• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2007, Vol. 29 ›› Issue (1): 88-89.

• 论文 • 上一篇    下一篇

自动组题算法中适应度函数的构建方法

张星 廖琪梅   

  • 出版日期:2007-01-01 发布日期:2010-05-30

  • Online:2007-01-01 Published:2010-05-30

摘要:

在基于GA的自动组题算法中,构建适应度函数是最大的一个难点。适应度函数构建的关键在于如何选择适应度函数的参数以及如何计算各个参数的值,本文从IRT理论和试题以 及算法的特殊性出发,确定了适应度函数的参数及其计算方法。该方法有着强大的理论支持和完善的用户需求接口,同时也为试卷的分析提供了新的方法。

关键词: 遗传算法 项目反应理论 适应度函数 组题

Abstract:

Fitness function’s building is the most difficult point in GA-based automatic test creation algorithms. How to select the fitness function’s factors and how to calculate them are the key points in the fitness function’s building. The paper presents a method of selecting and calculating the factors in the fitness function which comes from the Item Response Theory, the test pool’s structuring and the algorithm’s construction. This method has a st trong support from IRT and a perfect user interface, and provides a new way of test paper analysis

Key words: genetic algorithm;item response theory;fitness function;test creation