Article Per Year (5 Year)
p-Index From 2021 - 2026
0.408
P-Index
This Author published in this journals
All Journal CSRID
Cucut Susanto
a:1:{s:5:"en_US";s:25:"Universitas Dipa Makassar";}
Computer Science & IT

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

Found 2 Documents
Search
Journal : csrid

 1 
Komparasi Algoritma Decision Tree, Naive Bayes dan KNN dalam Klasifikasi Kanker Payudara Muhammad Abdul Jabbar; Erfan Hasmin; Sunardi; Cucut Susanto; Wilem Musu
Computer Science Research and Its Development Journal Vol. 14 No. 3: October 2022
Publisher : LPPM Universitas Potensi Utama

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22303/csrid.14.3.2022.258-270

Abstract

Breast cancer is a type of cancer that is commonly formed in breast cells and the cancer cells grow out of control. Cancer can occur in all genders. In 2020, the Global Cancer Observatory recorded a death rate of 684,996,000 and new cases of 2,261,419[1]. From the mortality rate, both men and women should be aware of their health by taking actions such as early detection and avoiding the risk of causing cancer. The source of data in this study came from the UCI Machine Learning Repository. This study aims to compare three data mining algorithms for classifying breast cancer. In this study, the algorithms used in making comparisons are the Decision Tree Algorithm, Naive Bayes, and KNN using 2 cross-validation methods, Hold-Out and K-Fold. The results of the test showed that the KNN algorithm always produced excellent accuracy performance compared to the Naive Bayes and Decision Tree algorithms, namely 98% in the Hold-Out method and 96% in the K-Fold method, while Naive Bayes is 95% on the Hold-Out method and 95% on the K-Fold method, Decision Tree is 94% on the Hold-Out method and 93% on the K-Fold method. Keywords—Breast Cancer, Decision Tree, Naive Bayes, KNN
Perbandingan Logistic Regression, Random Forest, dan Perceptron pada Klasifikasi Pasien Gagal Jantung Taufik Zulhaq Jasman; Erfan Hasmin; Sunardi; Cucut Susanto; Wilem Musu
Computer Science Research and Its Development Journal Vol. 14 No. 3: October 2022
Publisher : LPPM Universitas Potensi Utama

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22303/csrid.14.3.2022.271-286

Abstract

According to WHO (World Health Organization) data, heart disease accounts for one-third of all fatalities worldwide. Heart failure causes the death of approximately 17.9 million people worldwide and is more prevalent in Asia. With the use of technology, biostatistical analysis data may be processed using data mining techniques to uncover correlation patterns for each data from previous data that can then be used to forecast diseases based on these patterns. Algorithms with high accuracy can at least help medical experts prevent heart failure deaths. The ultimate goal of this research is to compare the algorithms and methods of Standard Scalers and robust scalers on datasets that have implemented the SMOTE method and without SMOTE to find algorithms and techniques with the highest accuracy and performance from the best algorithms using ROC curve values ​​and AUC values. In this study, researchers will use Logistic Regression, Random Forest, and Perceptron. Comparing the Standard Scaler and Robust Scaler methods as data standardization. And using the SMOTE technique and not using SMOTE to overcome unbalanced classes. As a result, the Random Forest algorithm with SMOTE technique and Standard Scaler is suitable for classifying and predicting heart failure.
Co-Authors Erfan Hasmin Muhammad Abdul Jabbar Sunardi Sunardi Taufik Zulhaq Jasman Wilem Musu