Claim Missing Document
Check
Articles

Found 2 Documents
Search
Journal : Indonesian Research Journal on Education

Utilization of Google Colaboratory to Determine the Relationship Between the National Defense Index (IBNU) and the Human Development Index (HDI) in Indonesia in 2023 Rifai, Bachtiar; Poniman, Aris; Gultom, Rudy Agus Gemilang; Wadjdi, Achmad Farid; Rifai, Salsabila Azzahra
Indonesian Research Journal on Education Vol. 4 No. 4 (2024): irje 2024
Publisher : Fakultas Keguruan dan Ilmu Pendidikan, Universitas Pahlawan Tuanku Tambusai

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31004/irje.v4i4.1546

Abstract

This study uses Random Forest model and statistical analysis to evaluate the relationship between the Index of State Defense (IBNu) and the Human Development Index (HDI) at the provincial level in Indonesia. The data includes the average IBNu, HDI in 2023, the manager's prediction, and data from the Indonesian Democracy Index (IDI) in 2021-2023. The model shows excellent prediction performance with a Mean Squared Error (MSE) of 0.0018 and an R² value of 0.97, indicating that the prediction results are very close to the actual data. The residual plot also shows that the model has no significant bias as the prediction error is evenly distributed around the zero line. Correlation analysis found that HDI has a strong positive correlation with education variables, such as Expected Years of Schooling and Average Years of Schooling, while IBNu is negatively correlated with economic growth, indicating that an increase in IBNu may have the effect of decreasing economic growth. The classification performance of the model was assessed using the confusion matrix, which showed accuracy, precision, recall, and F1 scores of 0.8, indicating the model's ability to distinguish between the various classes well. These results suggest that IBNu policies aimed at improving HDI may not always be effective and that a more comprehensive policy approach is needed. The proposed recommendations include the integration of national defense and human development programs into the school curriculum, balanced budget allocation, innovative programs for the community, support for the balance between security and welfare, and continuous research and evaluation.
Analisis Ulasan Pengguna Aplikasi Layanan Publik Berbasis Pemrosesan Bahasa Alami dan BERTopic Jabar, Jaka Abdul; Gultom, Rudy Agus Gemilang; Wadjdi, Achmad Farid
Indonesian Research Journal on Education Vol. 5 No. 1 (2025): Irje 2025
Publisher : Fakultas Keguruan dan Ilmu Pendidikan, Universitas Pahlawan Tuanku Tambusai

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31004/irje.v5i1.2162

Abstract

Penelitian ini mengevaluasi ulasan pengguna Polri Super App di Google Play Store untuk mengidentifikasi masalah yang memengaruhi kepuasan pengguna. Dengan menggunakan teknik pemodelan topik berbasis Natural Language Processing (NLP), khususnya BERTopic, kami menganalisis 12.230 ulasan yang dikumpulkan antara Mei 2022 hingga Desember 2024. Hasil penelitian menunjukkan bahwa sebagian besar ulasan memberikan peringkat rendah, dengan lebih dari 50% pengguna memberikan peringkat satu bintang. Tema utama yang teridentifikasi mencakup masalah teknis seperti pembayaran yang kedaluwarsa, aplikasi yang sering tertutup sendiri (force close), waktu muat yang lama, dan kesulitan dalam verifikasi kode OTP. Temuan ini menyoroti perlunya perbaikan mendesak pada aspek fungsional inti aplikasi untuk meningkatkan kepuasan pengguna. Selain itu, penelitian ini menegaskan bahwa ulasan pengguna dapat menjadi sumber wawasan berharga dalam memahami persepsi masyarakat terhadap layanan publik digital, sehingga membantu pengembang meningkatkan kualitas layanan dan mendukung transformasi digital yang lebih efektif di sektor publik.