Data analysis and modeling for engineering and medical applications

Abstract

Multidimensional and multi class problems are found widely not only in chemical and food industry but also in clinical world. The integration between engineering, computing and life sciences has lead to the development of machine learning and data mining technique which facilitate the information extraction contained in dataset. The methods are not only able to identify the important variables characterizing the samples, but are also able to generate accurate and reliable rules which can be used to distinguish classes. In this work, data mining methods are applied for food product quality classification and for early illness detection. The work was then continued by developing a model which is able to capture the dynamics of blood glucose measurement of ICU patients for blood glucose control purposes.