Melola Discovery

← Melola Discovery

Open access Jurnal SINTA 2

Exploring Pre-Trained Model and Language Model for Translating Image to Bahasa

IJCCS (Indonesian Journal of Computing and Cybernetics Systems) · Vol. 17 No. 4 (2023) · pp. 347–358

DOI
10.22146/ijccs.76389
ISSN
2460-7258, 1978-1520
Published
2023-10-31
Publisher
IndoCEISS in colaboration with Universitas Gadjah Mada, Indonesia.
Language
eng
License
creativecommons.org/licenses/by-sa/4.0

Abstract

In the last decade, there have been significant developments in Image Caption Generation research to translate images into English descriptions. This task has also been conducted to produce texts in non-English, including Bahasa. However, the references in this study are still limited, so exploration opportunities are open widely. This paper presents comparative research by examining several state-of-the-art Deep Learning algorithms to extract images and generate their descriptions in Bahasa. We extracted images using three pre-trained models, namely InceptionV3, Xception, and EfficientNetV2S. In the language model, we examined four architectures: LSTM, GRU, Bidirectional LSTM, and Bidirectional GRU. The database used was Flickr8k which was translated into Bahasa. Model evaluation was conducted using BLEU and Meteor. The performance results based on the pre-trained model showed that EfficientNetV3S significantly gave the highest score among other models. On the other hand, in the language model, there was only a slight difference in model performance. However, in general, the Bidirectional GRU scored higher. We also found that step size in training affected overfitting. Larger step sizes tended to provide better generalizations. The best model was generated using EfficientNetV3S and Bidirectional GRU with step size=4096, which resulted in an average score of BLEU-1=0,5828 and Meteor=0,4520.

Keywords

Collaboration network of the authors

6 authors · 7 co-authorship links · 1 across different organisations

Hover or tap a name to highlight its co-authors. Double-click to open the author's page.

  • Universitas Amikom Purwokerto
  • Universitas Gadjah Mada
  • Different organisations
  • Same organisation
  • Affiliation unknown
  • Thicker line = more joint articles · bigger circle = more articles
Show as table
Co-authorOrganisationJoint articles
Irfan Rifai AzzizUniversitas Amikom Purwokerto2
Agus HarjokoUniversitas Gadjah Mada1
Uswatun HasanahUniversitas Amikom Purwokerto1

Affiliations come from OpenAlex; journal OAI-PMH metadata carries none. Authors without an OpenAlex ID are matched by name, which can merge or split people.

Metadata source: OAI-PMH (metadata jurnal) https://journal.ugm.ac.id/ijccs/oai, first harvested 26 Okt 2026, last seen 26 Okt 2026.
Original metadata (oai_dc) 
<oai_dc:dc xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/  http://www.openarchives.org/OAI/2.0/oai_dc.xsd">
	<dc:title xml:lang="en-US">Exploring Pre-Trained Model and Language Model for Translating Image to Bahasa</dc:title>
	<dc:creator>Nurhopipah, Ade</dc:creator>
	<dc:creator>Suhaman, Jali</dc:creator>
	<dc:creator>Widianto, Anan</dc:creator>
	<dc:subject xml:lang="en-US">Computer Science</dc:subject>
	<dc:subject xml:lang="en-US">BLEU; Image Caption Generation; Meteor; Language Model; Pre-trained Model</dc:subject>
	<dc:description xml:lang="en-US">In the last decade, there have been significant developments in Image Caption Generation research to translate images into English descriptions. This task has also been conducted to produce texts in non-English, including Bahasa. However, the references in this study are still limited, so exploration opportunities are open widely. This paper presents comparative research by examining several state-of-the-art Deep Learning algorithms to extract images and generate their descriptions in Bahasa. We extracted images using three pre-trained models, namely InceptionV3, Xception, and EfficientNetV2S. In the language model, we examined four architectures: LSTM, GRU, Bidirectional LSTM, and Bidirectional GRU. The database used was Flickr8k which was translated into Bahasa. Model evaluation was conducted using BLEU and Meteor. The performance results based on the pre-trained model showed that EfficientNetV3S significantly gave the highest score among other models. On the other hand, in the language model, there was only a slight difference in model performance. However, in general, the Bidirectional GRU scored higher. We also found that step size in training affected overfitting. Larger step sizes tended to provide better generalizations. The best model was generated using EfficientNetV3S and Bidirectional GRU with step size=4096, which resulted in an average score of BLEU-1=0,5828 and Meteor=0,4520.</dc:description>
	<dc:publisher xml:lang="en-US">IndoCEISS in colaboration with Universitas Gadjah Mada, Indonesia.</dc:publisher>
	<dc:contributor xml:lang="en-US"></dc:contributor>
	<dc:date>2023-10-31</dc:date>
	<dc:type>info:eu-repo/semantics/article</dc:type>
	<dc:type>info:eu-repo/semantics/publishedVersion</dc:type>
	<dc:type xml:lang="en-US"></dc:type>
	<dc:format>application/pdf</dc:format>
	<dc:identifier>https://journal.ugm.ac.id/ijccs/article/view/76389</dc:identifier>
	<dc:identifier>10.22146/ijccs.76389</dc:identifier>
	<dc:source xml:lang="en-US">IJCCS (Indonesian Journal of Computing and Cybernetics Systems); Vol 17, No 4 (2023): October; 347-358</dc:source>
	<dc:source>2460-7258</dc:source>
	<dc:source>1978-1520</dc:source>
	<dc:language>eng</dc:language>
	<dc:relation>https://journal.ugm.ac.id/ijccs/article/view/76389/37707</dc:relation>
	<dc:rights xml:lang="en-US">Copyright (c) 2023 IJCCS (Indonesian Journal of Computing and Cybernetics Systems)</dc:rights>
	<dc:rights xml:lang="en-US">http://creativecommons.org/licenses/by-sa/4.0</dc:rights>
</oai_dc:dc>