<![CDATA[Diabetes or can be called with diabetes or blood sugar disease is a disease that is hard to cure but can be controlled blood sugar levels. This causes people with diabetes is increasing every year. This study aims to determine which algorithm that has the best classification accuracy, so that it can be used to assist in classifying whether a person has diabetes or not. The data used is the Pima Indians Diabetes dataset obtained from the UCI machine learning. Processing of data mining is divided into two stages, namely stage of data preprocessing and feature selection. Results of the research that has been done, C4.5 algorithm has an accuracy of 73.82% and increased to 74.87%, subsequent to the selection of attributes. While naïve Bayes has an accuracy rate of 76.30% and increased to 77.47%. The end result of this research is naïve bayes algorithm is better than C4.5 algorithms because it has a better accuracy rateKeyword: C4.5, Naïve Bayes, Diabetes]]>
Copyrights © 2015
