Data mining explores a huge amount of data to extract the information to be meaningful. In the field of public health, data mining hold a crucial contribution in predicting disease in early stage. In order to detect diseases, the patients need to conduct various tests. In the context of disease predicion, Data mining techniques aims to reduce the test that patients need to accomplish. Also the techniques is used to increase the accuracy rate of detection. Nowadays, diabetes attacks many adults in the world. Moreover, in order to reduce the number of adult having diabetes, an effective and efficient diabetes detection mechanism should be found. This report will apply some data mining techniques on diabetes dataset that has been downloaded at UCI Machine Learning Repository.Three kind of classification algorithm such as Naïve Bayes Classifier, Multilayer Perceptrons (MLP’s) and Desicion Tree (J.48) have been performed on this dataset. Obtained outcomes indicated that Naïve Bayes Classifier achieved the highest accuracy with 76,30%. As the result, this algorithm is a good method to classify and diagnose diabetes diseases on studying dataset.
Copyrights © 2021