapping procedure for attaining reduced-dimension representations of multidimensional data. This procedure has a theoretical basis for topological correctness in that the ratio of the metrics in the two representations is maintained constant throughout. It is also demonstrated that comparing the reduced-dimension depiction of data with the results of clustering and spanning tree operations in the full-dimension space can validate such reduced-dimension mappings. Demonstrations are carried out using a body of semiconductor data, with five independent variables and one dependent variable. (C) 1998 Published by Elsevier Science S.A. All rights reserved.
引用
收藏
页码:22 / 29
页数:8
相关论文
共 7 条
[1]  
[Anonymous], 1987, Algorithmics: The Spirit of Computing
[2]   ART-2 - SELF-ORGANIZATION OF STABLE CATEGORY RECOGNITION CODES FOR ANALOG INPUT PATTERNS [J].
CARPENTER, GA ;
GROSSBERG, S .
APPLIED OPTICS, 1987, 26 (23) :4919-4930
[3]  
Hart P.E., 1973, Pattern recognition and scene analysis
[4]   SELF-ORGANIZED FORMATION OF TOPOLOGICALLY CORRECT FEATURE MAPS [J].
KOHONEN, T .
BIOLOGICAL CYBERNETICS, 1982, 43 (01) :59-69
[5]   NONLINEAR PRINCIPAL COMPONENT ANALYSIS USING AUTOASSOCIATIVE NEURAL NETWORKS [J].
KRAMER, MA .
AICHE JOURNAL, 1991, 37 (02) :233-243
[6]  
Pao Y. H., 1989, ADAPTIVE PATTERN REC
[7]   Dimension reduction, feature extraction and interpretation of data with network computing [J].
Pao, YH .
INTERNATIONAL JOURNAL OF PATTERN RECOGNITION AND ARTIFICIAL INTELLIGENCE, 1996, 10 (05) :521-535