Optimization of Certainty Factor Algorithm to Overcome Uncertainty in Expert System Identification of Pests and Diseases of Cocoa

• Published in: 2022 2nd International Conference on Electronic and Electrical Engineering and Intelligent System (ICE3IS) pp. 310 - 315
• Main Authors: Gunawan, Rakhmat Dedi, Ahmad, Imam, Parjito, Anggono, Harry, Vernando, Ego Hengki, Jaya, Firman
• Format: Conference Proceeding
• Language: English
• Published: IEEE 04.11.2022
• Subjects: certainty factors; consistency ratio; Deep learning; diseases and pests of cocoa plant; Electrical engineering; Expert systems; Indexes; Intelligent systems; Production; Uncertainty
• DOI: 10.1109/ICE3IS56585.2022.10010123

Lampung Province is one of the regions with the highest cocoa production, but many factors can interfere with cocoa production. One of them is the factor of pests and diseases of cocoa that cannot be identified and prevented beforehand. An expert system is a system that adopts expert knowledge or experts in a particular field. However, the diagnosis usually contains uncertainty in the form of answers and statements used. Certainty Factor (CF) algorithm, is one of the algorithms that is able to overcome uncertainty by providing a value for the level of trust from experts and users. However, because the CF value is based on the level of confidence in the symptoms experienced by the user, it creates a problem, namely the consistency of the user's answers. So, it is necessary to make improvements to the CF algorithm, to overcome the inconsistency of user answers through the use of Consistency Ratio (CR). Through the use of CR, inconsistencies are overcome through comparisons between the consistency index and the random index. Based on the results of the accuracy test, the proposed model is able to produce an accuracy rate of 86.67% in the diagnosis of pests and diseases on cocoa plants.