Comparative Study on Perceived Trust of Topic Modeling Based on Affective Level of Educational Text

Abstract

Latent dirichlet allocation (LDA) is a representative topic model to extract keywords related to latent topics embedded in a document set. Despite its effectiveness in finding underlying topics in documents, the traditional algorithm of LDA does not have a process to reflect sentimental meanings in text for topic extraction. Focusing on this issue, this study aims to investigate the usability of both LDA and sentiment analysis (SA) algorithms based on the affective level of text. This study defines the affective level of a given set of paragraphs and attempts to analyze the perceived trust of the methodologies in regards to usability. In our experiments, the text of the college scholastic ability test was selected as the set of evaluation paragraphs, and the affective level of the paragraphs was manipulated into three levels (low, medium, and high) as an independent variable. The LDA algorithm was used to extract the keywords of the paragraph, while SA was used to identify the positive or negative mood of the extracted subject word. In addition, the perceived trust score of the algorithm was evaluated by the subjects, and this study verifies whether there is a difference in the score according to the affective levels of the paragraphs. The results show that paragraphs with low affect lead to the high perceived trust of LDA from the participants. However, the perceived trust of SA does not show a statistically significant difference between the affect levels. The findings from this study indicate that LDA is more effective to find topics in text that mainly contains objective information.

Everyday life information seeking in South Korea during the COVID-19 pandemic: daily topics of information needs in social Q&A

“…The widely used LDA topic modeling algorithm applying the "scikit-learn" module offered in Python was employed. LDA has often been used to identify a multitude of topics, including health (Bahng and Lee, 2020;Abd-Alrazaq et al, 2020), education (Chen et al, 2016;Im et al, 2019), and political issues (Hagen, 2018). In this study, it was used to identify natural clusters of questions about COVID-19.…”

A descriptive framework for the field of knowledge management

“…The second limitation is related to the results obtained from employing topic modeling algorithm. Although topic modeling is effective in finding underlying topics in documents, the user trust in the discovered topics is still an issue [76]. Therefore, further investigation of the evaluation methods for topic modeling is recommended to further trust the topics discovered by employing this unsupervised technique.…”

Topic Modeling: Perspectives From a Literature Review

“…In addition, both the researchers were from the Republic of Korea (Sungkyunkwan University). Another application is perceived trust in educational texts by professors Jaehyun Park, Minyeong Kim, and other co-authors [61]. The two professors were from Incheon National University in Korea.…”