Journal of Agroecology and Natural Resource Management Print ISSN: 2394-0786, Online ISSN: 2394-0794, Volume 2, Number 2; April-June 2015 pp. 134-140 © Krishi Sanskriti Publications http://www.krishisanskriti.org/janrm.html A Classification Based Target Specific Expert System for Cotton Crop Vandit Rastogi 1 and Arun Solanki 2 1,2 Gautam Buddha University E-mail: 1 vandit92@gmail.com, 2 ymca.arun@gmail.com Abstract—Diseases detection at early stage helps the crops to overcome disease disorder and treat appropriately. Proposed work developed an expert system named Classification Based Target Specific Expert System (CTSES) used to identify the diseases at early stage. In this work CTSES finds the diseases in cotton crops. A rich knowledge base is developed. The knowledge base contains all the facts related to domain. Different rules are developed and used for inferencing. A fuzzy inferencing mechanism provides the reliability of occurrence of disease in cotton plant. The main feature of CTSES is a classification module which helps to determine the probability of occurrence of a disorder and the class to which the disorder belongs. The classification module serves at two level: (1) Classification and grouping of disorder, having same causing agents such as virus, bacteria, fungi etc., based on feature vector extraction technique. (2) Reclassification based on Support Vector Machines (SVM), as provided by the widely used ‘libsvm’ implementation. The classification module helps the system to find results faster from large database by reducing the number of searches and decreasing time complexity. Algorithms are used for classification, based on the concept of SVM, and for finding the reliability of disorder occurrence. 1. INTRODUCTION Artificial intelligence is the branch of computer science concerned with making computers behave like humans. Artificial intelligence includes the following areas of specialization:  Expert systems  Natural Languages  Neural Networks  Robotics [1] An expert system is computer software that attempts to act like a human expert on a particular subject area. Intelligent systems are often used to advise non-experts in situations where a human expert in unavailable. The basic components of an expert system are: User Interface, Inference Engine and Knowledge Base [2]. In this research a new expert system named Classification Based Target Specific Expert System (CTSES) is proposed. 2. RELATED WORK Authors exposed the automatic computation system to analyse the cotton leaf spot diseases. Three features, namely color feature variance, shape and texture feature variance, are extracted by PSO [3]. Crops are classified on the foundation of shape, color and texture with SVM, BPN, Fuzzy along with Edge, CMYK features and GA feature selection are combined for training and testing the cotton diseases dataset [4]. Three different color models for extracting the damaged image from cotton leaf images were implemented, namely RGB color model, HSI color model, and YCbCr color model [5]. Authors reported an image-processing based algorithm to extract plant disease symptoms from colored images. The processing algorithm developed starts by converting the RGB image of the diseased plant or leaf, into the H, I3a and I3b color transformations [6]. ESDIABETES was developed to help people monitor and control the blood glucose level [7]. Development environment was proposed that supports the integration of high level knowledge into host projects, data integration from conventional database systems and system's verification, debugging and profiling [8]. A fuzzy expert system framework was proposed which constructs large scale knowledge based system effectively for diabetes [9]. 3. CTSES ARCHITECTURE CTSES served at two levels of functional processes as user and domain expert. Further, it involves development activities that allow end users to build their own decision support system. End users will have provision to use their own set of decision making parameters to build the target-specific decision support system. Fig. 1 shows the three-tier architecture of CTSES. There are three components named as: Knowledge Base, Inference Module and the User Interface. 3.1 Knowledge Base The Knowledge base is divided into two parts as dynamic and static. The static part of the knowledge base involves the data