The Research On The Automatic Generation Of Concept Weights And Optimal Composition For Test Item In E-learning System

In the information society, e-learning is an important research field. In the e-teaching and learning system, online testing is a very important way to evaluate students'learning performance. Questions in an item bank are the main factors to compose a test sheet. It's not an easy thing to form a test sheet from the item bank. First, we should define parameters for the test item (for example concept weights). Therefore, how to select appropriate test items in order to effectively construct a suitable test-sheet is a big issue for educators and test experts.This research's main objective is as follows:Objective 1: To analyze the parameters of test items in item banks. This paper proposes a structure that automatically analyzes the parameters of Chinese test items. This structure utilizes latent semantic analysis (LSA) to analyze the relationships of keywords among all test items in an item bank. It also uses the similarity measure to calculate the similarity degree of keywords. We also propose an algorithm for keyword clustering. The concept weights of each test item will be generated automatically with the proposed automatic generate concept weights (AGCW).Objective 2: To propose an appropriate item-selection strategy. An Immune Algorithm (IA) is applied to improve the efficiency of composing optimal test sheet from item banks and compares with a genetic algorithm(GA).I successfully used AGCW to analyze the relationships of keywords in all the test items. And I successfully constructed a suitable test-sheet. The results demonstrated my method effectively that can reduce the working of the test experts or relevant domains teachers.