Article Per Year (5 Year)
p-Index From 2019 - 2024
0.408
P-Index
This Author published in this journals
All Journal Eksponensial
Rito Goejantoro
aboratorium Statistika Komputasi FMIPA Universitas Mulawarman
Computer Science & IT Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Mathematics Other

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

Found 2 Documents
Search

 1 
Perbandingan Klasifikasi Metode Naive Bayes dan Metode Decision Tree Algoritma (J48) pada Pasien Penderita Penyakit Stroke di RSUD Abdul Wahab Sjahranie Samarinda Irene Lishania; Rito Goejantoro; Yuki Novia Nasution
EKSPONENSIAL Vol 10 No 2 (2019)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (554.727 KB)

Abstract

Classification is a technique to form a model of the data that has not been classified, then the model can be used to classify new data. Naive Bayes is a classification using probability method based on the Bayes theorem with a strong assumption of independence. The decision tree algorithm (J48) is an implementation of the algorithm (C4.5) that produces decision trees. In this research, will be compared the results of classification accuracy with the naive Bayes method and the decision tree algorithm (J48) in stroke patients. That is, a person who has stroke will be classified by using the data of patients in Abdul Wahab Sjahranie Samarinda Hospital with 7 factors, namely age, gender, blood pressure, diabetes mellitus, dyslipidemia, uric acid levels and heart disease. The results showed that the decision tree algorithm (J48) method has the higher level of accuracy than the method naive Bayes for stroke classification.
Perbandingan Pengelompokan K-Means dan K-Medoids Pada Data Potensi Kebakaran Hutan/Lahan Berdasarkan Persebaran Titik Panas Athifaturrofifah Athifaturrofifah; Rito Goejantoro; Desi Yuniarti
EKSPONENSIAL Vol 10 No 2 (2019)
Publisher : Program Studi Statistika FMIPA Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (786.777 KB)

Abstract

The cases of forest/land fires in Indonesia seem endless, almost every year in the dry season similar problems always occur. Some areas in Indonesia often occur in forest fires and result in losses of up to trillions of rupiah. Various ways have been made to help the government in minimizing the potential for forest or land fires, one of them is by monitoring hot spots. In this study using data hot spots with parameters of latitude, longitude, brightness, fire radiation power and confidence by using the method of grouping K-Means and K-Medoids. The difference between these two methods is that the K-means method uses the mean as the center of the cluster, while K-Medoids uses representative objects (medoids) as the center of the cluster. This study aims to compare the results of the grouping of K-Means method with K-Medoids by using 42 data. The results of this study indicate that the K-Means method produces Silhouette Coefficient scores greater than K-Medoids. So that, K-Means can provide more accurate grouping results with a greater Silhouette Coefficient value.
Co-Authors Athifaturrofifah Athifaturrofifah Desi Yuniarti Irene Lishania Yuki Novia Nasution