基于基因表达式编程的计算机组卷算法研究-《计算机技术与发展》

文章信息/Info

Title:: Research on Computer Test Paper Generation Algorithm Based on Gene Expression Programming

作者:: 韩啸¹; 毕波¹ ; 唐锦萍²; 1. 东北石油大学数学与统计学院,黑龙江大庆 163000; 2. 黑龙江大学数据科学与技术学院,黑龙江哈尔滨 150080

Author(s):: HAN Xiao¹; BI Bo¹; TANG Jin-ping²; 1. School of Mathematics and Statistics,Northeast Petroleum University,Daqing 163000,China; 2. School of Data Science and Technology,Heilongjiang University,Harbin 150080,China

Keywords:: gene expression programming; intelligent test paper generation; mathematical model; genetic operator; intelligent algorithm

摘要:: 随着计算机技术的发展,传统的手动组卷方法难以满足新时代下的各种领域的需求,为解决传统手动组卷在性能、速度、题型分配等方面的缺陷,基于计算机技术的智能组卷问题日渐变为人们关注的问题。然而目前的组卷算法存在成功率低、计算时间久、知识点覆盖不完整、难度系数难以把握、生成的试卷难以满足要求等问题,导致了生成的试卷无法达到理想的效果。为改善上述问题,引入了基因表达式编程算法,通过使用适当的遗传算子,采用线性定长的编码方式,构造了新的智能组卷方法,避免了传统组卷算法成功率低以及适应性差等问题,解决了多约束条件下试卷的分数分配、章节分配、难度等一系列问题。实验证明,该算法有着较高的效率,能够快速地生成满足要求的试卷。

Abstract:: With the development of computer technology,the traditional manual test paper generation method can’t meet the needs of various fields in the new era. In order to solve the shortcomings of traditional manual test paper generation in performance, speed,assignment of questions and other aspects,the intelligent test paper generation problem has gradually attracted more interest. However,there are some problems in the current test paper generation algorithm, such as low success rate, long calculation time, incomplete coverage of knowledge points,difficulty coefficient control,and the generated test paper can hardly meet the requirements. As a result,the generated papers can’t achieve the desired results.In order to solve above problems,a new intelligent test paper generation method is constructed by using gene expression programming algorithm. By introducing appropriate genetic operators and linear fixed length coding method,it avoids the shortcomings of premature and fast convergence of traditional test paper generation algorithm,and solves a series of problems such as score allocation, chapter allocation and difficulty of test paper under multi- constraints. Experiment shows that the proposed algorithm has high efficiency and can quickly generate test papers that meet the requirements.