Anzeige
Erschienen in:
2011 | OriginalPaper | Buchkapitel
An Effective Method to Find Better Data Mining Model Using Inferior Class Oversampling
verfasst von : Hyontai Sug
Erschienen in: Convergence and Hybrid Information Technology
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Decision trees are known to have very good performance in the task of data mining of classification, and sampling is often used to determine some proper training sets. Among many parameters the accuracy of generated decision trees depends upon training data sets much, so we want to find some better classification models from the given data sets by oversampling the instances that have higher error rates. The resulting decision trees have better accuracy for classes that had lower error rates, but have worse accuracy for classes that have higher error rates. In order to take advantage of the better accuracy and compensate the worse accuracy, we suggest using class association Experiments with real world data sets showed promising results.