Claim Missing Document
Check
Articles

Found 2 Documents
Search

WEB Based Geographic Information System For School Maping In Bungo District Melky Ardiyansa; Sepriano Sepriano; Fatima Felawati
Sustainability (STPP) Theory, Practice and Policy Vol. 2 No. 2 (2022): Sustainability: Theory, Practice and Policy October Edition
Publisher : Pusat Kajian Berkelanjutan UIN Sulthan Thaha Saifuddin Jambi

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (853.375 KB) | DOI: 10.30631/sdgs.v2i2.1460

Abstract

The Bungo Regency Government has done several things in realizing the Sustainable Development Goals (SDGs) which are the current global focus that are mutually agreed upon, including Indonesia. This is intended to make the data easier to display. Displaying information in web form will make it easier for people to see it. The method in this research is the Waterfall method. In designing maps using Google Maps, the software used in building this application is PHP, HTML, CSS, as a programming language, MySQL as a database server, and Codeigniter. GIS mapping of education in Bungo Regency is a system that provides information to the public about the location of schools in Bungo Regency and their supporting facilities
Analyzing Public Sentiment on the Proposal to Return Regional Head Elections to DPRD on Platform X Using the C4.5 Algorithm Ade Novia Maulana; Wan Moh Yusoff bin Wan Yaacob; Fatima Felawati
Journal of Information System and Informatics Vol 8 No 2 (2026): April
Publisher : Asosiasi Doktor Sistem Informasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.63158/journalisi.v8i2.1483

Abstract

This study examines public sentiment among X users toward the proposal to return regional head elections (Pilkada) to an indirect electoral mechanism through the Regional People’s Representative Council (DPRD), using a decision-tree classifier based on the C4.5 approach. A dataset of 4,127 tweets collected via X API v2 between December 2024 and January 2026 was analyzed using a seven-stage text preprocessing pipeline. Sentiment labels were generated through a hybrid lexicon-based approach, followed by manual verification of 500 stratified tweets by two independent annotators, yielding substantial inter-annotator agreement (Cohen’s Kappa = 0.78). TF-IDF was used for feature extraction, and the dataset was divided using an 80:20 stratified train-test split. The classifier achieved 81% accuracy, 82% precision, 79% recall, and an F1-score of 80%, outperforming Naive Bayes (74%) and Support Vector Machine (79%) baselines on the same dataset. The sentiment distribution showed that 45% of tweets were negative, 32% were positive, and 23% were neutral, indicating a predominantly critical response among X users toward the proposal. These findings describe discourse on X during the study period and should not be interpreted as representative of broader public opinion. Overall, the study highlights the potential of machine learning methods for analyzing Indonesian political discourse on social media.