Title: Knowledge Based System for Diagnosis Tomato Diseases
Authors: Heba I.A. Alqedra & Samy S. Abu-Naser
Volume: 9
Issue: 8
Pages: 100-110
Publication Date: 2025/08/28
Abstract:
Background: Tomato diseases are among the most serious threats to agricultural production worldwide, causing significant reductions in crop yield, deterioration of fruit quality, and severe economic losses for farmers. According to global agricultural studies, plant pests and diseases account for nearly 40% of annual crop losses, with tomatoes being particularly vulnerable due to their perishable nature and susceptibility to multiple pathogens. Objectives: This paper presents TomatoPulse, a knowledge-based expert system designed to support the early detection and diagnosis of seven economically important tomato diseases: Fusarium Wilt, Bacterial Wilt, Early Blight, Late Blight, Gray Mold, Powdery Mildew, and Tomato Mosaic Virus. The primary goal is to assist farmers, agricultural engineers, and students in accurately identifying disease symptoms and recommending suitable treatment options. Methods: The system integrates a rule-based expert system implemented in CLIPS with a user-friendly interface developed in Delphi. The interface guides users through symptom selection and provides corresponding diagnostic outcomes. The knowledge base was constructed from validated agricultural datasets, expert consultations, and published scientific literature. Results: Preliminary evaluation by agricultural specialists and students indicated high satisfaction with the system's accuracy, response time, and ease of use. Diagnostic results demonstrated consistency with expert assessments and published agricultural guidelines. Conclusion: TomatoPulse highlights the potential of expert systems in agricultural decision support, offering a reliable and accessible diagnostic tool that reduces dependency on immediate expert availability, promotes informed disease management, and contributes to sustainable tomato production, particularly in resource-limited farming communities.