<![CDATA[Indonesia is a vast country with many islands suitable for the development of livestock business. In reality, the livestock sector has not been able to encourage public and private participation. To overcome these problems, some of the budget of the Ministry of Agriculture is allocated in the form of social assistance expenditures, such as for community empowerment and poverty alleviation in the form of goods to farmer groups. One of the forms of assistance allocated to farmer groups is the provision of livestock. Determination of potential recipients is still not effective and sometimes leads to the giving of livestock assistance is not right on target, so that every expenditure of state money does not provide maximum benefits for the community. In this research, K-Means Naive Bayes (KMNB) method is considered capable of giving accurate classification results on the determination of livestock recipients. The KMNB learning approach is formed by combining clustering and classification techniques. K-Means is used as a pre-classification component to group the same data at an early stage. Furthermore, for the second grouping of data will be classified by category Accepted or not using Naive Bayes. Thus, the data with the wrong group during the first stage will be classified according to the category in the second stage. Based on the test results by comparing the results of grouping on conventional K-Means method it is proven that KMNB gives the highest accuracy of 100% while conventional K-Means has an accuracy of 95.91%]]>
Copyrights © 2018
