ANALYSIS OF FUZZY K-NEAREST NEIGHBORS WITH KNN AND FUZZY LOGIC APPROACH ON DUCK BROILER PRODUCTION

Rapid growth and the ability to turn feed into valuable meat are two advantages of broiler ducks. The success of broiler duck production is reflected in measurable performance indicators such as mortality rate, feed consumption, final body weight, feed conversion ratio (FCR). There are still many disadvantages in the cage management pattern, one of the main factors is the high mortality rate. Therefore, based on the results of research on the subject of broiler duck production, researchers tried to analyze production elements using various data processing techniques, including artificial neural network-based clussification and fuzzy classifiers that have been proven to have very good results for classification data. However, in practice there are situations where the distribution of training and testing data is the same but different. From the results of the previous research analysis, the fuzzy k-nearest neighbor algorithm was used to process broiler duck production data. Based on the test results, the accuracy value of the KNN algorithm was 87%, and the accuracy value of fuzzy logic was 98%. Because the data that the researcher prepared had irregular characters which caused the KNN method to experience many errors during data processing. Furthermore, the researcher combined the KNN and Fuzzy Logic methods into fuzzy k-nearest neighbor. which with the FKNN method obtained an accuracy value of 83%. can optimize the KNN method which previously experienced many errors when processing data.