Development of a Case-based Reasoning System for Onion Disease Diagnosis and Treatment

Abstract

Onion is a bulb crop widely grown for local use and sold in markets in Ethiopia by businessmen and families in the countryside. Although onions have been utilized for cooking daily which shows there economic significance, there are various barriers to increasing onion production in Ethiopia. They include pests and diseases, productivity constraints, and inadequately developed abilities. Vegetable disease is mostly diagnosed and treated by domain experts. Specialists in the field can occasionally address issues of agricultural productivity. In this study, a decision-making system based on the variables and symptoms was developed to give recommendations for onion disease diagnosis and therapy. To create the case-based reasoning system, relevant information was gathered through organized and unstructured interviews with domain experts that were chosen using the purposive sample technique. Through direct interactions with a subset of overall specialists in the field, a prototype's functionality was examined to determine if the stated goal was met. Test cases were chosen from instances during platform validation, and the results showed that 85.7% of users approved of the system. Additionally, the system's performance was evaluated in terms of recall and precision, where it records results of 76% and 82%, respectively.