Virtual Learning Environment - Evaluation of Learner’s Behavior Using Topic Models

online learning platforms come with a number of difficulties. To identify the student who does not do the given assignment within the allotted time. Researchers have been attempting to solve this issue in the literature of late, however most algorithms are unable to produce linearly separable learner clusters and correctly classify the input documents. In an attempt to overcome these problems, the suggested methodology builds clusters of linearly separable learners by applying topic models such as Latent Dirichlet Allocation (LDA). First, the necessary features are retrieved and converted into an appropriate LDA of words and phrases. The topic-modeling algorithm (LDA) is then fed the words to create clusters of related content or learners. A number of experiments were carried out to assess how well various predictive models performed. The results show the topic-modeling algorithm LDA attains significant clustering of documents over the other state-of-art.