Diagnosis of Metabolic Syndrome: A Diversity Based Hybrid Model
العنوان: | Diagnosis of Metabolic Syndrome: A Diversity Based Hybrid Model |
---|---|
المؤلفون: | Nahla Barakat |
المصدر: | Machine Learning and Data Mining in Pattern Recognition ISBN: 9783319419190 MLDM |
بيانات النشر: | Springer International Publishing, 2016. |
سنة النشر: | 2016 |
مصطلحات موضوعية: | medicine.medical_specialty, business.industry, 030209 endocrinology & metabolism, 02 engineering and technology, Type 2 diabetes, Disease, medicine.disease, Ensemble learning, Support vector machine, 03 medical and health sciences, 0302 clinical medicine, 0202 electrical engineering, electronic engineering, information engineering, Medicine, 020201 artificial intelligence & image processing, Pairwise comparison, Medical diagnosis, Metabolic syndrome, business, Intensive care medicine, Dyslipidemia |
الوصف: | Metabolic Syndrome (MetS) is a serious disorder, which is mainly characterized by central obesity, abnormal glucose tolerance, hypertension and dyslipidemia. It has been shown that 25% of adults around the world have MetS. The main concern is that those with MetS are more likely to develop type 2 Diabetes, which is found to be the fourth leading cause of global death by disease. Other life threatening complications of MetS include cardiovascular diseases (CVD), heart attack and stroke. It has also been shown that early screening and detection of people at risk may help in preventing or delaying MetS and its further complications. Within this context, data mining and machine learning can be valuable tools for identifying those people, based on their success in diagnosis and prognosis of related diseases like type 2 Diabetes. In this paper, we propose a hybrid diversity based model for diagnosis of Metabolic Syndrome. The proposed model utilizes two learning algorithms only, in particular; a Support Vector Machine (SVM) as the base-level classifier and a different classification algorithm at the meta level. The choice of the meta level classifier is based on different pairwise diversity measures. This is then followed by a final voting stage. Results on real life data set for diagnosis of MetS show that the proposed model is a promising technique, which compares favorably to other well established ensemble methods, and the choice of meta classifiers based on diversity measures was beneficial in this case. |
ردمك: | 978-3-319-41919-0 |
URL الوصول: | https://explore.openaire.eu/search/publication?articleId=doi_________::ec6a63f79672669f03f445c61ae20104 https://doi.org/10.1007/978-3-319-41920-6_14 |
حقوق: | CLOSED |
رقم الأكسشن: | edsair.doi...........ec6a63f79672669f03f445c61ae20104 |
قاعدة البيانات: | OpenAIRE |
ردمك: | 9783319419190 |
---|