پیش بینی وضعیت بیماران مبتلا به هپاتیت با استفاده از تکنیک‌های داده‌کاوی در استان همدان

نوع: Type: thesis

مقطع: Segment: masters

عنوان: Title: پیش بینی وضعیت بیماران مبتلا به هپاتیت با استفاده از تکنیک‌های داده‌کاوی در استان همدان

ارائه دهنده: Provider: hamid reza moghadam kiya

اساتید راهنما: Supervisors: hamid reza dezfolian (Ph.D)

اساتید مشاور: Advisory Professors:

اساتید ممتحن یا داور: Examining professors or referees: دکتر پروانه سموئی – دکتر فرید عزیزی جلیلیان

زمان و تاریخ ارائه: Time and date of presentation: 10/08/99 – ساعت 11 لغایت 12

مکان ارائه: Place of presentation: به صورت مجازی

چکیده: Abstract: Today, the advent of integrated information systems and the growth of information technology have led to significant growth in medicine. Databases in the field of health contain a large amount of clinical data, and the use of data mining methods in this branch of science has led physicians to be of great help in all topics, especially the diagnosis and prediction of diseases. In this study, in order to predict the status of patients with hepatitis in Hamadan province, data related to 1217 patients with hepatitis B and C during the period of April 2014 to February 2017 in Hamadan province have been analyzed. In order to predict the clinical status of patients, disease outcome and diagnosis of hepatitis, by RapidMiner software.First superior features were selected with expert opinion and then classification Algorithms such as, Naïve Bayes, Neoral Network, Decision Trees and Support Vector Machine were used with Metacast and Adaboost operators for improving the results. So According to the compare of results, the C4.5 Decision Tree Algorithm with Metacast and Adaboost operators were used to predict the target properties. Finally the C4.5 Decision Tree models were studied and analyzed as much as possible due to some information deficiencies. In the following for clustering the patients, first superior features were selected with expert opinion too, and then the results of Aggregation Hierarchical Clustering, DBSCAN Density-Based Clustering and K-Mean Clustering Algorithms have been reviewed and analyzed. From the obtained results, it seems that considering that hepatitis is a behavioral disease and is a function of cultural, economic, social and belief variables, patients' clinical condition, disease outcome in individuals and diagnosis of hepatitis varies according to the level of culture and health education in urban, rural and habitat are different. Keywords: Hepatitis, Data Mining, Decision Tree C4.5, K-Means, Adabost, Metacast