QML-Morven: A Novel Framework for Learning Qualitative Models

作者: George M. Coghill , Wei Pang

DOI:

关键词:

摘要: In this report, a novel qualitative model learning (QML) framework named QML-Morven is presented. an extensible and currently includes three QML subsystems, which employ either symbolic or evolutionary approaches as their strategies. uses the formalism of Morven, fuzzy simulator, to represent reason about models, it also utilises Morven verify candidate models. Based on framework, series experiments were designed carried out to: (1) results obtained by previous system ILP-QSI; (2) investigate factors that influence precision minimum data requirement for successful learning; (3) address scalability issue systems.

参考文章(42)
Fernando J. Von Zuben, Leandro Nunes de Castro, The Clonal Selection Algorithm with Engineering Applications 1 ,(2000)
Alen Varsek, Qualitative model evolution international joint conference on artificial intelligence. pp. 1311- 1316 ,(1991)
George MacLeod Coghill, Mycroft: a framework for constraint based fuzzy qualitative reasoning Heriot-Watt University. ,(1996)
Stephen P. Ellner, John Guckenheimer, Dynamic Models in Biology ,(2006)
Mark Eric Wiegand, Constructive qualitative simulation of continuous dynamic systems Heriot-Watt University. ,(1991)
Yumi Iwasaki, Herbert A. Simon, Theories of causal ordering: reply to de Kleer and Brown Artificial Intelligence. ,vol. 29, pp. 661- 665 ,(1986) , 10.1016/0004-3702(86)90091-3
David T. Hau, Enrico W. Coiera, Learning Qualitative Models of Dynamic Systems inductive logic programming. ,vol. 26, pp. 177- 211 ,(1997) , 10.1023/A:1007317323969
Gregory L. Kearns, Wolfgang A. Ritschel, Handbook of basic pharmacokinetics-- including clinical applications American Pharmacists Association,. ,(1992)
Jon Timmis, Camilla Edmonds, A Comment on Opt-AiNET: An Immune Network Algorithm for Optimisation genetic and evolutionary computation conference. pp. 308- 317 ,(2004) , 10.1007/978-3-540-24854-5_32