Powered by OpenAIRE graph
Found an issue? Give us feedback
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/ ILTEK Jurnal Teknolo...arrow_drop_down
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
ILTEK Jurnal Teknologi
Article . 2025 . Peer-reviewed
License: CC BY NC
Data sources: Crossref
addClaim

Klasifikasi Citra Tulisan Tangan Aksara Sunda Berbasis Inception-ResNetV2 dengan Transfer Learning

Authors: Clara Diva Paramitha; Achmad Junaidi; Muhammad Muharrom Al Haromainy;

Klasifikasi Citra Tulisan Tangan Aksara Sunda Berbasis Inception-ResNetV2 dengan Transfer Learning

Abstract

Aksara Sunda adalah aksara yang semakin jarang digunakan sehingga masyarakat awam sering tidak familiar dengan bentuknya, terutama saat membaca tulisan tangan yang selalu memiliki variasi tergantung penulisnya sehingga ada keterbatasan dalam mengenali bentuknya. Penelitian ini berfungsi untuk menguji performa model deep learning untuk tugas klasifikasi 23 kelas Aksara Sunda serta menguji kombinasi model terhadap berbagai parameter agar dapat memberikan hasil optimal. Penelitian ini menggunakan Inception-ResNetV2 yang dikombinasikan dengan metode fine-tuning transfer learning untuk diuji terhadap tiga optimizer dan learning rate sebanyak 20 epoch. Data pada penelitian ini gabungan dari data GitHub dan data pengumpulan mandiri. Pengujian ini menggunakan rasio 80:20 untuk data latih dan data uji. Optimizer yang diuji adalah SGD, Adam, dan RMSProp. Hasil pengujian menunjukkan bahwa tiap-tiap optimizer mampu memberikan hasil teroptimalnya pada parameter tertentu. Melihat skor performa, RMSProp 0.0001 berhasil mencapai nilai akurasi data uji tertinggi pada 99.15%, diikuti oleh SGD 0.01 dengan akurasi data uji 98.66%, lalu disusul Adam 0.0001 dengan akurasi data uji 96.61%. Akan tetapi, melihat grafik kurva, optimizer SGD lebih stabil dibandingkan RMSProp—yang mengalami guncangan di awal—ataupun Adam—yang mengalami gejala overfitting ringan. Hasil kontradiktif ini dapat menjadi pembelajaran untuk penelitian selanjutnya. 

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
Powered by OpenAIRE graph
Found an issue? Give us feedback
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
0
Average
Average
Average
gold