Article Per Year (5 Year)
p-Index From 2021 - 2026
0.659
P-Index
This Author published in this journals
All Journal Jurnal Manajemen Informatika Indonesian Journal of Electrical Engineering and Computer Science
Prasetyo Agung, Ignatius Wiseto
Unknown Affiliation
Computer Science & IT

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

Found 2 Documents
Search

 1 
Breast cancer identification using machine learning and hyperparameter optimization Arifin, Toni; Prasetyo Agung, Ignatius Wiseto; Junianto, Erfian; Rachman, Rizal; Wibowo, Ilham Rachmat; Agustin, Dari Dianata
Indonesian Journal of Electrical Engineering and Computer Science Vol 36, No 3: December 2024
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijeecs.v36.i3.pp1620-1630

Abstract

Breast cancer identification can be analyzed through genomic analysis using gene expression data, one type of which is mRNA. This involves analyzing gene expression patterns of breast tissue samples to distinguish breast cancer from healthy tissue or to differentiate subtypes of different breast cancers. This research developed the right computational model for breast cancer classification using machine learning and hyperparameter optimization algorithms. The primary objective of this research is to utilize various machine learning algorithms to classify breast cancer based on gene expression and enhance the models developed in previous studies. This paper provides an extensive literature review of prior breast cancer classification research and offers new theoretical perspectives. This research used a problem-solving approach with conventional machine learning techniques, most notably the decision tree. It also evaluates other machine learning algorithms for comparison, including k-nearest neighbor, naïve bayes, random forest, extra tree classifier, and support vector machine. The evaluation process used classification reports that provide insight into the precision, recall, F1-score, and accuracy of each machine learning model. The evaluation results show that the performance of the decision tree algorithm model is superior and impressive, achieving 99.73% accuracy and a score of 1 for precision, recall, and F1-score.
Klasifikasi Jenis Jerawat Secara Otomatis Dengan Convolutional Neural Network Menggunakan Arsitektur Resnet-50 Anshori, Muhammad Iqbal; Zafar Sidiq, Muhammad Ali; Yaqin, Rifki Ainul; Prasetyo Agung, Ignatius Wiseto
Jurnal Manajemen Informatika JAMIKA Vol 15 No 1 (2025): Jurnal Manajemen Informatika (JAMIKA)
Publisher : Program Studi Manajemen Informatika, Fakultas Teknik dan Ilmu Komputer, Universitas Komputer Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34010/jamika.v15i1.13712

Abstract

Acne is a common skin problem that requires different treatments based on its type, such as blackheads, conglobata, and papulopustular. This research develops an automatic acne type classification system using deep learning-based Residual Network (ResNet-50) architecture. With its 50 layers, ResNet-50 is effective in image classification. The objective of of this research is to classify the type of acne from skin images on the face, so that it can help diagnosis and treatment. face, so that it can help diagnosis and treatment. The method used in this research includes several main stages, namely the collection of the dataset, model training using CNN with ResNet-50 architecture, model testing, and performance evaluation. model, and performance evaluation. The dataset was obtained from Roboflow, consisting of three classes: acne-comedonica, acne-conglobata, and acne-papulopustulosa. The process involves image preprocessing, data augmentation, and model parameter adjustment, including Adam's dropout and optimizer techniques. The model can achieve 98.35% accuracy with loss of 0.0489 and the highest validation accuracy of 92.86% with a validation loss of 0.1976. In addition, confusion matrix analysis shows an accuracy result of 93%, which indicates the performance of the model in distinguishing between acne classes effectively. These results show that the model is effective in classifying the types of acne and can have a significant impact in assisting a more accurate and faster diagnosis. more accurate and quicker diagnosis.
Co-Authors Agustin, Dari Dianata Anshori, Muhammad Iqbal Erfian Junianto Pratama, Devina Gracia Purwadhi Purwadhi Rizal Rachman, Rizal Toni Arifin Wibowo, Ilham Rachmat Yaqin, Rifki Ainul Zafar Sidiq, Muhammad Ali