智能组卷系统中试题难度正态分布算法的研究  

A normal-distribution algorithm for question difficulty in the smart test-paper generation system

作　　者：潘刚 AN Gang(School of Intelligent Manufacturing,Sichuan University of Arts and Science,Dazhou 635000,China;Dazhou Industrial Technology Institute of Intelligent Manufacturing,Sichuan University of Arts and Science,Dazhou 635000,China)

机构地区：[1]四川文理学院智能制造学院,四川达州635000 [2]四川文理学院达州智能制造产业技术研究院,四川达州635000

出　　处：《云南民族大学学报（自然科学版）》2018年第4期326-334,共9页Journal of Yunnan Minzu University:Natural Sciences Edition

基　　金：四川文理学院科研项目(2016KZ007Y);四川文理学院2017年度智能制造产业开发研究专项(2017ZZ004Y;2017ZZ003G);四川文理学院2015年度特色培育一般项目(2015TP001Y);四川省教育厅项目(18ZB0507)

摘　　要：针对当前智能组卷系统中存在的试卷区分度不高、试题难度呈非正态分布及不能根据应试者层次选择试题难度的问题,深入分析了试卷区分度、试题难度及其正态分布的关系及问题产生的原因,提出了一种适合在基于遗传算法的智能组卷系统中控制试题难度呈正态分布的算法,并借助项目反应理论对该算法进行优化处理,使试卷难度可以根据应试者的不同层次要求进行自由调整.最后根据蒙特卡罗理论设计了一个仿真试验系统对该算法的性能进行验证,实验结果表明,该算法的性能达到了预期效果.In order to solve some problems in the smart test-paper generation system such as vague difficulty discrimination,abnormal distribution of question difficulty or failure in determining the proper question difficulty,this paper gives an in-depth analysis of the relationship among different test items and the causes of the problems. It proposes and optimizes an algorithm,which follows the normal distribution of question difficulty in the smart test-paper generation system on the basis of a genetic algorithm and the item response theory so that the question difficulty is proper for the examinees. Meanwhile,it uses the Monte Carlo theory in the simulation experiment to test this algorithm,whose result is satisfactory.

关 键 词：试卷区分度 试题难度 正态分布 应试者层次 智能组卷系统 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]