<> "The repository administrator has not yet configured an RDF license."^^ . <> . . "Exploiting context when learning to classify"^^ . "This paper addresses the problem of classifying observations when\nfeatures are context-sensitive, specifically when the testing set involves a context\nthat is different from the training set. The paper begins with a precise definition of\nthe problem, then general strategies are presented for enhancing the performance\nof classification algorithms on this type of problem. These strategies are tested on\ntwo domains. The first domain is the diagnosis of gas turbine engines. The\nproblem is to diagnose a faulty engine in one context, such as warm weather,\nwhen the fault has previously been seen only in another context, such as cold\nweather. The second domain is speech recognition. The problem is to recognize\nwords spoken by a new speaker, not represented in the training set. For both\ndomains, exploiting context results in substantially more accurate classification."^^ . "1993" . . . . . . . . "Peter"^^ . "Turney"^^ . "Peter Turney"^^ . . . . . . "Exploiting context when learning to classify (PDF)"^^ . . . . . . . . . "NRC-35058.pdf"^^ . . . "Exploiting context when learning to classify (Indexer Terms)"^^ . . . . . . "indexcodes.txt"^^ . . "HTML Summary of #1863 \n\nExploiting context when learning to classify\n\n" . "text/html" . . . "Artificial Intelligence" . . . "Machine Learning" . . . "Statistical Models" . .