Article Per Year (5 Year)
p-Index From 2021 - 2026
0.61
P-Index
This Author published in this journals
All Journal JITK (Jurnal Ilmu Pengetahuan dan Komputer) Jurnal Teknologi Pendidikan : Jurnal Penelitian dan Pengembangan Pembelajaran
Muhtyas Yugi
Unknown Affiliation
Computer Science & IT Education Languange, Linguistic, Communication & Media Other

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search

 1 
IMPROVING HANDWRITTEN DIGIT RECOGNITION USING CYCLEGAN-AUGMENTED DATA WITH CNN–BILSTM HYBRID MODEL Muhtyas Yugi; Utomo, Fandy Setyo; Barkah, Azhari Shouni
JITK (Jurnal Ilmu Pengetahuan dan Teknologi Komputer) Vol. 11 No. 2 (2025): JITK Issue November 2025
Publisher : LPPM Nusa Mandiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33480/jitk.v11i2.6982

Abstract

Handwritten digit recognition presents persistent challenges in computer vision due to the high variability in human handwriting styles, which necessitates robust generalization in classification models. This study proposes an advanced data augmentation strategy using Cycle-Consistent Generative Adversarial Networks (CycleGAN) to improve recognition accuracy on the MNIST dataset. Two architectures are evaluated: a standard Convolutional Neural Network (CNN) and a hybrid model combining CNN for spatial feature extraction and Bidirectional Long Short-Term Memory (BiLSTM) for sequential pattern modeling. The CycleGAN-based augmentation generates realistic synthetic images that enrich the training data distribution. Experimental results demonstrate that both models benefit from the augmentation, with the CNN-BiLSTM model achieving the highest accuracy of 99.22%, outperforming the CNN model’s 99.01%. The study’s novelty lies in the integration of CycleGAN-generated data with a CNN–BiLSTM architecture, which has been rarely explored in previous works. These findings contribute to the development of more generalized and accurate deep learning models for handwritten digit classification and similar pattern recognition tasks.
IMPROVING HANDWRITTEN DIGIT RECOGNITION USING CYCLEGAN-AUGMENTED DATA WITH CNN–BILSTM HYBRID MODEL Utomo, Fandy Setyo; Barkah, Azhari Shouni; Muhtyas Yugi
JITK (Jurnal Ilmu Pengetahuan dan Teknologi Komputer) Vol. 11 No. 2 (2025): JITK Issue November 2025
Publisher : LPPM Nusa Mandiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33480/jitk.v11i2.6982

Abstract

Handwritten digit recognition presents persistent challenges in computer vision due to the high variability in human handwriting styles, which necessitates robust generalization in classification models. This study proposes an advanced data augmentation strategy using Cycle-Consistent Generative Adversarial Networks (CycleGAN) to improve recognition accuracy on the MNIST dataset. Two architectures are evaluated: a standard Convolutional Neural Network (CNN) and a hybrid model combining CNN for spatial feature extraction and Bidirectional Long Short-Term Memory (BiLSTM) for sequential pattern modeling. The CycleGAN-based augmentation generates realistic synthetic images that enrich the training data distribution. Experimental results demonstrate that both models benefit from the augmentation, with the CNN-BiLSTM model achieving the highest accuracy of 99.22%, outperforming the CNN model’s 99.01%. The study’s novelty lies in the integration of CycleGAN-generated data with a CNN–BiLSTM architecture, which has been rarely explored in previous works. These findings contribute to the development of more generalized and accurate deep learning models for handwritten digit classification and similar pattern recognition tasks.
Enhancing Computational Thinking of Islamic Education Students through CT-Based Prompt Engineering: A Quasi-Experimental Study on AI Multimodal Media Design Imam Kharits Najibulloh; Ahmad Latif; Muhtyas Yugi; Beny Riswanto
Jurnal Teknologi Pendidikan : Jurnal Penelitian dan Pengembangan Pembelajaran Vol. 11 No. 2 (2026): April
Publisher : UNDIKMA

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33394/jtp.v11i2.20197

Abstract

This study evaluates the effectiveness of Prompt Engineering strategies based on Computational Thinking (CT) in enhancing the ability of Islamic Religious Education (PAI) students to design multimodal learning media (images, videos, and games) powered by Artificial Intelligence (AI). Using a quasi-experimental design involving 35 students, the research integrates the four pillars of CT decomposition, pattern recognition, abstraction, and algorithm design as logical foundations for composing AI instructions. The results show a significant increase in students’ CT scores, from an average of 56.80 to 85.10, with an N-Gain Score of 0.65 (effective category). Students successfully produced educational image media (92%), interactive game logic (84%), and animated videos (78%) with high theological accuracy. The abstraction pillar was found to be the most crucial in minimizing AI “hallucinations” in sensitive religious content, while algorithm design enabled the creation of systematic game flows. This strategy successfully transformed students’ roles from mere users to logical and critical instruction designers (Prompt Engineers). The study recommends integrating CT-Prompting into the PAI curriculum as a core competency to produce innovative and valid digital content in the era of artificial intelligence.
Co-Authors Ahmad Latif Azhari Shouni Barkah Beny Riswanto Fandy Setyo Utomo Imam kharits najibulloh