Data Distribution Modelling in Supervised Learning Algorithm is for The classification of Prospective Recipient Candidate

Abstract

Indonesia is a country with the largest Muslim population in the world. The number of practices of worship in Islam may affect several things. One of these practices is zakat. This is because zakat can help people with lower economic levels, even zakat has its role in reducing poverty. For this reason, zakat management is very important to be optimized. This research applies a classification to the data of prospective mustahik of BAZNAS Riau 2020. The purpose of this research was to find out the performance of classification algorithm in determining the feasibility of tithe recipient and to give the knowledge to the stakeholder in this case is BAZNAS Riau. The classification algorithms used are Probabilistic Neural Network (PNN), K-Nearest Neighbor (KNN), and Naive Bayes Classifier (NBC). The division of training and testing data is carried out using K-fold Cross-Validation and Hold out. The findings obtained are that the NBC algorithm has better performance with an accuracy of 97.12% based on the K-fold cross-validation division technique.