Article Per Year (5 Year)
p-Index From 2021 - 2026
1.642
P-Index
This Author published in this journals
All Journal Jurnal Universitas Paramadina Inquiry: Jurnal Ilmiah Psikologi G-Couns: Jurnal Bimbingan dan Konseling Jurnal Komputer Teknologi Informasi Sistem Komputer (JUKTISI) Triwikrama: Jurnal Ilmu Sosial JIPI (Jurnal IPA dan Pembelajaran IPA) Liberosis: Jurnal Psikologi dan Bimbingan Konseling Jurnal Kajian Hukum Dan Kebijakan Publik SITEKNIK: Sistem Informasi, Teknik dan Teknologi Terapan
Devi Wulandari
Unknown Affiliation
Arts Humanities Automotive Engineering Biochemistry, Genetics & Molecular Biology Chemistry Civil Engineering, Building, Construction & Architecture Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Education Engineering Environmental Science Law, Crime, Criminology & Criminal Justice Physics Social Sciences Veterinary Other

Published : 11 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 11 Documents
Search

 1   2 
Leakage-Aware Benchmarking of Lightweight Models for Robust Handwritten Hanacaraka Character Recognition Abdulloh, Ferian Fauzi; Sharazita Dyah Anggita; Ikmah; Ali Mustopa; Majid Rahardi; Devi Wulandari
Jurnal Komputer Teknologi Informasi Sistem Komputer (JUKTISI) Vol. 5 No. 1 (2026): Juni 2026
Publisher : LKP KARYA PRIMA KURSUS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62712/juktisi.v5i1.1039

Abstract

Handwritten Hanacaraka character recognition is important for preserving Javanese script, but reported results can be difficult to compare because datasets, preprocessing procedures, and train-test separation protocols vary across studies. This study presents a leakage-aware benchmark of lightweight models on a public handwritten Hanacaraka dataset containing 20 basic character classes. A data audit removed 17 unreadable image files and retained 1,562 valid images. Two experimental settings were evaluated: a perceptual-hash grouped split for leakage-aware testing and a random-stratified split as an optimistic upper-bound scenario. The leakage-aware benchmark compared HOG with SVM, HOG with Random Forest, MobileNetV2 head-only training, fine-tuned MobileNetV2, and a confusion-aware MobileNetV2 variant. Fine-tuned MobileNetV2 achieved the best leakage-aware result with 53.82% accuracy and 49.59% macro-F1, while robustness testing under image distortions produced 47.85% accuracy and 44.53% macro-F1. In the optimistic random-stratified experiment, an ensemble of EfficientNetB0 and MobileNetV2 with test-time augmentation reached 74.11% accuracy and 74.24% macro-F1. The results indicate that stricter evaluation substantially lowers performance and that visually similar classes remain difficult. Therefore, future Hanacaraka recognition work should report leakage control, robustness, and confusion analysis, not only clean-set accuracy.
Co-Authors Abdul L. Mawardi Abdulloh, Ferian Fauzi Ali Mustopa Ali Mustopa, Ali Anggita, Sharazita Dyah Dany Miftahul Ula Ekariana S Pandia Fakhriyyah Arsya Favian Afrheza Fattah Ikmah Maida Riyani Majid Rahardi Mawardi Mawardi Raissa Rizqi Marwa Rita Elviany Selly Ruwinda Sofwa Syifa Fauzia Tiara Putri Pertiwi Tri Mustika Sarjani Wiraguna, Sidi Ahyar Zahra Atha Assyifa