Model and Application of Web-based e-Assessment With Combination of Rule Base Reasoning (RBR) and Case Base Reasoning (CBR) Methods

Abstract

E-Assessment or known as the Computer Based Assessment, Computer-Based Testing, or the Computerized Testing is a method in the management of the test with a response that was recorded electronically. Application of rule-based and cases in e-assessment can improve the functionality of the system assessment. Rule-based reasoning is a way of reasoning based on specific situations using the short-term memory and long-term memory as a basis to perform an action. While the case-based reasoning is based reasoning previous cases. By utilizing a computer, this reasoning process of changing the subject of human actors into the computer. Conducted by human reasoning is replaced by machine-based reasoning inference based on the rules and the facts of the case. In the assessment activity, both models this reasoning can be used to determine the format about assessment, how the implementation of assessment, similarity analysis of calculation and similarity analysis of assessment suggestions. To formulate rules can be used ID3 algorithm that will generate decision trees and tables of rules determining the format of questions and how assessment. As for similarity calculation of value analysis and advice in the assessment to determine the weight and the weight of the proximity attribute similarity assessment of cases using a similarity function of new cases with old cases using nearest neighbor algorithm. The combination of rule-based reasoning and case-based reasoning be integrated into the framework of e-assessment system based on web. Keywords: E-Assessment, Rule-based reasoning, Case-based reasoning, ID3, Similarities, Nearest Neighbor.