Classification of Determination the Recipients of the Program Keluarga Harapan (PKH) Using K-Nearest Neighbor Algorithm

Classification is one method of data mining techniques for classifying data by system according to predetermined rules. In this study, the algorithm used was K-Nearest Neighbor (KNN) algorithm and the data used in this study was the community data obtained from Kantor Desa Gampong Uteun Geulinggang, Dewantara, in Aceh Utara. The main focus of this study was to analyzing and applying K-NN algorithm in a web-based system to classify data on beneficiaries of the Program Keluarga Harapan (PKH) in Gampong Uteun Geulinggang, Dewantara, Aceh Utara. This study used 16 (sixteen) criterias, namely house status, floor area, floor type, gas cylinder, refrigerator, air conditioner, water heater, television, jewelry/gold (10 grams), computer/laptop, bicycle, motorcycle, car, cow, buffalo and goat. The data was classified into 3 classes, namely worthy, not unworthy and very unworthy. The results of this study with a value of k = 3, obtained a precision value of 97%, recall value of 95% and the accuracy value of 97%