Article Per Year (5 Year)
p-Index From 2021 - 2026
9.028
P-Index
This Author published in this journals
All Journal Jurnal Media Infotama Bulletin of Electrical Engineering and Informatics Jurnal Pseudocode Jurnal Informatika Upgris Sistemasi: Jurnal Sistem Informasi Journal of Economic, Bussines and Accounting (COSTING) INTECOMS: Journal of Information Technology and Computer Science JSAI (Journal Scientific and Applied Informatics) Journal of Technopreneurship and Information System (JTIS) Just TI (Jurnal Sains Terapan Teknologi Informasi) Jurnal Abdimas Serawai Jurnal Sistem Informasi Dan E-Bisnis Jurnal Ilmu Komputer Dan Informatika JPMTT (Jurnal Pengabdian Masyarakat Teknologi Terbarukan) JCOSIS (Journal Computer Science and Information Systems) Jurnal Komputer, Informasi dan Teknologi
Putra, Erwin Dwika
Program Studi Teknik Informatika, Fakultas Teknik, Universitas Muhammadiyah Bengkulu
Religion Humanities Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Education Electrical & Electronics Engineering Engineering Environmental Science Languange, Linguistic, Communication & Media Law, Crime, Criminology & Criminal Justice Library & Information Science Nursing Public Health Social Sciences

Published : 71 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 71 Documents
Search

 4   5   6   7   8 
Pengembangan Metode Deep Metric Learning Berbasis Attention untuk Verifikasi Tanda Tangan Offline pada Lingkungan Data Terbatas Marissa Utami; Erwin Dwika Putra
JSAI (Journal Scientific and Applied Informatics) Vol 9 No 2 (2026): Juni
Publisher : Fakultas Teknik Universitas Muhammadiyah Bengkulu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36085/jsai.v9i2.10792

Abstract

This study aims to develop an attention-based Deep Metric Learning method for offline signature verification in environments with limited data availability. The main challenge in offline signature verification lies in the high intra-class variation and the similarity between genuine and forged signatures, particularly when the amount of training data is limited. The proposed method employs a Siamese Convolutional Neural Network architecture combined with an attention mechanism to enhance discriminative feature extraction capabilities. The dataset used in this study was obtained from offline sources and Kaggle, consisting of genuine and forged signature images. The research process includes preprocessing, signature pair generation, feature extraction, embedding generation using Deep Metric Learning, and optimization using Contrastive Loss. Experimental results demonstrate that the proposed method achieved an Accuracy of 91.12%, Precision of 92.27%, Recall of 92.43%, F1-score of 90.75%, and an Equal Error Rate (EER) of 4.88%. These results indicate that the integration of the attention mechanism and Deep Metric Learning effectively improves the system's capability to recognize signature patterns under limited data conditions.
Penerapan Metode Ekstraksi Fitur Geometris, Hog, dan Hu Moment Pada Citra Tanda Tangan Digital Menggunakan Support Vector Machine (SVM) Muhammad Hikmal Febrian; Erwin Dwika Putra; Ardi Wijaya; Muntahanah
JSAI (Journal Scientific and Applied Informatics) Vol 9 No 2 (2026): Juni
Publisher : Fakultas Teknik Universitas Muhammadiyah Bengkulu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36085/jsai.v9i2.10910

Abstract

The increasing use of electronic documents has heightened the need for fast, accurate, and objective digital signature verification systems. This study proposes a digital signature recognition system by combining Geometric Features, Histogram of Oriented Gradients (HOG), and Hu Moment feature extraction with a Support Vector Machine (SVM) classifier using the Radial Basis Function (RBF) kernel. A dataset of 500 signature images from 50 individuals was divided into training, validation, and testing sets using an 80:10:10 ratio. The proposed workflow includes image preprocessing, feature extraction, feature vector construction, model training, and evaluation using a confusion matrix. Experimental results show that the combined feature extraction methods effectively represent both global and local signature characteristics. The proposed model correctly classified 46 of 50 testing samples, achieving 92.00% accuracy, 88.00% precision, 92.00% recall, and an 89.33% F1-score, demonstrating its effectiveness for automatic digital signature recognition and electronic document authentication.
IMPLEMENTASI CONVOLUTIONAL NEURAL NETWORK MENGGUNAKAN RESNET 50 UNTUK MENGKLASIFIKASI TINGKAT KEMANTANGAN  BUAH PEPAYA Mellysa Pratama; Erwin Dwika Putra
Journal of Economic, Bussines and Accounting (COSTING) Vol. 9 No. 3 (2026): Journal of Economic, Bussines and Accounting (COSTING)
Publisher : Institut Penelitian Matematika, Komputer, Keperawatan, Pendidikan dan Ekonomi (IPM2KPE)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31539/nvnckc75

Abstract

Buah pepaya adalah salah satu jenis buah tropis yang banyak disukai karena kaya akan nutrisi. Namun, cara menilai kematangan buah pepaya masih sering dilakukan dengan cara manual, yang dapat menyebabkan kesalahan dalam proses pemisahan dan penyebarannya. Dengan demikian, penelitian ini bertujuan untuk mengembangkan sistem otomatis untuk mengklasifikasikan tingkat kematangan buah pepaya dengan menggunakan metode Convolutional Neural Network (CNN) yang berbasis arsitektur ResNet50. Dataset yang digunakan terdiri dari gambar buah pepaya yang dibagi menjadi empat tingkat kematangan, yaitu masih mentah, setengah matang, sudah matang, dan sudah busuk. Gambar tersebut kemudian menjalani proses preprocessing yang mencakup pengubahan ukuran gambar menjadi 224×224 piksel, penyesuaian nilai piksel, serta augmentasi data melalui teknik seperti rotasi, zoom, dan pembalikan horizontal untuk memperbanyak variasi data pelatihan. Model dilatih dengan menggunakan metode transfer learning dengan memanfaatkan bobot yang sudah ada dari dataset ImageNet. Evaluasi kinerja model dilakukan melalui penggunaan matriks kebingungan dan matriks klasifikasi yang mencakup akurasi, precision , recall, dan skor F1. Hasil dari proses pelatihan menunjukkan bahwa model mencapai akurasi pelatihan sebesar 91,72% dan akurasi validasi sebesar 83,56%, dengan nilai  validasition loss sebesar 0,4958. Temuan ini mengindikasikan bahwa model dapat mengklasifikasikan citra buah pepaya dengan performa yang relatif baik dan konsisten. Penelitian ini diharapkan dapat mendukung proses otomatisasi dalam mengidentifikasi tingkat kematangan buah pepaya di sektor pertanian dan industri pangan. Kata Kunci: klasifikasi citra , buah pepaya ,CNN, Resnet-50,deep learning.
IMPLEMENTASI CONVOLUTIONAL NEURAL NETWORK MENGGUNAKAN RESNET 50 UNTUK MENGKLASIFIKASI TINGKAT KEMANTANGAN  BUAH PEPAYA Mellysa Pratama; Erwin Dwika Putra
INTECOMS: Journal of Information Technology and Computer Science Vol. 9 No. 3 (2026): INTECOMS: Journal of Information Technology and Computer Science
Publisher : Institut Penelitian Matematika, Komputer, Keperawatan, Pendidikan dan Ekonomi (IPM2KPE)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31539/tr9sbe10

Abstract

Papaya is a very popular tropical fruit variety because it is rich in nutrients. However, the method of assessing the ripeness of papaya fruit is still often done manually, which can cause errors in the separation and distribution process. Thus, this study aims to develop an automatic system to classify the ripeness level of papaya fruit using the Convolutional Neural Network (CNN) method based on the ResNet50 architecture. The dataset used consists of papaya fruit images divided into four stages of ripeness, namely unripe, half-ripe, and unfit. The images then undergo a preprocessing process that includes resizing the image to 224 × 224 pixels, adjusting pixel values, and data augmentation through techniques such as rotation, zoom, and horizontal flipping to increase the variety of training data. The model is trained using a transfer learning approach by utilizing existing weights from the imagenet dataset. Model performance evaluation is carried out through the use of a confusion matrix and a classification matrix that includes accuracy, precision, recall, and F1 score. The results of the training process show that the model achieved a training accuracy of 91.72% and a validation accuracy of 83.56%, with a validation loss value of 0.4958. These findings indicate that the model can classify papaya fruit images with relatively good and consistent performance. This research is expected to support the automation process in identifying the ripeness level of papaya fruit in the agricultural and food industry sectors.  Keywords: image classification, papaya, CNN, Resnet-50, deep learning.  
Analisis Perbandingan Metode Multimedia Development Live Cycle Pada Augmented Reality Naufal Afif Hawari; Erwin Dwika Putra
Jurnal Media Infotama Vol 18 No 1 (2022): April
Publisher : UNIVED Press

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37676/jmi.v18i1.1759

Abstract

Indonesia is one of the countries that has high biodiversity. This biodiversity is spread throughout the territory of Indonesia. However, endemic and rare animals in Indonesia are increasingly decreasing due to human greed. If this continues, the endangered species are likely to become extinct. therefore the need for an information media that is different from the usual, namely with Augmented Reality. Augmented Reality is a technology that can combine the real world with the virtual world and can combine the real world with virtual objects in it. In this research, augmented reality is used as a media of information and to learn about the biodiversity of the people, especially the animals that exist in Indonesia. This study aims to test the best method for making Augmented Reality. In this application, it displays 3D objects from deer animals. There are 4 menus in the application, one of which is detailed information about deer animals. objects can appear if the marker can be detected properly in the program. Key Words : Augmented Reality, Multimedia Development Live Cycle, Android
Segmentasi Warna Kulit Menggunakan Ruang Warna YCBCR Untuk Deteksi Wajah Manusia Yovi Apridiansyah; erwin dwika putra; Diana Diana; Angtyas Candra Pratama
Jurnal Media Infotama Vol 19 No 1 (2023): April
Publisher : UNIVED Press

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37676/jmi.v19i1.3808

Abstract

Image processing is a system where the process is done by entering in the form of imagery and the result is also in the form of imagery. At first this image processing was done to improve the quality of the image, but with the development of the world of computing which is characterized by increasing the capacity and speed of computer processes. In this research image processing will be used for skin color detection which is the initial process in image processing. Success in detecting pixel imagery can be a category of skin or not skin seen from the results of processing processes such as the process of detecting human faces. Skin detection is also a process of finding skin pixels of color in the area of the image or video, this process is usually used as a preprocessing step to find areas that could potentially have a human face. In this study by applying the color space YcbCr we were able to detect faces in humans with a high percentage accuracy rate with a precession percentage of 77%, Recal 87% and an accuracy rate of 72% from the results of 10 image samples.
Implementasi Algoritma Weighted Product Untuk Penilaian Quis Pada Aplikasi Belajar Online Siswa Sekolah Menengah Kejuruan Yuza Reswan; Pahrizal Pahrizal; Erwin Dwika Putra; Sahrudin Sahrudin
Jurnal Media Infotama Vol 20 No 1 (2024): April 2024
Publisher : UNIVED Press

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37676/jmi.v20i1.5701

Abstract

Online learning media (Online learning) uses a technological approach that is more suited to today's student learning systems compared to conservative methods of learning in the classroom. By looking at the varying number of questions, the weighting of the exercises varies with the number of questions contained in the exercises. As for the examples of questions that consist of 20 questions, each question is worth 5, so if the student answers 20 questions correctly then the student gets a score of 100. The problems encountered in research on citizenship education learning are still manual so that students' understanding of the concepts and rules of citizenship education is still limited. less because it is rote. The problem formulation based on the background above is that the weighting of the questions and answers is not yet known. The benefit of this research is as an information medium to introduce quiz exercises in applications with citizenship education subject matter for students at Vocational High School 08, North Bengkulu Regency with the concept of learning while playing.
Analisis Kesadaran Masyarakat terhadap Program Pengelolaan Sampah Menggunakan Integrasi Sentiment Analysis dan BERTopic pada Data Media Sosial Indonesia Erwin Dwika Putra; Marissa Utami
Jurnal Sistem Informasi dan E-Bisnis Vol 8 No 2 (2026): June
Publisher : LPPMPP Yayasan Sejahtera Bersama Bengkulu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54650/jusibi.v8i2.697

Abstract

Waste management remains one of the major challenges in achieving sustainable development, where the success of waste management programs is highly dependent on public awareness and participation. With the rapid growth of social media usage, public conversations generated on digital platforms provide valuable insights into community perceptions regarding environmental issues. This study aims to analyze public awareness of waste management programs by integrating IndoBERT-based Sentiment Analysis and BERTopic-based Topic Modeling on Indonesian social media data. Data were collected from the X (Twitter) platform using waste-management-related keywords and processed through several preprocessing stages, including text cleaning, tokenization, stopword removal, and normalization. The sentiment analysis results demonstrate that the IndoBERT model achieved an accuracy of 93.82%, with sentiment distribution consisting of 44.88% positive, 29.84% neutral, and 25.28% negative sentiments. Furthermore, BERTopic successfully identified five dominant discussion topics, namely waste banks, recycling, temporary and final disposal sites (TPS and TPA), urban sanitation facilities, and environmental education. The integration of sentiment and topic analysis produced an Awareness Index score of 0.694, which falls into the high-awareness category. The findings indicate that Indonesian society generally demonstrates a positive perception and a relatively high level of awareness toward waste management programs, although concerns regarding operational services and waste management infrastructure remain evident. This study contributes a novel approach by combining IndoBERT, BERTopic, and an Awareness Index framework to provide a comprehensive assessment of public awareness based on social media analytics, which can support evidence-based policymaking and sustainable waste management strategies.
Analisis Kesesuaian Enterprise Architecture terhadap Standar TOGAF Menggunakan Sentence-BERT dan Semantic Similarity marissa utami; Erwin Dwika Putra
Jurnal Sistem Informasi dan E-Bisnis Vol 8 No 2 (2026): June
Publisher : LPPMPP Yayasan Sejahtera Bersama Bengkulu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54650/jusibi.v8i2.698

Abstract

Digital transformation encourages organizations to implement Enterprise Architecture (EA) to align business strategies with information technology. However, evaluating EA documents against TOGAF standards is still largely performed manually, making the process time-consuming and potentially subjective. This study aims to analyze the compliance of Enterprise Architecture with TOGAF standards using Sentence-BERT and Semantic Similarity. The dataset consists of TOGAF standard documents and publicly available Enterprise Architecture documents. The proposed method includes text preprocessing, sentence embedding generation using Sentence-BERT, document similarity measurement through Cosine Similarity, and calculation of the Enterprise Architecture Compliance Score (EACS). The results show that Business Architecture achieved the highest similarity score of 0.86, followed by Application Architecture (0.82), Technology Architecture (0.78), and Data Architecture (0.74). The calculated EACS value of 0.80 indicates a high level of compliance with TOGAF standards. These findings demonstrate that the Sentence-BERT-based approach effectively captures semantic similarities between documents and provides a more objective evaluation compared to conventional methods. The main contribution of this study is the development of EACS as a quantitative indicator to support Enterprise Architecture evaluation and governance in a more efficient and measurable manner.
Systematic Literature Review terhadap Model Deep Learning untuk Smart Waste Classification Erwin Dwika Putra; Anita Septiani Putri
JUKOMIKA (Jurnal Ilmu Komputer dan Informatika) Vol. 9 No. 1 (2026): July
Publisher : LPPMPP Yayasan Sejahtera Bersama Bengkulu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54650/jukomika.v9i1.715

Abstract

The rapid advancement of Deep Learning has significantly contributed to the development of Smart Waste Classification systems for improving automated waste sorting and sustainable waste management. However, comprehensive studies that systematically summarize the evolution of Deep Learning models, datasets, evaluation methods, and future research directions remain limited. This study aims to conduct a Systematic Literature Review (SLR) on Deep Learning models applied to Smart Waste Classification by following the PRISMA 2020 guidelines. The literature search was performed using Publish or Perish across several scientific databases, covering publications from 2020 to 2025. A total of 100 studies were initially identified, and after the identification, screening, eligibility, and inclusion processes, 40 studies were selected for qualitative synthesis. The findings indicate that Convolutional Neural Network (CNN) remains the most widely adopted architecture, followed by ResNet, MobileNet, EfficientNet, YOLO, and Vision Transformer. TrashNet is the most frequently used dataset, while accuracy, precision, recall, F1-score, and mean Average Precision (mAP) are the dominant evaluation metrics. Current research trends emphasize transfer learning, lightweight architectures, and the integration of Deep Learning with the Internet of Things (IoT) and edge computing. This review provides comprehensive insights into recent developments and identifies research opportunities for developing more accurate, efficient, and practical Smart Waste Classification systems. Keywords— Systematic Literature Review (SLR); Deep Learning; Smart Waste Classification; Convolutional Neural Network (CNN); Computer Vision; Transfer Learning; Waste Management.
Co-Authors Abdiansah, Abdiansah Abdullah, Dedy Abdullah, Dedy Ade Rangga Saputera Agastra Galih Setiawan Ahmad Sayyeid Al Jadd Alber Uci Saputra Altra Yudha Mawlana Andi Nugroho Angtyas Candra Pratama Anita Septiani Putri Apridiansyah, Yovi Ayumi, Vina Azildjian Arma Yuda Azildjian Arma Yuda Bony Triwahyuda Charles Roenal Krisubiyantoro Dede Erwan Dedy Abdullah Dedy Abdullah Dedy Agung Prabowo Depmi Hardianto Deslianti, Dwita Deslianti, Dwita Diana Diana Ermatita - Erzi Hidayat Fikri Agnesa Putra Filda Rahayu Firdianti Sukemi Hadiguna Setiawan Handrie Noprison Hardianto, Depmi Harry Witriyono Harry Witriyono Herianto Herianto hidayah, agung kharisma Ilham Fahriansyah Ilham Fahriansyah Ilham Fahriansyah, Ruliyani Jastrawan, Noris Jefri Zulkarnain Khairullah khairullah Krismiyani Krismiyani Krismiyani, Krismiyani M Khairunnas M. Alfarisi M. Alfarisi M. Husni Rifqo Mariana Purba Mariana Purba Marissa Utami Marrisa Utami Mellysa Pratama Moh. Rere Valentino Zantohar Monsya Juansen Muhammad Hikmal Febrian Muhammad Husni Rifqo Muhammad Husni Rifqo Muntahanah Naufal Afif Hawari Noprisson, Handrie Noris Jastrawan Nuri David Maria Veronika Pahrizal, Pahrizal Pitersa Susanto Purba, Mariana Putra, Zulhamdi Rahmat Arif Ma’ruf Rega Satya Putra Rifqo, Muhammad Husni Rio Eka Prayuda Rizky Rahmat Saputra Rizky Rahmat Saputra Ruliyani Sahrudin Sahrudin Sandhy Fernandez Sarwati Rahayu Sonita, Anisya Stefanus Santosa Sulis Sandiwarno ujang juhardi Umniy Salamah Vendi Handoyo Handoyo Veronika, Nuri David Maria Wachyu Hari Haji Wahyu Adianto Wijaya, Ardi Yetman Erwadi Yulia Darmi Yuza Reswan Zalia Apriyanti Zulhamdi Putra