Learning internal representations by error propagation
Neurocomputing: foundations of research
C4.5: programs for machine learning
C4.5: programs for machine learning
Machine learning, neural and statistical classification
Machine learning, neural and statistical classification
Neural Networks in Computer Intelligence
Neural Networks in Computer Intelligence
Logistic Regression Using the SAS System: Theory and Application
Logistic Regression Using the SAS System: Theory and Application
Artificial Neural Networks: An Introduction to ANN Theory and Practice
Predicting breast cancer survivability: a comparison of three data mining methods
Artificial Intelligence in Medicine
Data mining a diabetic data warehouse
Artificial Intelligence in Medicine
Hi-index | 0.00 |
From last decade, we are confronted with the rapid growth of diabetic patients who have become one of the most important burdens of public health. Accompanied with different complications, diabetes has considerable influences on the quality of individual living and the use of medical resources in the world in the 21st century. The purpose of this study is twofold. First, from the comparison standpoint logistic regression and neural networks were adopted to pursue the underlying characteristics of the glycemic control of the achieving target, or poor control level, so as to provide guidelines for physicians and diabetes educators. Second, for the cross validity purpose, 512 middle-aged patients, enrolled in Diabetes Healthcare Quality Improvement Program, were divided into training data and holdout data in a teaching hospital in Taiwan. Armed with the comparison, the finding revealed that neural networks is more accuracy than logistic regression. The important factors influence glycemic control are Years of diabetes onset, Education status, Body mass index, Months of enrolled in Diabetes Healthcare Quality Improvement Program, and Patient-Physician relationship.