Big data has become available due to the large number of applications that generate them with minimum effort. However, this has resulted in new challenges such as limited storage capacity and difficulty in processing large amounts of scattered data. Additionally mathematical properties of these data are not clear. This limits the ability of mathematical parametric models to deal with big data while the advantage of nonparametric and empirically designed models becomes more evident. We discuss common machine learning techniques typically used in classification applications in addition to fusion strategies used to enhance the performance of classifier systems. Additionally, we discus feature selection techniques used to reduce the large dimensionality of data with an example of an application in the medical area
Dr. Fuad M. AlkootCyberfusion Center for Technological Consultations HITN, PAAET
Fuad M. Alkoot has obtained his PhD in Electronic Engineering from University of Surrey – CVSSP in 2001. He has obtained his MSc in Electrical Engineering from Rochester Institute of Technology, USA, in 1989 and his BSc in Electrical Engineering in 1987. He is currently a Lecturer at PAAET-HITN. His research interests include statistical pattern recognition, combiner design and fusion strategies.
Fill the form below to get directions to our event location
If you have any questions about the event, please contact us directly. We will respond for sure.