基于类别相似性优化偏好的个性化学习资源推荐算法
Personalized Learning Resource Recommendation Algorithm to Optimize Preferences Based on Category Similarity
张进军1
作者信息
- 1. 安徽警官职业学院信息管理系,安徽合肥 230031
- 折叠
摘要
由于部分学科数据在网络中的获取难度较高,个性化学习资源推荐可能面临数据稀疏问题,仅考虑资源的类内学习偏好相似性,忽略了类间相似性,从而导致资源推荐的准确性偏低.基于此,提出基于类别相似性优化偏好的个性化学习资源推荐方法.通过学习者的学习偏好数据构建学习者的偏好模型,获取对各个知识点的反馈和评价,进而确定知识点的难易度.针对知识点难易度因素,利用学生的学习行为数据分析获得资源特征量和操作资源特征量,计算知识点的投入程度和难易度系数.基于这些信息,制订个性化学习资源推荐策略,并通过类别相似性和启发式思想进行优化,生成优化后的个性化学习资源推荐结果,以提高推荐准确性和可靠性.实验结果表明,所提方法的推荐效果较好,推荐点击率较高,具有一定的应用价值.
Abstract
Due to the high difficulty of obtaining data in some subjects in the network,personalized learning resource recommendation may face the problem of sparse data. Only the intra-class learning preference similarity of resources is considered,and the inter-class similarity is ignored,resulting in the low accuracy of resource recommendation. We propose a personalized learning resource recommendation method based on category similarity. The learners' preference model is constructed through the learning preference data,so as to obtain the feedback and evaluation of each knowledge point,and then determine the difficulty of the knowledge points. According to the factors of difficulty of knowledge points,the learning behavior data is used to obtain the characteristics of resources and the characteristics of operation resources,and the input degree and difficulty coefficient of knowledge points are calculated. Based on these information,the personalized learning resource recommendation strategies are formulated,and optimized through category similarity and heuristic ideas to generate the optimized personalized learning resource recommendation results to improve the accuracy and reliability of the recommendation. The experimental results show that the proposed method has good recommendation effect and high recommendation click rate,which has certain application value.
关键词
类别相似性/个性化推荐/学习资源/学习偏好/优化算法Key words
category similarity/personalized recommendation/learning resource/learning preference/optimization algorithm引用本文复制引用
基金项目
高校科研计划校级自然科学研究重点项目(2023)(2023zkxm003)
出版年
2024
NETL
NSTL
万方数据