Article Per Year (5 Year)
p-Index From 2021 - 2026
3.212
P-Index
This Author published in this journals
All Journal IKRA-ITH Informatika : Jurnal Komputer dan Informatika MIND (Multimedia Artificial Intelligent Networking Database) Journal Jurnal Teknologi Terpadu Jurnal Informatika Terpadu Bulletin of Computer Science Research The Indonesian Journal of Computer Science DBESTI: Journal of Digital Business and Technology Innovation
Adriansyah, Ahmad Rio
Sekolah Tinggi Teknologi Terpadu Nurul Fikri
Computer Science & IT Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Education Electrical & Electronics Engineering Engineering Library & Information Science

Published : 22 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 22 Documents
Search

 1   2   3 
Prediksi Retensi Mahasiswa Menggunakan Algoritma Random Forest dengan Optimasi Algoritma Genetika PUDY PRIMA; AHMAD RIO ADRIANSYAH; ALFIAN NUR USYAID
MIND (Multimedia Artificial Intelligent Networking Database) Journal Vol 11, No 1 (2026): MIND Journal
Publisher : Institut Teknologi Nasional Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26760/mindjournal.v11i1.115-126

Abstract

AbstrakKetidakseimbangan kelas (imbalanced data) memicu bias mayoritas pada model konvensional dalam memprediksi retensi mahasiswa. Penelitian ini mengusulkan model peringatan dini (early warning system) dengan mengintegrasikan teknik penyeimbangan data Synthetic Minority Over-sampling Technique (SMOTE) dan pengklasifikasi Random Forest (RF). Untuk menghindari inefisiensi pencarian hyperparameter manual, Algoritma Genetika (GA) diaplikasikan guna melakukan optimasi secara global. Pengujian terhadap dataset historis mahasiswa STT Terpadu Nurul Fikri angkatan 2021 membuktikan bahwa kombinasi SMOTE dan GA-RF sangat efektif. Model hibrida ini mencapai akurasi global 99%, dengan nilai Precision 1,00 dan Recall 0,67 pada deteksi kelas minoritas (dropout). Analisis ekstraksi fitur (Feature Importance) mengungkap bahwa ketahanan studi mahasiswa didominasi oleh performa Indeks Prestasi Semester (IPS) di tahun pertama serta faktor administratif berupa jalur pendaftaran seleksi mandiri.Kata kunci: prediksi Dropout, Ketidakseimbangan Data, SMOTE, Random Forest, Algoritma GenetikaAbstractClass imbalance triggers majority bias in conventional models for predicting student retention. This study proposes an early warning model integrating the Synthetic Minority Over-sampling Technique (SMOTE) for data balancing and a Random Forest (RF) classifier. To avoid manual hyperparameter tuning inefficiencies, a Genetic Algorithm (GA) is applied for global optimization. Testing on the 2021 historical student dataset of STT Terpadu Nurul Fikri proves the effectiveness of combining SMOTE and GA-RF. The hybrid model achieved 99% global accuracy, with 1.00 Precision and 0.67 Recall in minority class (dropout) detection. Feature Importance analysis reveals that student study retention is predominantly driven by first-year Grade Point Average (GPA) performance and administrative factors, specifically the independent selection admission path.Keywords:  Dropout Prediction, Imbalanced Data, SMOTE, Random Forest, Genetic Algorithm
Implementasi Local-First RAG dengan Hybrid Retrieval IndoBERT dan BM25 untuk Pendukung Keputusan Akademik Romi Wahyudi Hasibuan; Ahmad Rio Adriansyah; Henry Saptono
Bulletin of Computer Science Research Vol. 6 No. 4 (2026): June 2026
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bulletincsr.v6i4.1088

Abstract

The need for rapid and accurate access to academic information for educational institution stakeholders, such as campus management and academic advisors, still relies on manual administrative processes, thereby leading to operational inefficiencies. This study develops a locally hosted, open-source Retrieval-Augmented Generation (RAG) system as an automated solution for academic information services while alleviating administrative burdens. The system integrates the Mistral-7B-Instruct LLM with a hybrid search approach within the Elasticsearch ecosystem, combining IndoBERT-based dense retrieval for narrative academic guideline documents and BM25-based sparse retrieval for structured student data. Evaluation was conducted using ROUGE-1, ROUGE-2, and ROUGE-L metrics against 60 test data points generated by Claude Sonnet 4.6. The system successfully answered 58 out of 60 queries, achieving a ROUGE-L f1-score of 0.29. An asymmetrical pattern was observed, where recall values were consistently higher than precision across all metrics, which indicates the impact of the language generation capacity gap between Mistral 7B and the reference model. An average input prompt length ranging from 1,270 to 1,380 tokens contributed to an average latency of 30 seconds per query, representing the primary contemporary challenge of the system. This research is expected to serve as a baseline for developing open-source RAG systems within Indonesian language domains, specifically in the context of higher education academic administration.
Co-Authors ., Suhendi Abdullah, Muhammad Yahya Ahmad Mujahid, Ahmad Al Fajri, Fikri Pratama Al Khairi, Said ALFIAN NUR USYAID Alimi, Aldi Akbar Atmam Al Faruqi, Hamdan Ainul Edi Wibowo Faisal Ayash Fikrian Hamdan Ainul Atmam Al Faruqi Haya Rasikhah Ishom Muhammad Drehem Krisna Panji Lukman Rosyidi Maharani, Siti Zahra Muhammad Abdul Karim Muhammad Fikri Muhammad Rizki Herfian Nabila, Zakiah Prasetyo, Kurniawan Dwi Prima, Pudy Raka Agus Maulana Ridwan Subula Romi Wahyudi Hasibuan Rufaidah Taqiyyah Sahrudin, Asep Saptono, Henry Setyo Nugroho Sirojul Munir Suharya, Suharya Zaki Imaduddin