Document Details

Document Type : Thesis 
Document Title :
Development of Data-Mining Models for Big Data Analytics Case Study: Improving Prediction Accuracy of Hospital Acquired Infection
تطوير نماذج التنقيب عن البيانات لتحليل البيانات الضخمة دراسة حالة: (تحسين دقة التنبؤ بالعدوى المكتسبة في المستشفيات)
 
Subject : Faculty of Computing and Information Technology 
Document Language : Arabic 
Abstract : This research focuses on developing data-mining models for big data analytics to improve prediction accuracy for Healthcare-Associated Infections (HAIs) as a case study. A new data-mining classification algorithm was created to overcome some of the issues plaguing this endeavor. The new algorithm was assessed against the best data mining techniques defined by this study and was found to minimize prediction times of HAIs by three to five days. The early prediction enables immediate intervention by clinical staff, which speeds up the recovery time and minimizes harm to the patient. Traditional algorithms are inapplicable to the target domain analytic process. Big Data raises the bar as a result of using additional features. It is characterized mainly by a tremendous amount of data in different forms. It also deals with rapid data flow rates that are generated from multiple sources, and to top it off, the quality of the data is questionable. The research surveys data-mining algorithms in healthcare to define strengths and weaknesses. Accordingly, it defines the most suitable techniques and compares them to similar approaches. The proposed classification algorithm was evaluated with real patients’ data from King Abdullah Medical City, Makkah. It contains more than 28,000 cases that consist of laboratory results, radiology findings, surgical histories, and physicians’ notes. The outcomes prove that the developed classification algorithm is superior to the others. 
Supervisor : Dr. Amin Noman 
Thesis Type : Doctorate Thesis 
Publishing Year : 1440 AH
2019 AD 
Added Date : Wednesday, August 21, 2019 

Researchers

Researcher Name (Arabic)Researcher Name (English)Researcher TypeDr GradeEmail
عمر سامي باعيسىBaeissa, Omar SamiResearcherDoctorate 

Files

File NameTypeDescription
 44905.pdf pdf 

Back To Researches Page