Hierarchical Clustering for Micro-Learning Units Based on Discovering Cluster Center by LDA
Ding Li, Yueqin Zhang, Jian Chen · 2018
Through Massive Open Online Courses (MOOCs), Micro-Learning is leading a trend of attractive paradigm. Micro-learning contents constitute a series of small knowledge points, which can be used to explain complex concepts more effectively. However, with the massive release of micro-learning resources, there are large amount of quite similar contents among micro-learning units. In order to help learners reuse these learning units efficiently. Thus, we try to analyze the text contents of micro-learning lectures which are regarded as micro-learning units. However, with the massive release of micro-learning resources, there are large amount of quite similar contents among micro-learning units. It is badly in need of solution for managing, searching and recommending micro-learning units. In order to solve this problem, we aim to provide a clustering approach, in which, the contents of micro-learning units consisted of titles, descriptions and details are used for our clustering approach. In details, the LDA model is used to find cluster center of micro-learning units, which is called as root cluster center. And then, the Hierarchical Agglomerative Clustering (HAC) is used to cluster the micro-learning units based on their root cluster centers. The preliminary results show our proposal can improve the precision of clustering, and verify the effectiveness of our proposal.