以貝氏分類器預測腎臟病初期症狀之研究

• Main Authors: Hui-Chin Lee, 李惠卿
• Other Authors: none
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/34127482443064469263

碩士 === 國立高雄第一科技大學 === 資訊管理研究所 === 104 === In recent years, due to aging population, rich material life, environmental changes and the shifting of dietary habits, together with busy and stressful life, most of the Taiwanese people ignore the warning signs of poor health status for themselves. Besides, the growing food safety crisis leads to an increasing number of chronic kidney disease in our society. Based on the data of Department of Health and National Health Insurance Department, it shows that: the annual increase by incidence and prevalence of kidney dialysis is about 6 to 7 million people, which costs about 33 billion NT dollars in the year of 2014. As kidney is an organ of silence, the early chronic kidney disease has no obvious symptoms, which makes it hard to uncover. When a patient obviously feels uncomfortable, it has often resulted in kidney function decline. The risky factors lead to kidney disease coming from: aging, hypertension, diabetes, lipid abnormalities, metabolic syndrome, analgesics and herbal abuse, or a family history of kidney disease. If the chronic kidney disease can be detected and treated earlier, the problem of increasing number of patients with kidney disease would be hopefully mitigated. In this study, by sampling the early chronic kidney patient cases and general diagnostic data for medical treatment, we will respectively establish frameworks to predict the probability of a patient’s chronic kidney disease by analyzing the risky factors associated with the accompanied diseases as variables through Naïve Bayes and Bayesian classifier. We hope the research result can be regarded as a valuable reference to improve the quality of health care and medical act in Taiwan.