Published in Journal of Smart Technology and Engineering, 2026
This study developed a CNN-based image classification system using MobileNetV2 and EfficientNetV2-S to address the health risks posed by common seafood allergens like shrimp, crab, shellfish, and fish.
Published in Jurnal Tekno Kompak, 2025
Penelitian ini membandingkan model Bi-LSTM dan HIVE-COTE 2.0 dalam memprediksi klasifikasi harga penutupan saham menggunakan fitur-fitur hasil feature engineering, dan menyimpulkan bahwa HIVE-COTE 2.0 lebih unggul dengan \(F_1\)-score 98,7% serta merekomendasikan persentase perubahan harga sebagai fitur paling berpengaruh.
Published in Eurasian Journal of Mathematical and Computer Applications, 2024
This research employs eleven machine learning algorithms to predict patient hospital readmissions from historical data while also identifying the most influential features for that prediction.
Published in The 1st International Conference on Emerging Issues in Technology, Engineering, and Science (ICE-TES), 2021
This paper proposes a method that merges multiple breast cancer histopathological image datasets into a unified three-class dataset, applies Vahadane preprocessing and progressive resizing training, and achieves an \(F_1\) score of around 99%, comparable to other state-of-the-art models.