Application of C4.5 and Naïve Bayes Algorithm for Detection of Potential Increased Case Fatality Rate Diarrhea

Case Fatality Rate or mortality percentage due to some extraordinary events (outbreaks) diarrhea in Indonesia is still above target expected by the government. Several factors have been known to be the cause of diarrhea, but the most influential factor to increase Case Fatality Rate of diarrhea is not known. Therefore, the purpose of this research is to create classification from diarrhea outbreaks data to obtain the data patterns in the form of classification rule that can be applied to detect Case Fatality Rate of diarrhea. Classification used the C4.5 algorithm and Naïve Bayes algorithm. C4.5 algorithm is a popular algorithm with decision tree approach, while Naïve Bayes algorithm is a popular with probabilistic approach in classification. Research implementation uses the stages in Knowledge Discovery in Databases. After obtaining the classification rule, this rule evaluated by Confussion Matrix and Receiver Operating Characteristic Curve. The evaluation was done by using training data and testing data. The evaluation result in this case indicates that C4.5 algorithm has a higher accuracy level than Naïve Bayes. While the factors that most influence in the Case Fatality Rate increase in diarrhea diseases are shelter and sanitation.