An Integrated Approach for Sentiment Analysis and Topic Modeling of a Digital Bank in Indonesia using Naïve Bayes and Latent Dirichlet Allocation Algorithms on Social Media Data

Setiawan, Johan (2023) An Integrated Approach for Sentiment Analysis and Topic Modeling of a Digital Bank in Indonesia using Naïve Bayes and Latent Dirichlet Allocation Algorithms on Social Media Data. In: 4th International Conference on Big Data Analytics and Practices (IBDAP), 25-27 Agustus 2023, Bangkok, Thailand.

Text An Integrated Approach for Sentiment Analysis and Topic Modeling of a Digital Bank in Indonesia using Naïve Bayes and Latent Dirichlet Allocation Algorithms on Social Media Data.pdf Download (2MB)

Abstract

Social media provides a public platform for expressing complaints and opinions. Researchers can use text mining techniques such as sentiment analysis and topic modeling on social media data to compare features and gauge public opinion on competing digital banks in Indonesia. The aim of this study is to classify sentiments and identify topics in social media data related to a specific digital bank. To accomplish this, the Naïve Bayes algorithm is used for sentiment analysis, while Latent Dirichlet Allocation is used for topic modeling. The social media data is sourced from Twitter and Instagram for Line Bank digital bank. The study finds that the Naïve Bayes algorithm performs well in classifying sentiments, achieving a maximum F1 score of 0.863. Positive sentiments are more prevalent in Twitter data, while negative sentiments are more prevalent on Instagram. Topic modeling using Latent Dirichlet Allocation algorithm identifies four optimal topics for positive sentiment and five for negative sentiment. The coherence value obtained is 0.426279 for positive sentiment and 0.397232 for negative sentiment.

Item Type:	Conference or Workshop Item (Paper)
Creators:	Setiawan, Johan
Contributors:
Keywords:	Analytical models, Sentiment analysis, Social networking (online), Blogs, Multimedia Web sites, Coherence, Prediction algorithms
Subjects:	000 Computer Science, Information and General Works > 000 Computer Science, Knowledge and Systems > 006 Special Computer Methods
Divisions:	Faculty of Engineering & Informatics > Information System
Date Deposited:	04 Feb 2026 02:58
URI:	https://kc.umn.ac.id/id/eprint/44557

Actions (login required)

View Item