An expert system for mango disease diagnosis and management

Abstract

This paper describes the development of a rule-based expert system for the diagnosis of diseases of Indian mango and to suggest the appropriate management. It follows an object oriented approach of presenting rules in the knowledgebase of expert system in the form of Object-Attribute-Value that allows developing knowledge base without using expert system shell software. Initially this system is developed for five major diseases (powdery mildew, anthracnose, bacterial canker, phoma blight and red rust), which may further be extended to different diseases of mango. This expert system is devised to show typical symptoms of the disease, weather parameters critical for rapid development of the diseases and suitable integrated disease management measures. After diagnosis, it also advises the management options of the different diseases. Expert system also include weather based forewarning of powdery mildew disease, which takes into account three weather parameters i.e., maximum and minimum temperatures, relative humidity and wind speed to find out whether weather is conducive for development of powdery mildew or not? If weather parameters are favourable, then the system advises its management options. It is a good enough for providing appropriate advice for early diagnosis and integrated management of diseases for enhancing mango productivity. It would be helpful in early and accurate identification of diseases and their management by the application of biological, cultural, physical and chemical management methods. The system would serve as an effective knowledge dissemination tool and would empower orchardist for effective decision making for the timely management of the different mango diseases.