基于遗传算法的智能组卷系统的研究与实现  

Research and implementation of automatic test paper generation system based on genetic algorithm

作　　者：赵博 宁慧[1] 张汝波 ZHAO Bo;NING Hui;ZHANG Rubo(College of Computer Science and Technology,Harbin Engineering University,Harbin 150001,China;College of Mechanical and Electronic Engineering,Dalian Minzu University,Dalian 116600,China)

机构地区：[1]哈尔滨工程大学计算机科学与技术学院,黑龙江哈尔滨150001 [2]大连民族大学机电工程学院,辽宁大连116600

出　　处：《应用科技》2021年第2期50-53,共4页Applied Science and Technology

基　　金：国家自然科学基金项目(61673084).

摘　　要：为解决传统组卷算法存在的工作量大、耗时长和试卷主观性强的问题,本文研究并实现了基于遗传算法的智能组卷系统。实现遗传算法的关键是设计适应度函数,适应度函数是描述试卷是否符合要求的函数,因此适应度函数设计的好坏直接决定了试卷的质量。为此研究设计的适应度函数能够对试卷难度、知识点考察数、区分度等多个试卷指标做出综合评判,生成令人满意的试卷。实验结果表明,基于遗传算法的智能组卷系统生成的试卷在难度、知识点覆盖数和区分度等指标上能够达到要求,满足课程考核的需要。In order to solve the problems of large workload,time-consuming and subjective test papers existing in traditional test paper generating algorithms,this paper studies and implements an intelligent test paper generating system based on genetic algorithm.The key to implementing genetic algorithm is to design the fitness function.The fitness function describes whether the test paper meets the requirements,so its design directly determines the quality of the test paper.The fitness function designed in this paper can make a comprehensive evaluation of multiple test paper indexes such as the difficulty of test paper,the number of knowledge points investigated and the degree of discrimination,and generate a satisfactory test paper.The experimental results have shown that the test papers generated by the genetic algorithm-based intelligent test paper composition system can meet the requirements in terms of difficulty,coverage of knowledge points,and degree of discrimination,and can meet the need of course assessment.

关 键 词：组卷系统 遗传算法 适应度函数 智能组卷 算法设计 系统实现 试卷评判 教学考核 

分 类 号：TP312[自动化与计算机技术—计算机软件与理论]