A Web-Based Soya Bean Expert System Using Bagging Algorithm with C4.5 Decision trees

Abstract – The Machine learning [1] is a mechanism concerned with a computer program that automatically improves its learning capabilities with experience. The Bootstrap aggregation (Bagging) is one of the most popular ensemble methods in Machine Learning. Bagging algorithm uses any classification method to increase the performance of the classifier. In this paper, bagging algorithm is used to increase the performance of the C4.5 classifier. Bagging algorithm runs the slightly altered data on a given classification method several times and combines the hypotheses for achieving higher accuracy in the simple classification method. A knowledge Base known as ‘Soya Bean Expert Knowledge Base’ is constructed by conducting programmed interviews with domain experts in Soya Bean crop production. A Soya bean expert system is developed to identify the disease of the crop with the use of bagging algorithm. A separate user interface for the Soya Bean expert system, consisting of three different interfaces namely, Enduser/ farmer, Expert and Admin is presented here. Enduser/ farmer module may be used for identifying the diseases for the symptoms entered by the farmer. Expert module may be used for adding rules and questions to data set by any domain expert. Admin module may be used for maintenance of the system. This expert system is a web based application for online users with JSP as front end and MYSQL as backend.