Article Per Year (5 Year)
p-Index From 2021 - 2026
5.745
P-Index
This Author published in this journals
All Journal International Journal of Electrical and Computer Engineering Jurnal Sistem Komputer Bulletin of Electrical Engineering and Informatics Jurnal Informatika Jurnal Ilmiah Teknik Elektro Komputer dan Informatika (JITEKI) Bulletin of Electrical Engineering and Informatics Telematika : Jurnal Informatika dan Teknologi Informasi Sinergi Jurnal Teknologi Informasi dan Ilmu Komputer JUITA : Jurnal Informatika International Journal of Advances in Intelligent Informatics Seminar Nasional Informatika (SEMNASIF) Register: Jurnal Ilmiah Teknologi Sistem Informasi JURNAL NASIONAL TEKNIK ELEKTRO Bulletin of Electrical Engineering and Informatics Jurnal Teknologi dan Sistem Komputer Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) JIKO (Jurnal Informatika dan Komputer) Jurnal Sisfokom (Sistem Informasi dan Komputer) ILKOM Jurnal Ilmiah Compiler MATRIK : Jurnal Manajemen, Teknik Informatika, dan Rekayasa Komputer Jurnal Nasional Pendidikan Teknik Informatika (JANAPATI) GERVASI: Jurnal Pengabdian kepada Masyarakat Systemic: Information System and Informatics Journal Journal of Information Systems and Informatics Buletin Ilmiah Sarjana Teknik Elektro International Journal of Engineering, Technology and Natural Sciences (IJETS) Indonesian Journal of Electrical Engineering and Computer Science International Journal of Advances in Data and Information Systems Journal of Innovation Information Technology and Application (JINITA) Science in Information Technology Letters Masyarakat Berkarya: Jurnal Pengabdian dan Perubahan Sosial JuTISI (Jurnal Teknik Informatika dan Sistem Informasi)
Saifullah, Shoffan
Universitas Pembangunan Nasional Veteran Yogyakarta
Agriculture, Biological Sciences & Forestry Humanities Civil Engineering, Building, Construction & Architecture Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Education Electrical & Electronics Engineering Engineering Industrial & Manufacturing Engineering Languange, Linguistic, Communication & Media Law, Crime, Criminology & Criminal Justice Mathematics Public Health Social Sciences Other

Published : 56 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 56 Documents
Search

 2   3   4   5   6 
EfficientNet B0 Feature Extraction with L2-SVM Classification for Robust Facial Expression Recognition Akbar, Ahmad Taufiq; Saifullah, Shoffan; Prapcoyo, Hari; Rustamadji, Heru; Cahyana, Nur Heri
Journal of Information System and Informatics Vol 7 No 2 (2025): June
Publisher : Universitas Bina Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51519/journalisi.v7i2.1071

Abstract

Facial expression recognition (FER) remains a challenging task due to the subtle visual variations between emotional categories and the constraints of small, controlled datasets. Traditional deep learning approaches often require extensive training, large-scale datasets, and data augmentation to achieve robust generalization. To overcome these limitations, this paper proposes a hybrid FER framework that combines EfficientNet B0 as a deep feature extractor with an L2-regularized Support Vector Machine (L2-SVM) classifier. The model is designed to operate effectively on limited data without the need for end-to-end fine-tuning or augmentation, offering a lightweight and efficient solution for resource-constrained environments. Experimental results on the JAFFE and CK+ benchmark datasets demonstrate the proposed method’s strong performance, achieving up to 100% accuracy across various hold-out splits (90:10, 80:20, 70:30) and 99.8% accuracy under 5-fold cross-validation. Evaluation metrics including precision, recall, and F1-score consistently exceeded 95% across all emotion classes. Confusion matrix analysis revealed perfect classification of high-intensity emotions such as Happiness and Surprise, while minor misclassifications occurred in more ambiguous expressions like Fear and Sadness. These results validate the model’s generalization ability, efficiency, and suitability for real-time FER tasks. Future work will extend the framework to in-the-wild datasets and incorporate model explainability techniques to improve interpretability in practical deployment Keywords: Facial Expression Recognition, EfficientNet, SVM, Deep Features, Emotion Classification
Privacy-Preserving U-Net Variants with pseudo-labeling for radiolucent lesion segmentation in dental CBCT Ismail, Amelia Ritahani; Azlan, Faris Farhan; Noormaizan, Khairul Akmal; Afiqa, Nurul; Nisa, Syed Qamrun; Ghazali, Ahmad Badaruddin; Pranolo, Andri; Saifullah, Shoffan
International Journal of Advances in Intelligent Informatics Vol 11, No 2 (2025): May 2025
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/ijain.v11i2.1529

Abstract

Accurate segmentation of radiolucent lesions in dental Cone-Beam Computed Tomography (CBCT) is vital for enhancing diagnostic reliability and reducing the burden on clinicians. This study proposes a privacy-preserving segmentation framework leveraging multiple U-Net variants—U-Net, DoubleU-Net, U2-Net, and Spatial Attention U-Net (SA-UNet)—to address challenges posed by limited labeled data and patient confidentiality concerns. To safeguard sensitive information, Differential Privacy Stochastic Gradient Descent (DP-SGD) is integrated using TensorFlow-Privacy, achieving a privacy budget of ε ≈ 1.5 with minimal performance degradation. Among the evaluated architectures, U2-Net demonstrates superior segmentation performance with a Dice coefficient of 0.833 and an Intersection over Union (IoU) of 0.881, showing less than 2% reduction under privacy constraints. To mitigate data annotation scarcity, a pseudo-labeling approach is implemented within an MLOps pipeline, enabling semi-supervised learning from unlabeled CBCT images. Over three iterative refinements, the pseudo-labeling strategy reduces validation loss by 14.4% and improves Dice score by 2.6%, demonstrating its effectiveness. Additionally, comparative evaluations reveal that SA-UNet offers competitive accuracy with faster inference time (22 ms per slice), making it suitable for low-resource deployments. The proposed approach presents a scalable and privacy-compliant framework for radiolucent lesion segmentation, supporting clinical decision-making in real-world dental imaging scenarios.
Urban Traffic Volume Prediction using LSTM and Bi-LSTM: Performance Evaluation on the Metro Interstate Dataset Pranolo, Andri; Saifullah, Shoffan; Putra, Agung Bella Utama; Dreżewski, Rafał; Wibawa, Aji Prasetya
ILKOM Jurnal Ilmiah Vol 17, No 3 (2025)
Publisher : Prodi Teknik Informatika FIK Universitas Muslim Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33096/ilkom.v17i3.3001.227-240

Abstract

Urban traffic forecasting underpins the mitigation of congestion, enhancement of road safety, and reduction of emissions in intelligent transportation systems. We benchmark Long Short-Term Memory (LSTM) and Bidirectional LSTM (Bi-LSTM) models on the Metro Interstate Traffic Volume dataset under an identical preprocessing and training pipeline for a fair comparison. Using a 24-hour multivariate input window (temperature, rainfall, snowfall, cloud cover), LSTM delivers the best overall balance of accuracy and efficiency on the full test sequence (RMSE = 0.196, MAPE = 2.36%, R² = 0.480; 7,344 s training). Bi-LSTM achieves competitive short-window accuracy but underperforms on the full sequence (RMSE = 0.231, MAPE = 2.92%, R² = 0.280; 12,672 s training). We attribute the Bi-LSTM gap to prediction "flattening" over long horizons, i.e., over-smoothed peaks from bidirectional averaging, despite its slightly stronger short-segment fit. Compared with prior RNN/GRU/CNN baselines on the same data, LSTM improves variance explanation while remaining deployable for near-real-time use. We also examine seasonality (daily/weekly cycles), weather effects, and data imbalance (peak versus off-peak) as factors that shape model error. These results support LSTM as a practical default for city-scale forecasting and motivate future work with attention/Transformer encoders and richer exogenous signals (incidents, events). The findings inform policy by enabling proactive traffic management that can reduce delays, emissions, and crash risk through earlier, data-driven interventions.
Fuzzy Inference System Mamdani dalam Prediksi Produksi Kain Tenun Menggunakan Rule Berdasarkan Random Tree Tundo, Tundo; Saifullah, Shoffan
Jurnal Teknologi Informasi dan Ilmu Komputer Vol 9 No 3: Juni 2022
Publisher : Fakultas Ilmu Komputer, Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25126/jtiik.2022924212

Abstract

Kain tenun merupakan salah satu produk yang diminati oleh banyak orang. Hal ini menjadi pemicu produsen untuk meningkatkan pengelolahannya. Salah satu usaha yang dilakukan adalah memprediksi produksi yang dapat dilakukan untuk mendapatkan jumlah optimal yang diperoleh, sehingga mendapatkan keuntungan yang besar. Dalam penelitian ini, untuk mendapatkan prediksi jumlah produksi kain tenun dilakukan dengan perhitungan komputerisasi menggunakan metode logika fuzzy Mamdani. Metode ini menggunakan konsep pohon keputusan random tree dalam membentuk rule. Rule yang dibuat berdasarkan pada kriteria dalam penentuan jumlah produksi kain tenun, diantaranya yaitu biaya produksi, permintaan, dan stok. Konsep pohon keputusan random tree dalam penelitian ini digunakan untuk membuat rule secara otomatis berdasarkan data yang tersedia. Pembentukan rule ini berdasarkan data-data kain tenun dan diimplementasikan dalam random tree, sehingga tidak perlu menggunakan pakar. Penelitian ini membuktikan bahwa prediksi yang dilakukan dapat membangun rule dengan nilai akurasi sebesar 100%. Hasil perbandingan prediksi dengan produksi sesungguhnya memiliki persentase error sebesar 3% dengan nilai kebenaran sebesar 97% (berdasarkan perhitungan Average Forecasting Error Rate (AFER)). Oleh karena itu ketika diimplementasikan dalam fuzzy Mamdani dapat menghasilkan prediksi produksi kain tenun yang optimal. AbstractWoven fabric is a product that is in demand by many people. It triggers producers to improve their management. One of the efforts made is to predict the production that can be done to get the optimal amount obtained, to get a significant profit. In this study, to obtain a prediction of the amount of woven fabric production is done by computerized calculations using the Mamdani fuzzy logic method. This method uses the concept of a random tree decision tree in forming rules. The rules are made based on the criteria in determining the amount of woven fabric production, including production costs, demand, and stock. The concept of a random tree decision tree in this study automatically generates rules based on available data. This rule's formation is based on woven fabric data and is implemented in a random tree, so there is no need to use experts. This study shows that the predictions made can build rules with an accuracy value of 100%. The comparison of predictions with actual production has an error percentage of 3% with a truth value of 97% (based on the calculation of the Average Forecasting Error Rate (AFER)). When implemented in Fuzzy Mamdani, it can produce optimal woven fabric production predictions with predicted results less than the actual production.
Robust Classification of Beef and Pork Images Using EfficientNet B0 Feature Extraction and Ensemble Learning with Visual Interpretation Taufiq Akbar, Ahmad; Saifullah, Shoffan; Prapcoyo, Hari; Yuwono, Bambang; Rustamaji, Heru Cahya
Register: Jurnal Ilmiah Teknologi Sistem Informasi Vol 11 No 1 (2025): January
Publisher : Information Systems - Universitas Pesantren Tinggi Darul Ulum

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26594/register.v11i1.4045

Abstract

Distinguishing between beef and pork based on image appearance is a critical task in food authentication, but it remains challenging due to visual similarities in color and texture, especially under varying lighting and capture conditions. To address these challenges, we propose a robust classification framework that utilizes EfficientNet B0 as a deep feature extractor, combined with an ensemble of Regularized Linear Discriminant Analysis (RLDA), Support Vector Machine (SVM), and Random Forest (RF) classifiers using soft voting to enhance generalization performance. To improve interpretability, we incorporate Gradient-weighted Class Activation Mapping (Grad-CAM) to visualize classification decisions and validate that the model focuses on relevant regions of the meat, such as red-channel intensity and muscle structure. The proposed method was evaluated on a public dataset containing 400 images evenly split between beef and pork. It achieved a hold-out accuracy of 99.0% and a ROC-AUC of 0.995, outperforming individual learners and demonstrating strong resilience to limited data and variation in imaging conditions. By integrating efficient transfer learning, ensemble decision-making, and visual interpretability, this framework provides a powerful and transparent solution for binary meat classification. Future work will focus on fine-tuning the CNN backbone, applying GAN-based augmentation, and extending the approach to multiclass meat authentication tasks.
Visualization of Islamic Boarding School Location at Yogyakarta with Web-Based Geodesain Alfiani, Oktavia Dewi; Wahyuningrum, Dwi; Saifullah, Shoffan; Haekal, Haekal
Telematika Vol 20 No 3 (2023): Edisi Oktober 2023
Publisher : Jurusan Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31315/telematika.v20i3.10885

Abstract

Purpose: This research produces a webGIS design that presents the geospatial location of buildings in the Krapyak Yogyakarta boarding school area to facilitate users outside the area when heading to the location of the boarding school whose buildings are scattered.Design/methodology/approach: By combining aerial photos from UAV mapping with Open Street Map. The combined results of both maps are presented in a webGIS built from HTML, CSS and OpenLayers scripting.Findings/result: Building a webGIS to present information on the location of Krapyak Islamic boarding schools that has been equipped with corrected coordinates and routes from the iconic city of Yogyakarta so that immigrants from outside the area can easily understand the use of the webgis. Originality/value/state of the art: From previous research, webGIS development only uses maps presented through openstreetmap where if users use existing online navigation applications have different coordinate system references (Soraya R, 2018). So by equalizing the map reference by combining the results of UAV mapping and correcting the shape of the building presented on openstreetmap, the spatial information from the webgis will have a position accuracy that is more in line with the truth.
Geographic-Origin Music Classification from Numerical Audio Features: Integrating Unsupervised Clustering with Supervised Models Pranolo, Andri; Sularso, Sularso; Anwar, Nuril; Putra, Agung Bella Utama; Wibawa, Aji Prasetya; Saifullah, Shoffan; Dreżewski, Rafał; Nuryana, Zalik; Andi, Tri
Buletin Ilmiah Sarjana Teknik Elektro Vol. 7 No. 4 (2025): December
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/biste.v7i4.13400

Abstract

Classifying the geographic origin of music is a relevant task in music information retrieval, yet most studies have focused on genre or style recognition rather than regional origin. This study evaluates Support Vector Machine (SVM) and Convolutional Neural Network (CNN) models on the UCI Geographical Origin of Music dataset (1,059 tracks from 33 non-Western regions) using numerical audio features. To incorporate latent structure, we first applied K-means clustering with the optimal number of clusters (k=2) determined by the Elbow and Silhouette methods. The cluster assignments were used as auxiliary signals for training, while evaluation relied on the true region labels. Classification performance was assessed with Accuracy, Precision, Recall, and F1-score. Results show that SVM achieved 99.53% accuracy (95% CI: 97.38–99.92%), while CNN reached 98.58% accuracy (95% CI: 95.92–99.52%); Precision, Recall, and F1 mirrored these values. The differences confirm SVM’s superior performance on this dataset, though the near-perfect scores also suggest strong separability in the feature space and potential risks of overfitting. Learning-curve analysis indicated stable training, and cluster supervision provided small but consistent benefits. Overall, SVM remains a reliable baseline for tabular music features, while CNNs may require spectro-temporal representations to leverage their full potential. Future work should validate these findings across multiple datasets, apply cross-validation with statistical significance testing, and explore hybrid deep models for broader generalization.
Otsu Method for Chicken Egg Embryo Detection based-on Increase Image Quality Suhirman Suhirman; Shoffan Saifullah; Ahmad Tri Hidayat; Rr Hajar Puji Sejati
MATRIK : Jurnal Manajemen, Teknik Informatika dan Rekayasa Komputer Vol. 21 No. 2 (2022)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/matrik.v21i2.1724

Abstract

Detection of chicken egg embryos using image processing has limitations and needs some processes for improvement. By human vision, the previous process used binoculars and candling using light/beams directed at the chicken eggs in the incubator. In this study, we propose the application of image segmentation using the Otsu method in detecting chicken egg embryos. This method uses image segmentation with increased image quality (preprocessing) by several methods such as resizing, grayscaling, image adjustment, and image enhancement. These processes produce a better image and can be used for input in the segmentation process. In addition, this study compares several segmentation methods in detecting chicken egg embryos, such as thresholding, Otsu basic, and k-means clustering. The results show that our proposed method produced segmentation images to detect chicken egg embryos of 200 datasets images. This method has a faster process and can create a uniform segmentation than other methods. However, other methods can also detect chicken egg embryos. The method’s accuracy proposed in this study increased by 1.5% compared to other methods. In addition, the resulting SSIM value has a percentage close to and more than 90%, which means that the segmentation of the results obtained can be used to detect chicken egg embryos.
Essay auto-scoring using N-Gram and Jaro Winkler based Indonesian Typos Herlina Jayadianti; Budi Santosa; Judanti Cahyaning; Shoffan Saifullah; Rafal Drezewski
MATRIK : Jurnal Manajemen, Teknik Informatika dan Rekayasa Komputer Vol. 22 No. 2 (2023)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/matrik.v22i2.2473

Abstract

Writing errors on e-essay exams reduce scores. Thus, detecting and correcting errors automatically in writing answers is necessary. The implementation of Levenshtein Distance and N-Gram can detect writing errors. However, this process needed a long time because of the distance method used. Therefore, this research aims to hybrid Jaro Winker and N-Gram methods to detect and correct writing errors automatically. This process required preprocessing and finding the best word recommendations by the Jaro Winkler method, which refers to Kamus Besar Bahasa Indonesia (KBBI). The N-Gram method refers to the corpus. The final scoring used the Vector Space Model (VSM) method based on the similarity of words between the answer keys and the respondent’s answers. Datasets used 115 answers from 23 respondents with some writing errors. The results of Jaro Winkler and N-Gram methods are good in detecting and correcting Indonesian words with the accuracy of detection averages of 83.64% (minimum of 57.14% and maximum of 100.00%). In contrast, the error correction accuracy averages 78.44% (minimum of 40.00% and maximum of 100.00%). However, Natural Language Processing (NLP) needs to improve these results for word recommendations.
Sentiment analysis of Indonesian reviews using fine-tuning IndoBERT and R-CNN Jayadianti, Herlina; Kaswidjanti, Wilis; Utomo, Agung Tri; Saifullah, Shoffan; Dwiyanto, Felix Andika; Drezewski, Rafal
ILKOM Jurnal Ilmiah Vol 14, No 3 (2022)
Publisher : Prodi Teknik Informatika FIK Universitas Muslim Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33096/ilkom.v14i3.1505.348-354

Abstract

Reviews are a form of user experience information on a product or service that can be used as a reference for potential consumers’ preferences to buy, use, or consume a product. They can be also used by business entities to find out public opinion about their product or the performance of their business products. It will be very difficult to process the review data manually and it will take a long time. Therefore, sentiment analysis automation can be used to get polarity information from existing reviews. In this study, IndoBERT with Recurrent Convolutional Neural Network (RCNN) was used to automate sentiment analysis of Indonesian reviews. The data used was a sentiment analysis dataset obtained from IndoNLU with sentiment consisting of negative sentiment, neutral sentiment, and positive sentiment. The results of the test showed that IndoBERT with the Recurrent Convolutional Neural Network (RCNN) had better results than the IndoBERT base. IndoBERT with Recurrent Convolutional Neural Network (RCNN) obtained 95.16% accuracy, 94.05% precision, 92.74% recall and 93.27% f1 score.
Co-Authors Abdul Fadlil Adityo Nugroho, Adityo Afiqa, Nurul Agus Sasmito Aribowo Ahmad Taufiq Akbar Ahmad Tri Hidayat Aji Prasetya Wibawa Akbar, Bagus Muhammad Alek Setiyo Nugroho Alfiani, Oktavia Dewi Alin Khaliduzzaman Alin Khaliduzzaman Alisya Amalia Putri Hasanah Andi Muhammad Dirham Dewantara Andiko Putro Suryotomo Andri Pranolo Anton Satria Prabuwono Anton Yudhana Arianti, Berliana Andra Arief Hermawan Arif - Umami Awang Hendrianto Pratomo Azlan, Faris Farhan Azrul Mahfurdz Bambang Yuwono Betty Yel, Mesra Budi Santosa Devia, Elmi Dharmawan, Tio Dreżewski, RafaÅ‚ Drezewski, Rafal Drezewski, Rafał Dwi Wahyuningrum Dwiyanto, Felix Andika Faqihuddin Al-anshori Ghazali, Ahmad Badaruddin Haekal, Haekal Herlina Jayadianti Heru Cahya Rustamaji Hidayat, Ahmad Tri Humairoh, Nanda Lailatul Ismail, Amelia Ritahani Isna Nur Aini Ivana Puspita Sari Judanti Cahyaning Junaidi Junaidi Kusuma, M. Apriandi Lean Karlo Tolentino Luh Putu Ratna Sundari Muhammad Nur Hendra Alvianto Nathalie Japkowicz Nisa, Syed Qamrun Noormaizan, Khairul Akmal Nur Heri Cahyana Nuril Anwar, Nuril Nuryana, Zalik Opi Irawansah Prapcoyo, Hari Putra, Agung Bella Utama Rabbimov Ilyos Rabbimov, Ilyos Rafal Drezewski Rafal Drezewski Rafal Drezewski Rochmat Husaini Rochmat Husaini Rustamadji, Heru Saidah, Andi Santosa, Budi Satya Ghifari Adipratama Seno Aji Putra Suhirman SUHIRMAN SUHIRMAN Sularso Sularso, Sularso Sunardi - Sunardi - Sunardi Sunardi Sunardi, Sunardi Taufiq Akbar, Ahmad Tri Andi, Tri Tundo Tundo Tundo, Tundo Tuti Purwaningsih, Tuti Wahyu Adjie Saputra Wilis Kaswidjanti Wilis Kaswidjanti Wilis Kaswijanti Yuhefizar Yuhefizar Yuli Fauziah Yuli Fauziyah Zulfikar Yusya Mubarak