Title :
Data driven knowledge extraction of materials properties
Author :
Kandola, J.S. ; Gunn, S.R. ; Sinclair, I. ; Reed, P.A.S.
Author_Institution :
Sch. of Eng. Sci., Southampton Univ., UK
Abstract :
The problem of modelling a large commercial materials dataset using advanced adaptive numeric methods is described. The various approaches are outlined, emphasising their characteristics with respect to generalisation, performance and transparency. A highly novel support vector machine (SVM) approach is taken incorporating a high degree of transparency via a full analysis of variance (ANOVA) expansion. Using an example which predicts 0.2% proof stress from a set of materials features, different modelling techniques are compared by benchmarking against independent test data
Keywords :
Bayes methods; knowledge acquisition; materials properties; multilayer perceptrons; ANOVA expansion; advanced adaptive numeric methods; analysis of variance expansion; data driven knowledge extraction; generalisation; large commercial materials dataset; performance; proof stress; support vector machine approach; transparency; Analysis of variance; Data engineering; Data mining; Gunn devices; Intersymbol interference; Material properties; Predictive models; Production; Support vector machines; Thermomechanical processes;
Conference_Titel :
Intelligent Processing and Manufacturing of Materials, 1999. IPMM '99. Proceedings of the Second International Conference on
Conference_Location :
Honolulu, HI
Print_ISBN :
0-7803-5489-3
DOI :
10.1109/IPMM.1999.792507
