基于二进制微分进化算法的学习资源推荐方法  

Learning Resource Recommendation Method Based on Binary Differential Evolution Algorithm

作　　者：王文举[1] 窦曙光 王鸾熠 姜中敏[1] Wang Wenju;Dou Shuguang;Wang Luanyi;Jiang Zhongmin(Shanghai University of Science and Technology,Shanghai 200093,China)

机构地区：[1]上海理工大学出版印刷与艺术设计学院,上海200093

出　　处：《中南民族大学学报（自然科学版）》2018年第1期125-131,共7页Journal of South-Central University for Nationalities：Natural Science Edition

基　　金：新闻出版广电总局重点实验室2017年度招标课题(ZBKT201710);上海市属高校应用型本科试点专业建设项目(10-16-309-001)

摘　　要：为解决当前方法学习资源推荐精度低、时效性不高、无法适应学习者动态调整学习目标的问题,提出了一种二进制的微分进化算法并以此为基础进行学习资源推荐的方法.该法对学习者与学习资源进行数学建模,依据推荐资源需要最佳匹配学习者需求的约束条件,将学习资源推荐转化成了最优化求解问题;根据课程知识点的逻辑组织结构图与学习资源的十字链表存储结构图,快速判定目标函数的关键参数值;采用所提出的一种二进制微分进化算法用于二进制数据表征的学习资源个体变量的变异、交叉、选择操作进行迭代运算直至满足条件找到学习资源最佳的选取组合.实验结果表明:所提出的学习资源推荐方法收敛速度优于以微粒群为基础的学习资源推荐方法,可方便学习者对学习目标范围动态灵活选取,所推荐的学习资源可完全满足不同学习者对学习资源难易度差异化的需求.To address the problem of low accuracy and low effectiveness of current methods which can′t adapt to learners′dynamic adjustment of learning goals,a binary differential evolution algorithm is proposed as a basis for learning resource recommendation.The learners and learning resources are mathematically modeled.According to the recommended conditions of learning resources should best match the learners′requirements,the learning resources recommendation is transformed into the optimal solution problem.According to the logical organization chart of the course knowledge point and the storage structure diagram of the cross?linked list of learning resources,the key parameter value of the objective function can be quickly determined.A binary differential evolution algorithm proposed is used to make the binary individual variables of the learning resource variance,crossover,selection,until the configured conditions are met to find the best selection of learning resources.The experimental results show that the convergence rate and stability of the proposed method are better than the recommended method of learning resources based on the particle swarm optimization algorithm,which can facilitate the learners to choose the learning target range dynamically and make the recommended learning resources be satisfied to different learners′demand for differentiated learning resources.

关 键 词：微分进化算法 学习资源 推荐服务 十字链表 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术] G434[自动化与计算机技术—计算机科学与技术]