Article Per Year (5 Year)
p-Index From 2020 - 2025
0.408
P-Index
This Author published in this journals
All Journal International Journal of Electrical, Computer, and Biomedical Engineering (IJECBE)
Prima Dewi Purnamasari
Universitas Indonesia
Computer Science & IT Electrical & Electronics Engineering Materials Science & Nanotechnology Medicine & Pharmacology

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Development Of Detection Model For Skin Diseases In Pets Using Image Processing And Deep Learning Techniques Taufiqoh, Salma Dewi; Purnamasari, Prima Dewi
International Journal of Electrical, Computer, and Biomedical Engineering Vol. 3 No. 2 (2025)
Publisher : Universitas Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62146/ijecbe.v3i2.114

Abstract

Early detection of skin diseases in pets is essential but often hindered by the cost and complexity of clinical diagnosis. This study introduces a deep learning–based system for identifying three common pet skin diseases—Ringworm, Scabies, and Earmite—using images captured with mobile phone cameras. The system integrates classical image preprocessing techniques, including Contrast Limited Adaptive Histogram Equalization (CLAHE) and Hue-Saturation-Value (HSV) segmentation, with a custom convolutional neural network (CNN) designed for disease-specific classification tasks. Two separate models were developed: a multi-class CNN model for classifying Ringworm, Scabies, and Undetected conditions, which achieved a test accuracy of 83%, and a binary CNN model for classifying Earmite versus Undetected, which achieved 100% accuracy, precision, and recall on both test and unseen validation sets. Compared to transfer learning models such as ResNet-50 and VGG16, the proposed CNN models demonstrated superior performance under limited-data conditions (72 images total), emphasizing the advantage of domain-specific model design and preprocessing. These findings suggest that disease-adapted CNN architectures, combined with targeted preprocessing, can support accurate and accessible veterinary screening using mobile devices. Future work will focus on expanding the dataset and deploying the model in a real-time mobile diagnostic application.
Defying Data Scarcity: High-Performance Indonesian Short Answer Grading via Reasoning-Guided Language Model Fine-Tuning Faza, Muhammad Naufal; Purnamasari, Prima Dewi; Ratna, Anak Agung Putri
International Journal of Electrical, Computer, and Biomedical Engineering Vol. 3 No. 3 (2025)
Publisher : Universitas Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62146/ijecbe.v3i3.148

Abstract

Automated Short Answer Grading (ASAG) is crucial for scalable feedback, but applying it to low-resource languages like Indonesian is challenging. Modern Large Language Models (LLMs) severely overfit small, specialized educational datasets, limiting utility. This study compares nine traditional machine learning models against two fine-tuning strategies for Gemma-3-1b-it on an expanded Indonesian ASAG dataset (n=220): (a) standard fine-tuning predicting only scores, and (b) a proposed reasoning-guided approach where the model first generates a score rationale using knowledge distillation before predicting the score. The reasoning-guided model (Gemma-3-1b-ASAG-ID-Reasoning) achieved state-of-the-art performance (QWK 0.7791; Spearman’s 0.8276), significantly surpassing the best traditional model in this study (SVR, QWK 0.6952). This work advances foundational LSA-based approaches for this task by introducing a more robust methodology and evaluation framework. Crucially, standard fine-tuning (Gemma-3-1b-ASAG-ID) suffered catastrophic overfitting (QWK 0.7279), indicated by near-perfect training but poor test scores. While the reasoning-guided LLM showed superior accuracy, it required over 35 times more inference time. Results demonstrate that distilled reasoning acts as a powerful regularizer, compelling the LLM to learn underlying grading logic rather than memorizing pairs, establishing a viable method for high-performance ASAG in data-scarce environments despite computational trade-offs.
Co-Authors Faza, Muhammad Naufal Ratna, Anak Agung Putri Taufiqoh, Salma Dewi