Claim Missing Document
Check
Articles

Found 14 Documents
Search

An HHO-Optimized LSTM Framework for Predicting Adverse Effects Associated with Reproductive and Breast Disorders ANGEL METANOSA AFINDA; IGA NARENDRA PRAMAWIJAYA; FAUZAN FIRDAUS
ELKOMIKA: Jurnal Teknik Energi Elektrik, Teknik Telekomunikasi, & Teknik Elektronika Vol 14, No 3: Published July 2026
Publisher : Institut Teknologi Nasional, Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26760/elkomika.v14i3.358

Abstract

Reproductive toxicity prediction is a major challenge in drug development as side effects are often difficult to detect early. SMILES representations provide a compact sequential format suitable for deep learning. This study proposes an HHO-optimized LSTM model to predict reproductive and breast-related side effects. Four architectural schemes were evaluated including L (LSTM only), CL (Convolution + LSTM), LD (LSTM + Dense), and CLD (Convolution + LSTM + Dense). Results show that the tuned L scheme achieved the best performance with accuracy increasing from 0.6304 to 0.6739 and F1-score from 0.6792 to 0.7097. These findings highlight the effectiveness of metaheuristic optimization in computational toxicology modeling.
Leveraging Assistive Image Captioning on VizWiz through Multi-Caption Supervision Fauzan Firdaus; Ema Rachmawati; Angel Metanosa Afinda; Edward Ferdian; Muhammad Maulana Yusuf
Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) Vol 10 No 5 (2026): October 2026
Publisher : Ikatan Ahli Informatika Indonesia (IAII)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29207/resti.v10i5.7472

Abstract

Most image captioning datasets provide several human captions for the same image, yet model training is frequently carried out in a way that effectively treats each image as having only one caption. Our study revisits this mismatch by examining the impact of training with multiple captions per image on caption generation performance. For this purpose, we utilized the VizWiz-Captions dataset, which reflects a real assistive scenario with images taken by blind or visually impaired users with five reference captions per image. We evaluated two conventional CNN-based encoder–decoder models, namely CNN–LSTM and CNN–GRU. In both models, the visual encoder was fixed using InceptionV3 image features, while the recurrent decoder was varied between LSTM and GRU. We performed experiments by using training data by varying the number of captions used per image from one to five (k = 1–5), while keeping the overall pipeline unchanged. The models were evaluated using standard multi-reference metrics (BLEU-1 to BLEU-4, CIDEr, and SPICE). Across both baselines, performance generally improves when more captions are included in training, although the gains are not perfectly monotonic for every metric. The strongest results are observed at k = 5. For CNN–LSTM, the best setting achieved BLEU-4 = 0.175, CIDEr = 0.206, and SPICE = 0.120. On the other hand, CNN–GRU achieved BLEU-4 = 0.192, CIDEr = 0.226, and SPICE = 0.119. These findings suggest that, even without architectural changes, simply leveraging the available caption diversity during training can provide a practical boost for VizWiz-style captioning tasks.
RANCANG BANGUN SISTEM EMERGENCY ENERGI LISTRIK UNTUK KONSUMEN RUMAH TANGGA GOLONGAN R1 Fauzan Firdaus; M. Hariansyah; Suratun .
JUTEKS Vol 5 No 1 (2018)
Publisher : Program Studi Teknik Elektro Universitas Ibn Khaldun Bogor

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (729.131 KB) | DOI: 10.32832/juteks.v5i1.2345

Abstract

RANCANG BANGUN SISTEM EMERGENCY ENERGI LISTRIK UNTUK KONSUMEN RUMAH TANGGA GOLONGAN R1. Telah dilakukan Rancang Bangun System Emergency Energi Listrik Untuk Konsumen Rumah Tangga  Golongan R1. Llistrik merupakan kebutuhan yang sangat penting terutama dalam bidang pengembangan teknologi informasi, sarana pendidikan dan rumah tangga. Penerapan sumber energi listrik alternatif sebagai energi cadangandengan pemanfaatan sumber energi DC (direct current) dari baterai (accu) yang di konversikan melalui tahapan rangkaian sehingga dapat dimanfaatkan sebagai sumber energi cadangan AC (Alternating current). Alat ini bekerja ketika kondisi PLN padam, sedangkan ketika sumber listrik dari PLN menyala maka energi listrik pln yang digunakan sebagai mengisi daya pada baterai (charge) melalui Transformator step down menjadi 24Vac kemudian disearahkanmelalui rangkaian penyearah yang berupa bridge rectifier bertujuan untuk menyearahkan tegangan AC menjadi tegangan DC 24Vdc untuk mencharge accu 24Vdc. Kemudian terdapat rangkaian inverter yang bertujuan untukmengkonversikan energi listrik 24Vdc menjadi listrik 220 AC melalui proses switching atau pensaklaran pada inverter yang kemudian di salurkan pada transformator step up CT (center tap) 600VA. Sebelum daya listrik di alirkan ke beban, maka disalurkan terlebih dahulu pada magnetic contactor sebagai alat pemindah energi listrik PLN secara otomatis
Benchmarking YOLOv12 Variants for Indonesian Traditional Cuisine Detection Firdaus, Fauzan; Ningsih, Lidya; Marsuki, Aminah Indahsari; Afinda, Angel Metanosa
Journal of Applied Informatics and Computing Vol. 10 No. 3 (2026): June 2026
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v10i3.12625

Abstract

YOLOv12 is one of the latest YOLO versions currently. Several studies have proven that YOLOv12 has better performance compared to previous versions. YOLOv12 itself has five model variants based on its architectural complexity, namely nano, small, medium, large and extra larges. This study tests the performance of YOLOv12 model variants (n, s, m, l, x) for traditional Indonesian culinary detection using a domain-specific object detection dataset. The dataset contains 718 images with 720 bounding-box instances annotated across 20 culinary classes, divided into 418/150/150 images for training/validation/testing. Data processing was performed in Roboflow with automatic orientation and stretching resizing to 640×640, while the training split was enriched using augmentation (horizontal and vertical flips) to increase sample diversity. All YOLOv12 variants were trained with the same configuration and environment, for 50 epochs using the Ultralytics framework with default hyperparameters on an NVIDIA A100-SXM4 80GB GPU. On the validation set, all variants achieved high detection accuracy (mAP@0.5 = 0.985–0.991), while differences emerged under a more stringent localization criterion (mAP@0.5:0.95). The best overall localization performance was achieved by YOLOv12-L (mAP@0.5:0.95 = 0.874), while YOLOv12-N provided the fastest inference (0.8 ms/image) with competitive accuracy (mAP@0.5:0.95 = 0.822). These findings provide preliminary guidance for selecting YOLOv12 variants based on the trade-off between accuracy and speed.