<![CDATA[Determining the status of poor families as recipients of assistance is very important so that poverty reduction assistance from the government can be channeled on target. Data mining utilizes experience or even mistakes in the past to improve the quality of the model and the results of its analysis, one of which is the ability possessed by data mining techniques, namely classification. The purpose of this study was to test K-Fold Cross Validation in the K-Nearst Neighbors algorithm in predicting receipt of village aid funds. In the beneficiary dataset used in this study, there were 159 records or tuples with four attributes (house condition, income, employment and number of dependents). The new data category prediction is done by using the Euclidean Distance manual calculation stage of five different K values. While using the Rapidminer application aims to test the accuracy of the dataset in five different K values. The results show that with K=15 and K=30 the new data (D160) has a "Not Eligible" category with an accuracy of 100%. Then with K=45, K=60 and K=75, the new data (D160) has the category "Eligible" with an accuracy rate of 81.25%.]]>
Copyrights © 2019
