Claim Missing Document
Check
Articles

Found 4 Documents
Search

A Systematic Literature Review on Machine Learning Techniques for Skin Disease Classification Nadiyah, Fadilah Karamun Nisaa; Alifah, Nayla Nur; Nurdiati, Sri; Khatizah, Elis; Najib, Mohamad Khoirun
Techno.Com Vol. 24 No. 2 (2025): Mei 2025
Publisher : LPPM Universitas Dian Nuswantoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62411/tc.v24i2.12696

Abstract

Skin diseases are health problems that require accurate diagnosis to evaluation and ultimately leading to treatment decisions. One of the crucial roles in the diagnostic process is medical imaging. Machine learning technology can assist in classifying skin diseases using image data and achieving high levels of accuracy in diagnosis. The purpose of this research is to review machine learning algorithms that can be utilized to develop image-based skin disease classification systems. The methodology employed is a Systematic Literature Review (SLR), which can be used to provide a comprehensive review of the application of machine learning in the classification of skin diseases. The literature search strategy was based on the Boolean technique, applied to the Scopus database. The selected articles were screened using predefined inclusion and exclusion criteria. The results indicate that the most used machine learning algorithm with achieved the highest classification accuracy is the Convolutional Neural Network (CNN). Keywords - Skin Disease, Machine Learning, Classification, CNN.
El nino index prediction model using quantile mapping approach on sea surface temperature data Nurdiati, Sri; Khatizah, Elis; Najib, Mohamad Khoirun; Fatmawati, Linda Leni
Desimal: Jurnal Matematika Vol. 4 No. 1 (2021): Desimal: Jurnal Matematika
Publisher : Universitas Islam Negeri Raden Intan Lampung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24042/djm.v4i1.7595

Abstract

El Nino is a global climate phenomenon caused by the warming of sea surface temperatures in the eastern Pacific Ocean. El Nino has a powerful effect on the intensity of rainfall in several areas in Indonesia. El Nino impacts can be minimized by predicting the El Nino index from the sea surface temperature in the Nino 3.4 area. Therefore, many researchers have tried to predict sea surface temperature, and many prediction data are available, one of which is ECMWF. But, in reality, the ECMWF data still contains systematic errors or bias towards the observations. Consequently, El Nino predictions using ECMWF data are less accurate. For that reason, this study aims to correct the ECMWF data in the Nino 3.4 area using statistical bias correction with a quantile mapping approach. This method uses ECMWF data from 1983-2012 as training data and 2013-2018 as testing data. For this case, the results showed that 60% of El Nino's predictions on the testing data had improved the mean value. Also, all of El Nino's predictions on the testing data have improved the standard deviation value. Moreover, data testing's expected error can be corrected for all months in the 1st to 4th lead times. But, in the 5th to 7th lead times, only November-June can be corrected.
A Lightweight CNN for Multi-Class Classification of Handwritten Digits and Mathematical Symbols Abisha, Nicholas; Redytadevi, Tita Putri; Nurdiati, Sri; Khatizah, Elis; Najib, Mohamad Khoirun
Techno.Com Vol. 24 No. 3 (2025): Agustus 2025
Publisher : LPPM Universitas Dian Nuswantoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62411/tc.v24i3.13138

Abstract

Recognizing handwritten digits and mathematical symbols remains a nontrivial challenge due to handwriting variability and visual similarity among classes. While deep learning, particularly Convolutional Neural Networks (CNNs), has significantly advanced handwriting recognition, many existing solutions rely on deep, resource-intensive architectures. This study aims to develop a lightweight and efficient CNN model for multi-class classification of handwritten digits and mathematical symbols, with an emphasis on deployability in resource-constrained environments such as educational platforms and embedded systems. The proposed model, implemented in Julia using the Flux.jl library, features a compact architecture with only two convolutional layers and approximately 55,000 trainable parameters significantly smaller than typical deep CNNs. Trained and evaluated on a publicly available dataset of over 10,000 grayscale 28×28-pixel images across 19 symbol classes, the model achieves a test accuracy of 91.8% while maintaining low computational demands. This work contributes to the development of practical handwritten mathematical expression recognition systems and demonstrates the feasibility of using Julia for developing lightweight deep learning applications.   Keywords - Digits, Mathematical Symbol, Classification, CNN
Performance Comparison of VGG16, MobileNetV2, and InceptionV3 Convolutional Neural Networks in Classifying Facial Dermatological Conditions Nadiyah, Fadilah Karamun Nisaa; Alifah, Nayla Nur; Nurdiati, Sri; Khatizah, Elis; Najib, Mohamad Khoirun
Jambura Journal of Mathematics Vol 7, No 2: August 2025
Publisher : Department of Mathematics, Universitas Negeri Gorontalo

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37905/jjom.v7i2.33082

Abstract

This study investigates the performance of three convolutional neural network (CNN) architectures (VGG16, MobileNetV2 and InceptionV3) in classifying two common facial dermatological conditions: acne and dark spots. A dataset of 235 facial skin images was augmented, then used to train and evaluate each model using standard classification metrics such as accuracy, precision, recall, and F1-score. The results demonstrate that MobileNetV2 achieved the highest classification accuracy of 93.13% while maintaining a relatively low computational cost. The model exhibited perfect precision (1.00) for the acne class and a high recall of 0.99 for the dark spots class, indicating its strong capability in accurately and sensitively identifying both lesion types. All three models demonstrated acceptable classification performance for both acne and dark spots classes, as evidenced by their precision, recall, and F1-scores exceeding 70%. This indicates that each model was capable of capturing relevant discriminative features of both lesion types.