Rozak, Fatur and Widiyatama, Oja and Rozie, Andri Fachrur and Nugraheni, Ekasari and Kurniasari, Dian (2026) Comparative Lexicon-Based Sentiment Analysis Towards Indonesian ’Cek Kesehatan Gratis’ Program in X. In: 2025 International Conference on Computer, Control, Informatics and its Applications (IC3INA), 15-16 October 2025, Jakarta, Indonesia.

Text Comparative_Lexicon-Based_Sentiment_Analysis_Towards_Indonesian_Cek_Kesehatan_Gratis_Program_in_X.pdf Download (2MB)

Abstract

Sentiment analysis aims to understand a person’s perspective or attitude toward a particular topic. One of its main challenges is the labeling approach, especially in the context of the Indonesian language. This study evaluates six labeling approaches: manual labeling, InSet Lexicon, modified InSet Lexicon, VADER, VADER-Translate, and SentIL. Each approach classifies data into three categories: positive, neutral, and negative. The dataset consists of 8,393 posts from the X application related to the Free Health Check Program. Three models were used for the evaluation process: Support Vector Machine (SVM), Convolutional Neural Network (CNN), and IndoBERT. The results show that the combination of the modified InSet Lexicon and IndoBERT produced the best performance, achieving an accuracy of 94.04%, precision of 93.29%, recall of 93.47%, and an F1-score of 93.31%. These findings highlight the importance of selecting lexicon-based methods that align with the study domain and the linguistic characteristics of the data to improve the accuracy and reliability of sentiment analysis in public policy.

Item Type:	Conference or Workshop Item (Paper)
Subjects:	Q Science > QA Mathematics
Divisions:	Fakultas Matematika dan Ilmu Pengetahuan Alam (FMIPA) > Prodi Matematika
Depositing User:	DIAN KURNIASARI
Date Deposited:	17 Apr 2026 02:42
Last Modified:	17 Apr 2026 02:42
URI:	http://repository.lppm.unila.ac.id/id/eprint/54837

Actions (login required)

View Item