Classification of Drought Impact by Drought Vulnerability Indicators in Probolinggo Regrency Using Naive Bayes

Abstract

Drought in Probolinggo is a big problem because most of the people in this work as farmers. Drought is a natural phenomenon, difficult to define due to differences in hydrometeorological variables and socio economic factors along with the stochastic nature of water demand in various regions. Resident vulnerability to drought hazard is varie. Vulnerability can be measured using vulnerability indicators such as economic factors, social factors, and ecological factors. This research used several vulnerability indicators to classified the impact of drought in three villages in Probolinggo Regency (Sumberkare, Tandonsentul, and Tegalsono). The classification method used in this research is Naïve Bayes. The 10-fold cross validation method was used to train the developed predictive model and the performance of the models evaluated. The accuracy of drought impact by the naive bayes is 85,90 %. Naïve Bayes classifier classify indicators of the impact of drought accurately.