Rainfall Prediction Using Naïve Bayes and K-Means Algorithms in Malang Region

Abstract

Rain is beneficial for agriculture and also for water reserves, but rain can also have negative impacts such as floods, disease outbreaks and other disasters if not handled properly, therefore to overcome or avoid the problem of excessive rainwater falling It is better to make predictions as an effort to prevent disasters that will occur. Rain will be able to be handled well if you are able to predict when the rain will fall, so that early mitigation can be carried out. With a prediction method using the Naïve Bayes algorithm, rainfall can be predicted because this method uses a probability approach for each data. The result of this method is that it is able to predict the fall of rain by 80%, so the Naïve Bayes method is able to predict when it will rain accurately, where this value comes from 10 test data results for the Malang Raya area. The incorrect value is 20% which is relatively small and this value is quite accurate as a basis for the probability of rain in the Malang Raya area.