Evaluating a rule evaluation support method based on objective rule evaluation indices

摘要: In this paper, we present an evaluation of novel rule support method for post-processing mined results with models based on objective indices. Post-processing is one the key issues in a data mining process. However, it difficult human experts to evaluate many thousands rules from large dataset noises completely. To reduce costs task, have developed models, which are obtained indices classification and evaluations expert each rule. performances learning algorithms constructing done case study meningitis as actual problem. Furthermore, also evaluated our four rulesets kinds UCI datasets.