Open AccessTopic Discovery of Online Course Reviews Using LDA with Leveraging Reviews Helpfulness
Author(s) -
Fetty Fitriyanti Lubis,
Yusep Rosmansyah,
Suhono Harso Supangkat
Publication year - 2019
Publication title -
international journal of electrical and computer engineering
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.277
H-Index - 22
ISSN - 2088-8708
DOI - 10.11591/ijece.v9i1.pp426-438
Subject(s) - perplexity , computer science , python (programming language) , helpfulness , popularity , devanagari , world wide web , class (philosophy) , artificial intelligence , data science , multimedia , programming language , language model , psychology , social psychology , character recognition , image (mathematics)
Despite the popularity of the Massive Open Online Courses, small-scale research has been done to understand the factors that influence the teaching-learning process through the massive online platform. Using topic modeling approach, our results show terms with prior knowledge to understand e.g.: Chuck as the instructor name. So, we proposed the topic modeling approach on helpful subjective reviews. The results show five influential factors: “learn easy excellent class program”, “python learn class easy lot”, “Program learn easy python time game”, and “learn class python time game”. Also, research results showed that the proposed method improved the perplexity score on the LDA model.