 |
English
|
正體中文
|
简体中文
|
Items with full text/Total items : 94286/110023 (86%)
Visitors : 21700257
Online Users : 570
|
|
|
Loading...
|
Please use this identifier to cite or link to this item:
http://asiair.asia.edu.tw/ir/handle/310904400/4022
|
| Title: | A Computerized Adaptive Test System Based On a-Nearest-Neighbor Strategy |
| Authors: | CHANG, CHUN CHIN |
| Contributors: | Department of Computer Science and Information Engineering |
| Keywords: | A Computerized Adaptive Test System Based On a-Nearest-Neighbor Strategy |
| Date: | 2006 |
| Issue Date: | 2009-11-18 13:13:37 (UTC+0) |
| Publisher: | Asia University |
| Abstract: | Computerized adaptive test can be used to test the students in accordance with their capabilities, which not only shorten the testing time but also accurately evaluate the performance of the students. The successful elements of CAT are defined by the accuracy of traits estimation and the control of testing items exposure rate. To date, the study in the controlling item exposure rate is conducted by the computerized simulation system, which was not an actual system available for testers.
The purpose of the study is to develop an online Computerized Adaptive Testing System for testing the primary school – the third grade on their mathematics competence. The system is building upon the testing item bank of “Mathematics Competence Indicators for primary school third grade”, Shu-Ching Wang (2005), and incorporating the techniques of testing theory, the controlling of testing item exposure rate and the Internet network.
Regarding the research methods, the Computerized Adaptive Testing System, which is incorporated with the Item Response Theory and a-Nearest-Neighbor Strategy, proceeds with controlling the testing item exposure rate and can accurately estimate testers competences. Furthermore, the records of response procedures of all students are collected as the data for controlling exposure rate.
The research results are as follows:
1. The Computerized Adaptive Testing System, building upon Testing Item response theory, can be used to analyze students problem solving and responding capabilities in an instant and accurate manner.
2. The Computerized Adaptive Testing System, building upon a-Nearest-Neighbor Strategy for testing item selection, can be used as an efficient approach to control item exposure rate.
3. The Computerized Adaptive Testing System, building upon a-Nearest-Neighbor Strategy for testing item selection, can effectively shorten the testing time and decrease the quantity of testing items. |
| Appears in Collections: | [資訊工程學系] 博碩士論文
|
All items in ASIAIR are protected by copyright, with all rights reserved.
|
