Inclusion of Efficient Rules in PRISM Algorithm for Data Classification
Yogesh Wanjari, Sanjay Nagpure, Gokul Chute, Yogeshwari Kamble · International Journal of Computer Applications · 2019
Data classification is the process of classify data into different types, forms or any other separate classes.Data may be classified for a different of reasons, including ease of access, to observe with regulatory requirements, and to meet various other business or personal intention.In some cases, data classification is a regulatory requirement, as data must be searchable and recoverable within specified time durations.the focus of this paper is on the description of rule based classification and ensemble learning as well as the discussion on some existing methods and techniques.In our proposed approach we are using PRISM algorithm for rule induction.Based on induced rule, test data will be classified.In this paper we proposed Maximized the classification accuracy, Minimize the error rate and also Minimize the classification time.After that experimental evaluation will be performed with basic PRISM algorithm and will show comparative analysis of basic PRISM algorithm and other data classification algorithm such as SVM, Decision Tree, Perceptron Model and Logistic Regression.After comparing these classification algorithms, we found that Maximum accuracy using PRISM algorithm.