Topic Modelling of Germas Related Content on Instagram Using Latent Dirichlet Allocation (LDA)

DOI

10.2991/ahsr.k.210127.060How to use a DOI?

Keywords

Topic Modelling, LDA, Text Mining, Data Mining, Latent Dirichlet Allocation

Abstract

Content generated by Instagram users related to the Healthy Living Community Movement (GERMAS) has provided new media information that is important for the community and, in particular, the health department. At present, Indonesia is facing a serious challenge in the form of a double burden of disease. Changes in people’s lifestyles are suspected to be one of the causes of a shift in disease patterns (epidemiological transition) in the last 30 years. Discussions on what topics occur in the community related to health, as well as community complaints, have not been identified. The Data Mining technique makes it possible to analyze and extract any topics that are contained from the data captions from Instagram. This study uses Latent Dirichlet Allocation (LDA) as a method for modeling topics. The results of evaluating the number of topics using topic coherence yielded the eight most appropriate topic segments. Based on the results of content analysis on each topic segment, it was found that the most dominant topic related to GERMAS was a healthy lifestyle diet.

Copyright

© 2021, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Muhammad Habibi
AU  - Adri Priadana
AU  - Andika Bayu Saputra
AU  - Puji Winar Cahyo
PY  - 2021
DA  - 2021/01/27
TI  - Topic Modelling of Germas Related Content on Instagram Using Latent Dirichlet Allocation (LDA)
BT  - Proceedings of the International Conference on Health and Medical Sciences (AHMS 2020)
PB  - Atlantis Press
SP  - 260
EP  - 264
SN  - 2468-5739
UR  - https://doi.org/10.2991/ahsr.k.210127.060
DO  - 10.2991/ahsr.k.210127.060
ID  - Habibi2021
ER  -