学习者个性化特征表示及相似度计算

Representation of LearnersRepresentation Learners' Personalization Features and Similarity Computing: A Domain Ontology Perspective

学习者的相似性反映了学习者之间相近的学习经历,将相近者组织成一个具有共同学习兴趣和目标的共同体,可以增强虚拟学习社区中各组成要素的耦合与连接,提高学习者的学习效果和黏度。本文基于领域本体提出一种能够描述学习者个性化特征的VSM模型,并以此为数据结构,设计了一种能有效计算学习者相似性的算法。实验结果表明,利用概念之间继承和包含关系,算出概念相关度,能快速计算出学习者特征向量之间的相似度;得出的结果较好地反映了专家分组经验,为提高虚拟学习社区的个性化和智能化提供了关键技术支持。

Learner similarity indicates the extent to which learners have similar learning experience. Those who are close to one another in terms of learning experience can form a community of shared learning interests and goals, enabling the virtual learning community to work well and consequently enhancing learning effectiveness. Drawing upon the theory of domain ontology, the article proposes the VSM Model which is capable of describing learners' personalization features. The model is used to generate an algorithm for effectively calculating learner similarity. Experiments show that it can compute learner similarity rapidly by making use of relation between concepts from domain ontology. Experiment results are in close match with experts' classification. It is argued that the VSM Model can be used as a key technology to support the development of personalized and intelligent virtual learning communities.