Claim Missing Document
Check
Articles

Found 2 Documents
Search

Prediksi ISPU Jakarta Menggunakan Random Forest Roris, Renaldi Putra; Saputra, Andhika; Fahrizal, Ahmad; Susilowati, Susi; Rianto, Harsih; Nuryamin, Yamin
Journal Automation Computer Information System Vol. 5 No. 2 (2025): November
Publisher : Indonesian Journal Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47134/jacis.v5i2.139

Abstract

Polusi udara Jakarta memerlukan sistem prediksi akurat untuk peringatan dini kesehatan publik. Penelitian ini mengembangkan model machine learning untuk memprediksi Indeks Standar Pencemar Udara (ISPU) harian maksimum menggunakan dataset 3.045 observasi dari lima stasiun pemantauan (Januari–Agustus 2024) dengan enam parameter polutan (PM10, PM2.5, SO2, CO, O3, NO2). Tiga algoritma dievaluasi: Linear Regression, Random Forest, dan Gradient Boosting. Random Forest mencapai kinerja terbaik dengan R² = 0,9575, RMSE = 4,44, dan MAE = 0,82, melampaui studi sejenis (R² = 0,78–0,89). Analisis feature importance mengungkapkan PM2.5 mendominasi prediksi ISPU dengan kontribusi 87,11%, jauh melebihi NO2 (4,94%) dan SO2 (2,84%). Penelitian memberikan tiga kontribusi: (1) model prediksi ISPU akurasi tertinggi untuk implementasi sistem peringatan dini operasional; (2) identifikasi PM2.5 sebagai target prioritas kebijakan pengendalian polusi berbasis bukti; dan (3) bukti empiris bahwa polusi bersifat kronis dan menyeluruh, memerlukan intervensi komprehensif untuk melindungi kesehatan 10+ juta penduduk Jakarta
Optimalisasi Aplikasi E-Learning Berbasis Moodle Untuk Pengembangan Dan Peningkatan Media Pembelajaran Secara Digital Budi, Eko Setia; Priyatna, Ade; Zuraidah, Eva; Fadillah, Nur; Firmansyah, Ilhan; Rana, Dipo Yudhis; Fitriyani, Fitriyani; Roris, Renaldi Putra
JPM: Jurnal Pengabdian Masyarakat Vol. 6 No. 3 (2026): January 2026
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/jpm.v6i3.2751

Abstract

Community Service is a routine activity carried out by lecturers of Higher Education in order to carry out the obligations of the Tridharma of Higher Education. Lecturers have an obligation to carry out the obligation to share knowledge and skills they have with the wider community who need help. One of the partners who needs support is Al Qomar School which currently has problems in the information system in the digital learning process. This often causes errors in the digital learning process. In addition, the partner also does not have a website application that can be used to display the school's digital learning that makes it easier for teachers to access digital learning quickly and efficiently. Based on this, this community service aims to provide Moodle-Based E-Learning Application Training for the Development and Improvement of Digital Learning Media at Al Qomar School that can be used by partners to improve learning performance better. The activity was carried out face-to-face at the Al Qomar School Hall, Jalan Kamal Raya No. 1 Kalideres, West Jakarta. The target output to be achieved in this activity is the publication of this training activity in national-scale electronic or print media. Improving the ability of the e-learning system aims to provide effective solutions in increasing learning flexibility, accommodating diverse learning styles, and facilitating learning accessibility. E-learning development using the Moodle platform, 100% of teachers are able to design learning with creative and innovative content