计算机工程 ›› 2008, Vol. 34 ›› Issue (4): 190-192.doi: 10.3969/j.issn.1000-3428.2008.04.067

• 人工智能及识别技术 • 上一篇    下一篇

基于改进微粒群优化的学习路径优化控制方法

陈其晖1,凌培亮1,萧蕴诗2   

  1. (1. 同济大学现代远程教育研究所,上海 200092;2. 同济大学电子与信息工程学院,上海 200092)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2008-02-20 发布日期:2008-02-20

Control Method for Learning Path Optimization Problem Based on Improved Particle Swarm Optimization

CHEN Qi-hui1, LING Pei-liang1, XIAO Yun-shi2   

  1. (1. E-Learning Institute, Tongji University, Shanghai 200092;2. School of Electronics and Information Engineering, Tongji University, Shanghai 200092)
  • Received:1900-01-01 Revised:1900-01-01 Online:2008-02-20 Published:2008-02-20

摘要: 对学生学习的路径控制在智能化教学系统中是一个重要的问题。该文以知识空间理论为基础建立了学习状态空间,通过改进的微粒群算法对该学习状态空间的学习路径进行最优化控制,并利用死亡惩罚函数法把约束最优化学习路径问题转化成了无约束的最优化学习路径控制问题,引入交换子和交换序的概念对微粒群算法进行改进。在结果分析中,通过动态参数法,即动态变化交换子保留概率的方法提高微粒群的收敛效果,达到了最优化学习路径控制的目的。

关键词: 知识空间, 学习路径, 微粒群优化, 动态参数

Abstract: One of the important problems in intelligent tutoring system is to control the student’s learning path. This paper studies learning state space based on knowledge space theory, and introduces the learning path optimization problem based on improved Particle Swarm Optimization(PSO) The constrained learning path optimization problem is transformed into the non-constrained optimal learning path control study by the death penalty function. The PSO is modified and constructed via presenting the concepts of swap operator and swap sequence in the paper. The method of dynamic parameters is processed through changing retain probability of swap operator and swap sequence. The experiments show the improved PSO can achieve good results to control the learning path.

Key words: knowledge space, learning path, Particle Swarm Optimization(PSO), dynamic parameter

中图分类号: