Comparison of The Data-Mining Methods in Predicting The Risk Level of Diabetes
Journal Title: EMITTER International Journal of Engineering Technology - Year 2016, Vol 4, Issue 1
Abstract
Mellitus Diabetes is an illness that happened in consequence of the too high glucose level in blood because the body could not release or use insulin normally. The purpose of this research is to compare the two methods in The data-mining, those are a Regression Logistic method and a Bayesian method, to predict the risk level of diabetes by web-based application and nine attributes of patients data. The data which is used in this research are 1450 patients that are taken from RSD BALUNG JEMBER, by collecting data from 26 September 2014 until 30 April 2015. This research uses performance measuring from two methods by using discrimination score with ROC curve (Receiver Operating Characteristic). On the experiment result, it showed that two methods, Regression Logistic method and Bayesian method, have different performance excess score and are good at both. From the highest accuracy measurement and ROC using the same dataset, where the excess of Bayesian has the highest accuracy with 0,91 in the score while Regression Logistic method has the highest ROC score with 0.988, meanwhile on Bayesian, the ROC is 0.964. In this research, the plus of using Bayesian is not only can use categorical but also numerical.
Authors and Affiliations
Andri Permana W. , Tessy Badriyah, Achmad Basuki
Keywords
Related Articles
- EP ID EP269401
- DOI 10.24003/emitter.v4i1.119
- Views 78
- Downloads 0
Designing and Building of 3D Adventure Game “Tetuko: Childhood of Ghatotkacha” Using Kinect
Nowadays, the young people are not interested in the local culture as a “wayang” puppet. This condition threatens the extinction of some of the local culture that should be a mainstay of the industry entering an era of c...
An Implementation of Error Minimization Position Estimate in Wireless Inertial Measurement Unit using Modification ZUPT
Position estimation using WIMU (Wireless Inertial Measurement Unit) is one of emerging technology in the field of indoor positioning systems. WIMU can detect movement and does not depend on GPS signals. The position is t...
Optimizing Fuzzy Rule Base for Illumination Compensation in Face Recognition using Genetic Algorithms
Fuzzy rule optimization is a challenging step in the development of a fuzzy model. A simple two inputs fuzzy model may have thousands of combination of fuzzy rules when it deals with large number of input variations. Int...
Feature Extraction For Application of Heart Abnormalities Detection Through Iris Based on Mobile Devices
World Health Organization (WHO) data in 2012 showed 17.5 million people worldwide died from heart disease or 31% of 56.5 million deaths and examining it by current methods in hospitals is not cheap. Iridology is an alte...
Comparative Study of Modulation-Based Individual Inverter Techniques for Direct and Inverse by using Star-Connection Induction Motor in Extra Low Voltage Application
In this study, the IEEE 519 Standard as a basis benchmarking for voltage (THDV) and current (THDI) in draft performance. Comparative Study based onthree-techniques of 2-Level Converter (2LC) by using a Star-Connection In...