Article Per Year (5 Year)
p-Index From 2021 - 2026
0.702
P-Index
This Author published in this journals
All Journal INFOMATEK: Jurnal Informatika, Manajemen dan Teknologi Jurnal Sains Riset Journal Serambi Engineering (JSE) Jurnal Bakti Dirgantara
Tri Mulya Dharma
Universitas Jabal Ghafur
Agriculture, Biological Sciences & Forestry Computer Science & IT Earth & Planetary Sciences Economics, Econometrics & Finance Education Electrical & Electronics Engineering Energy Engineering Environmental Science Industrial & Manufacturing Engineering Mechanical Engineering Social Sciences Other

Published : 4 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 4 Documents
Search

 1 
Peningkatan Literasi AI dan Pemanfaatan Prompting Artificial Intelligence Untuk Keterampilan dan Kreativitas Siswa SMK Negeri 3 Sigli Tri Mulya Dharma
Jurnal Bakti Dirgantara Vol. 3 No. 2 (2026): Jurnal Bakti Dirgantara
Publisher : Universitas Dirgantara Marsekal Suryadarma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35968/d1336h80

Abstract

The rapid advancement of Artificial Intelligence (AI) requires students to possess digital literacy. However, initial identification of 10 11th-grade DKV students at SMK Negeri 3 Sigli revealed prompting-writing limitations and a lack of structured AI training due to manual design methods. This Community Service (PKM) program, conducted on July 13, 2026, aimed to enhance AI literacy and prompting techniques for product content creation. The learning by doing method combined theoretical lectures with practical mentoring using Gemini AI. Progress was evaluated via Quizizz pre-tests and post-tests, measuring AI literacy concepts and technical prompting formulation. Quantitative results showed students average AI literacy understanding increased from 30% to 80%, while prompting skills rose from 10% to 65%. This training successfully equipped students with creative, safe, and responsible AI prompting capabilities.
Learning Rate Tuning of Transfer Learning Models for Fresh and Spoiled Beef Image Classification Tri Mulya Dharma; Mukhsin Nuzula; Aviv Fitria Yulia; Dwi Feriyanto; Ningsiah
Jurnal Serambi Engineering Vol. 11 No. 3 (2026): Juli 2026
Publisher : Faculty of Engineering, Universitas Serambi Mekkah

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Beef freshness assessment is a critical aspect of the food industry, where conventional methods relying on visual inspection and olfaction are subjective and prone to human error. This study aims to develop an automated beef freshness classification system utilizing deep learning to enhance the accuracy and objectivity of quality assessments. A total of 2.800 beef images were sourced from an open research data repository, divided equally into fresh and spoiled classes. Three Convolutional Neural Networks (CNN) architectures with transfer learning VGG16, ResNet50-V2, and Inception-V3 were evaluated. The models were systematically tested using learning rate tuning at 0.01, 0.001, and 0.0001 to optimize training convergence. Evaluation results showed that VGG16 outperformed other models in classifying beef freshness. VGG16 achieved a peak testing accuracy of 99.46% at a learning rate of 0.001. The main contribution of this study is the systematic evaluation of transfer learning architectures to establish an optimal baseline for beef quality assessment. By deploying the best performing model into a web based application, this approach offers a practical, objective, and accessible alternative to conventional manual inspection, enabling rapid early detection of beef freshness to improve food safety.
KLASIFIKASI TINGKAT ROASTING BIJI KOPI MENGGUNAKAN METODE DEEP LEARNING BERBASIS CITRA DIGITAL Tri Mulya Dharma; Cut Jora Sari; Salimuddin Salimuddin
Jurnal Sains Riset Vol 16, No 1 (2026): April 2026
Publisher : Lembaga Penelitian dan Pengabdian kepada Masyarakat (LPPM) Universitas Jabal Ghafur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47647/jsr.v16i1.4014

Abstract

Proses pemanggangan merupakan faktor penting dalam menentukan kualitas aroma dan rasa kopi yang khas. Namun, evaluasi biji kopi tradisional seringkali kurang optimal dan subjektif, karena sangat bergantung pada instruktur terlatih. Studi ini bertujuan untuk mengevaluasi kinerja model Deep Learning dalam mengklasifikasikan tingkat pemanggangan biji kopi melalui pengolahan citra digital. Pendekatan eksperimental kuantitatif diterapkan menggunakan tiga arsitektur Convolutional Neural Network (CNN): InceptionV3, MobileNetV2, dan ResNet50. Dataset terdiri dari 1.200 gambar biji kopi yang diklasifikasikan ke dalam empat tingkat warna: hijau, terang, sedang, dan gelap. Hasil pelatihan menunjukkan bahwa model MobileNetV2 berhasil mengenali objek gambar dengan tingkat akurasi tinggi sebesar 96%. Penelitian ini memberikan solusi praktis bagi industri kopi untuk menjaga kualitas produk yang konsisten dan mengurangi ketergantungan pada penilaian manusia yang subjektif.
Analisis Sentimen Terhadap Isu Reshuffle Kapolri Menggunakan Algoritma Machine Learning dan Deep Learning Cut Jora Sari; Tri Mulya Dharma; Salimuddin Salimuddin
INFOMATEK Vol 28 No 2 (2026): Desember 2026
Publisher : Fakultas Teknik, Universitas Pasundan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23969/infomatek.v28i2.70864

Abstract

Dinamika diskursus publik pada platform YouTube terhadap institusi penegak hukum memicu beragam polarisasi opini, khususnya mengenai wacana isu reshuffle Kapolri. Penelitian ini bertujuan untuk menganalisis kecenderungan sentimen publik serta menguji dan membandingkan kinerja enam model machine learning dan satu model deep learning guna menemukan model klasifikasi yang paling handal dalam isu reshuffle Polri. Data dihimpun dari komentar video YouTube berbasis kata kunci terkait, kemudian diproses melalui serangkaian pra-pemrosesan teks dan dibobotkan menggunakan metode pembobotan frekuensi kata. Pelabelan data dilakukan secara otomatis menggunakan kamus InSet Lexicon sentimen bahasa Indonesia, menghasilkan dataset yang didominasi oleh sentimen negatif sebesar 73,17% dan sentimen positif sebesar 26,83%. Dataset berlabel tersebut dibagi ke dalam data latih dan data uji dengan rasio 80:20 untuk menguji model Regresi Logistik, Support Vector Machine, K-Nearest Neighbors, Multinomial Naive Bayes, Random Forest, Decision Tree, dan Multilayer Perceptron. Hasil evaluasi pengujian menunjukkan bahwa model Multinomial Naive Bayes memberikan performa paling optimal dengan tingkat akurasi tertinggi mencapai 82,5% dan nilai F1-score sebesar 80,4%, disusul oleh arsitektur Multilayer Perceptron dengan akurasi 80,9% dan F1-score 80,0%. Sementara itu, model K-Nearest Neighbors mencatatkan kinerja terendah dengan akurasi 74,7%. Kesimpulan dari penelitian ini menegaskan bahwa pendekatan probabilistik Multinomial Naive Bayes merupakan metode yang paling efektif dan stabil untuk mengklasifikasikan data teks opini publik yang tidak seimbang. Temuan ini memberikan implikasi penting bagi pengambil kebijakan dan institusi kepolisian dalam memetakan persepsi publik secara cepat sebagai landasan evaluasi reformasi struktural penegakan hukum.
Co-Authors Aviv Fitria Yulia Cut Jora Sari Cut Jora Sari Dwi Feriyanto Ningsiah Nuzula, Mukhsin Salimuddin Salimuddin Salimuddin Salimuddin