A web-based intelligent disease-diagnosis system using a new fuzzy-logic based approach for drawing the inferences in crops

This paper suggests a new approach for providing intelligence in the system for diagnosis of diseases of the oilseed-crops. It reports the development of a web-based intelligent disease diagnosis system (WIDDS). The WIDDS is based on a new fuzzy logic approach. The approach is based on use of a rule-promotion methodology. This approach enables the drawing of inferences with the enhanced intelligence. The WIDDS also incorporates new features that improve the presently existing expert systems. The new features are (i) object-oriented (O-O) inference model, (ii) dynamic knowledge base creation strategy. The dynamically promoted rules are derived from those diagnosis sessions, which resulted in successful decisions. This enables more efficient decision-making in the future sessions, (iii) audio–visual–graphical user interface using text-to-speech (TTS) conversion tools. The WIDDS results in decreasing not only the number of interactive question–answer sessions with the clients but also leads to acceptable diagnosis. Further, the inferences are drawn faster compared to the traditional approach, which is the expert based reasoning method. The suggested WIDDS, which is based on rule-promotion approach, has been tested for three oilseeds crops – soybean, groundnut and rapeseed-mustard.

Research highlights
► WIDDS based on a new fuzzy logic approach is developed for Oilseeds crops.
► A new fuzzy logic approach - rule-promotion methodology is described and evaluated.
► WIDDS incorporates design features like O-O inference model, Dynamic KB and TTS.
► Maintaining rule-patterns has significantly improved efficiency of system.
► Results indicated that new methods have improved inference procedures of ES.