{"title":"KEDS: a knowledge-based equation discovery system for engineering problems","authors":"R. Rao, S. Lu","doi":"10.1109/CAIA.1992.200032","DOIUrl":null,"url":null,"abstract":"Many engineering phenomena of interest are characterized by non-homogeneity. The authors discuss how the intertwining of the partitioning and discovery processes enables KEDS to learn relationships from engineering data and to extract the structure underlying these relationships. They present the KEDS algorithm and discuss the interaction between the two discovery and partitioning phases. Some extensions to the basic algorithm are described that greatly improve the performance of KEDS and increase the representation power of the models by permitting a probabilistic partitioning of the problem space. The results from running the KEDS system on data from a simulator for an internal combustion engine are presented.<<ETX>>","PeriodicalId":388685,"journal":{"name":"Proceedings Eighth Conference on Artificial Intelligence for Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1992-03-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"8","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings Eighth Conference on Artificial Intelligence for Applications","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CAIA.1992.200032","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 8
Abstract
Many engineering phenomena of interest are characterized by non-homogeneity. The authors discuss how the intertwining of the partitioning and discovery processes enables KEDS to learn relationships from engineering data and to extract the structure underlying these relationships. They present the KEDS algorithm and discuss the interaction between the two discovery and partitioning phases. Some extensions to the basic algorithm are described that greatly improve the performance of KEDS and increase the representation power of the models by permitting a probabilistic partitioning of the problem space. The results from running the KEDS system on data from a simulator for an internal combustion engine are presented.<>