计算机多阶段自适应测验的组卷方法

发布时间：2018-04-21 11:07

本文选题：计算机多阶段自适应测验 + 自动组卷　；参考：《江西师范大学学报(自然科学版)》2017年05期

【摘要】：计算机多阶段自适应测验(MST)实施的关键是成功组建多个满足测验规范(即统计和非统计约束)的平行测验(或称测验面板),自动组卷(ATA)为实现测验平行提供了可能.现有的MST组卷方法研究主要包括以下几种:1)基于线性规范算法的组卷方法;2)基于启发式算法的组卷方法;3)基于蒙特卡洛算法的组卷方法;4)基于在线组卷的方法.该文讨论这几种方法的优缺点并进行比较,同时指出未来可进一步改进这种方法并开发基于认知诊断测验的自动组卷方法.
[Abstract]:The key to the implementation of multi-stage adaptive testing (MSTs) is to successfully set up several parallel tests (or test panels or test panels) that meet the test specifications (i.e. statistical and non-statistical constraints), which provide the possibility for the realization of test parallelism. The existing methods of MST test paper generation mainly include the following: 1) the method of generating test papers based on linear canonical algorithm (2) the method of generating papers based on heuristic algorithm (3) the method of generating papers based on Monte Carlo algorithm (4)) the method of generating papers online based on Monte Carlo algorithm. This paper discusses and compares the advantages and disadvantages of these methods, and points out that this method can be further improved in the future and an automatic test paper generation method based on cognitive diagnostic test can be developed.
【作者单位】：江西师范大学心理学院;
【基金】：国家自然科学基金(31660278,31760288) 教育部人文社科项目(11YJC190002) 高等院校博士点基金(20123604120001) 江西省社会科学规划重点项目(13JY01) 江西省教育科学规划项目(12YB088,13YB029) 江西师范大学青年英才培育计划资助项目
【分类号】：B841.7

【相似文献】