Insights of Data Mining for Small and Unbalanced Data Set Using Random Forests

Provided by: Science & Engineering Research Support soCiety (SERSC)
Topic: Data Management
Format: PDF
Because random forests are generated with random selection of attributes and use samples that are drawn by boostraping, they are good for data sets that have relatively many attributes and small number of training instances. In this paper an efficient procedure that considers the property of data set having many attributes with relatively small number of attributes in arrhythmia is investigated to predict cardiac arrhythmia is shown. Even though several research results have been published already to find better prediction accuracy based on other methods, a new and better result has been found with the suggested method.

Find By Topic