JOURNAL OF SCIENCE AND SOCIAL RESEARCH
Vol. 9 No. 3 (2026): June 2026

KLASIFIKASI SENTIMEN PUBLIK TERHADAP PROGRAM SDG 1 PENGENTASAN KEMISKINAN DI INDONESIA MENGGUNAKAN SVM DAN INDOBERT

Nia Mardiah (Universitas Satya Terra Bhinneka)
M. Imam Santoso (Universitas Satya Terra Bhinneka)
Putri Athirah Thaibur (Universitas Satya Terra Bhinneka)
Ayu Andini Br Sitepu (Universitas Satya Terra Bhinneka)



Article Info

Publish Date
22 Jun 2026

Abstract

Abstract: The first goal of the Sustainable Development Goals (SDGs), namely poverty eradication, is a major priority of Indonesia’s national development agenda. Public responses to poverty alleviation initiatives provide valuable insights for assessing the effectiveness of government policies. This study examines public sentiment regarding the implementation of SDG 1 using posts collected from the X platform and compares the classification performance of Support Vector Machine (SVM) and IndoBERT models. A dataset consisting of 1,002 Indonesian-language posts was gathered through web scraping and Application Programming Interface (API) techniques. The research process included data preprocessing, sentiment labeling through a lexicon-based approach, TF-IDF feature extraction for the SVM model, and fine-tuning for the IndoBERT model. Model performance was evaluated using a confusion matrix with accuracy, precision, recall, and F1-score metrics. The findings indicate that negative sentiment is more dominant than positive sentiment toward poverty alleviation programs. The SVM model achieved an accuracy of 80.60%, while IndoBERT reached 80.10%. These results suggest that SVM performed slightly better on the dataset used in this study. Overall, the findings demonstrate that artificial intelligence-based sentiment analysis can support monitoring public perceptions and evaluating poverty reduction policies in Indonesia more effectively and comprehensively for evidence-based decision making processes nationwide. Keywords: sentiment analysis; SDG 1; poverty; SVM; IndoBERT.   Abstrak: Tujuan pertama Sustainable Development Goals (SDGs), yaitu menghapus kemiskinan, menjadi salah satu agenda strategis pembangunan nasional Indonesia. Respons masyarakat terhadap berbagai program pengentasan kemiskinan dapat digunakan sebagai bahan evaluasi terhadap efektivitas kebijakan yang diterapkan pemerintah. Penelitian ini menganalisis sentimen publik mengenai implementasi SDG 1 dengan memanfaatkan unggahan pada platform X serta membandingkan kinerja model Support Vector Machine (SVM) dan IndoBERT dalam proses klasifikasi. Sebanyak 1.002 unggahan berbahasa Indonesia dikumpulkan melalui teknik scraping dan pemanfaatan API. Tahapan penelitian meliputi praproses data, pemberian label sentimen menggunakan pendekatan leksikon, pembentukan fitur TF-IDF pada SVM, serta fine-tuning model IndoBERT. Kinerja model dievaluasi menggunakan confusion matrix dengan indikator accuracy, precision, recall, dan F1-score. Hasil penelitian memperlihatkan bahwa opini negatif lebih dominan dibandingkan opini positif terhadap program pengentasan kemiskinan. Model SVM memperoleh tingkat akurasi sebesar 80,60%, sedangkan IndoBERT mencapai 80,10%. Temuan tersebut menunjukkan bahwa SVM memiliki performa yang sedikit lebih unggul pada dataset yang digunakan. Penelitian ini mengindikasikan bahwa analisis sentimen berbasis kecerdasan buatan berpotensi menjadi alat pendukung dalam memonitor persepsi masyarakat serta mengevaluasi kebijakan pengurangan kemiskinan di Indonesia. Kata kunci: analisis sentimen; SDG 1; kemiskinan; SVM; IndoBERT.

Copyrights © 2026






Journal Info

Abbrev

JSSR

Publisher

Subject

Computer Science & IT Economics, Econometrics & Finance Education Social Sciences

Description

Journal of Science and Social Research is accepts research works from academicians in their respective expertise of studies. Journal of Science and Social Research is platform to disclose the research abilities and promote quality and excellence of young researchers and experienced thoughts towards ...