Search for abstracts of CN`99-15
Home > Description
Topic: Unusual Change Detection and Mining Unstructured Data
Abstract: In this talk we review and present a novel approach to data mining based on the Gestalt theory in Computer Vision and human perception. We also explain mathematical background of "a contrario" detection theory in data mining and image processing. Several new applications of image processing for text and log-files mining and data segmentations will be presented. Relations between "a contrario" modelling and deep learning will be discussed.
Key Speaker: Alexander BALINSKY (Cardiff University, UK)
Prof. Alexander Balinsky received his PhD degree in Mathematical Physics from the Landau Institute of Theoretical Physics in 1990 and was Research Fellow in the Department of Mathematics at The Technion-Israel Institute of Technology from 1993 till 1997. He joined Cardiff University in 1997. He is a Professor in the Cardiff School of Mathematics and WIMCS (Wales Institute of Mathematical and Computational Sciences) Chair in Mathematical Physics. His current research interests lie in the areas of spectral theory, stability of matter, image processing and machine learning. He has participated in EU TMR network on Partial Differential Equations and Quantum Mechanics (1996-2001). He was PI on three years grant from United State-Israel Binational Science Foundation (1996-1999), on three years EPSRC Research Grant 2003-2006. He was founding member of Cardiff Communication Research Center. In 2007-2011 he had joint with Hewlett-Packard EPSRC CASE award, and from October 2011 joint with Hewlett-Packard 50%-50% PhD Scholarship. He also did consultancy work for Reuters, London on mathematical models for Internet Security.
His Impact Case Study "Meeting the Challenges of Data Security: Detecting Unusual Behaviour and Mining Unstructured Data" was featured in the leading article 'The impact of impact' in Times Higher. Out of almost 7,000 case studies that were submitted to REF2014 across all units of assessment from all universities, this case study was one of only 8 highlighted in the article! He was invited by Springer to prepare an article about this Impact Case Study to be publish in a special book "REF 2014 Impact Cases: UK Success Stories in Industrial Mathematics". The book was published in February 2015.
Currently he is PI on joint with Hewlett-Packard and Dyfed Powys Police on “Crime Analysis and Predictive Policing”.