Sistem Pakar Diagnosis Hama Penyakit Tanaman Jeruk Keprok Borneo Prima (Citrus Reticulata) Dengan Fuzzy Tsukamoto

Abstract

Borneo Prima Citrus (Citrus Reticulata) is a monoculture commodity that is susceptible to pests or diseases. The role of an expert is constrained by the fact that expert knowledge is not well documented, the detection process is still manual and expensive, and the distance between agricultural extension workers and plantation sites is far. This problem can be solved by an expert system. The Fuzzy-Tsukamoto method is suitable for solving citrus disease diagnosis problems which have many rules with several symptoms that can represent several diseases. Fuzzy Tsukamoto applies monotonous reasoning to each rule. If in monotonous reasoning, the system only has one rule, in the Tsukamoto method the system consists of several rules. Because it uses the basic concept of monotonous reasoning, in Fuzzy-Tsukamoto's method, every consequence of the IF-THEN rule is represented by a fuzzy set on a monotonic membership function. There were 13 citrus trees indicated to be infected with the disease, which was observed from 500 trees. The accuracy of the results obtained between the expert system and expert knowledge is 84% of 13 diseases and 36 symptoms. The result shows that there are two diseases that are not suitable and 11 diseases according to experts.