<![CDATA[Oil palm cultivation is one of the main agricultural commodities in Indonesia, as oil palm plants have a relatively high economic value. However, oil palm cultivation is vulnerable to disease attacks that can threaten productivity. The lack of knowledge among farmers in managing plantations and addressing these problems can exacerbate damage to plants due to disease attacks. To address this issue, this research will develop an application using a Case-Based Reasoning (CBR) approach to diagnose diseases in oil palm plants with the nearest neighbor calculation algorithm. CBR will solve new problems by reusing knowledge to solve old problems that are similar and already have solutions. The nearest neighbor algorithm is useful for calculating similarity between new cases and old cases. The data for the research involves 7 types of diseases, 20 symptoms, and 345 cases of disease attacks on oil palm plants. The application testing uses 25 test cases and 320 case bases. From the test results, the application shows an accuracy rate of 31.00, with the highest similarity value reaching 0.98. In this research, it can be seen that the generated similarity values are high but the accuracy value is low. This can be caused by several factors, such as the data of the cases, feature representation, and insufficient diagnosis, which result in high similarity values due to the assignment of values to attribute categories and attribute value proximity.]]>
Copyrights © 2024
