Document Details
|
Document Type |
: |
Thesis |
|
Document Title |
: |
ASSESSING THE EFFECTIVENESS OF BIG DATA ANALYTICS TECHNIQUES: CASE OF HEALTHCARE SECTOR تقييم فعالية التقنيات التحليلية للبيانات الكبيرة: قطاع الرعاية الصحية |
|
Subject |
: |
Faculty of Computing and Information Technology |
|
Document Language |
: |
Arabic |
|
Abstract |
: |
One of the rapidly emerging fields is the big data field. Big data is a buzz word that describes data characterized by the large volume, variety, velocity, and veracity. Big data analytics and tools are used to extract valuable information and predictions from such data in an efficient way. One of the promising fields for the application of big data technology is the healthcare domain. Unfortunately, big data has not yet been applied extensively in this field to its extreme potential. This stems from many reasons including the variety of data sources leading to heterogeneous data of various types (structured, semi-structured and unstructured). Processing and analyzing such data are an important challenge. One of the most critical problems in healthcare is predicting the likelihood of hospital readmission in case of chronic diseases such as diabetes to be able to allocate necessary resources such as beds, rooms, specialists, and medical staff, for an acceptable quality of service. Unfortunately, relatively few research studies in the literature attempted to tackle this problem; the majority of the research studies are concerned with predicting the likelihood of the diseases themselves. Numerous machine learning techniques are suitable for prediction. Nevertheless, there is also a shortage of inadequate comparative studies that specify the most suitable techniques for the prediction process. The goal of this thesis is to collect healthcare big data from different sources. This is followed by studying available big data analytics techniques suitable for processing such complex data and understanding each technique. Towards this goal, this thesis proposed a methodology for big data analytics. Also, it presents a comparative study among common techniques in the literature for predicting the likelihood of hospital readmission in the case of diabetic patients. The contribution of this study is assessing the possibility of improving and/or integrating machine learning techniques and tailoring them for improved information and predictions for enhanced healthcare. Those techniques are decision trees (DTs), logistic regression (LR), linear discriminant analysis (LDA), artificial neural networks (ANNs), support vector machine (SVM), Naïve Bayesian (NB), random forest (RF), AdaBoost and gradient boosting (GB). The comparative study is based on realistic data gathered from a number of hospitals in the United States. Many experiments were conducted on those techniques where the comparative studies revealed that ensemble-based learning techniques (boosting and bagging) for example GB, RF and AdaBoost showed the best performance, while the NB classifier, LR analysis, and LDA were the worst. |
|
Supervisor |
: |
Dr. Kamal Jumbi |
|
Thesis Type |
: |
Doctorate Thesis |
|
Publishing Year |
: |
1441 AH
2020 AD |
|
Added Date |
: |
Friday, June 12, 2020 |
|
Researchers
| سماح هزاع العجماني | Al-Ajmani, Samah Hazaa | Researcher | Doctorate | |
|
Back To Researches Page
|