Concordia University

http://www.concordia.ca/content/shared/en/news/encs/computer-science/2018/05/16/Seminar-by-Javad-Sadri.html

notice

Seminar by Dr. Javad Sadri (Concordia University)

May 16, 2018

Speaker: Dr. Javad Sadri (Concordia University)                                                                                        

Title:  Clustering Big Data Networks Based on Fractal Theory: Applied to Motif Discovery


Date: Wednesday, May 16th, 2018


Time: 10:30 am - 12 noon


Room: EV 11.119

ABSTRACT

Motif discovery in big data networks is the process of finding meaningful, unfamiliar and frequent patterns to discover knowledge from data. Motif discovery in complex and social networks have received a lot of attention in recent years. However, it is still a challenging task in big data analytics and data mining. Motif discovery has numerous applications for revealing hidden frequent structures, functional building blocks, or knowledge discovery in domains such as engineering, cheminformatics, bioinformatics, genomics, neurobiology, sociology, and ecology.

In the first part of this talk, based on fractal theory and a novel cluster growing algorithm, a motif localization method for complex and big data networks will be presented. In our method, for each big data network, a corresponding multifractal network is generated which then is adaptively partitioned into several clusters. Afterwards, top ranked clusters are chosen to predict locations of motifs in the input network. We show that the proposed method provides an efficient solution for motif localization. Also pruning non-promising areas of complex networks speeds up current motif discovery algorithms. Experimental results show that our algorithm efficiently deals with complex networks representing large datasets with high-dimensionality such as WikiTalk, BioGRID, Amazon, Facebook, Twitter, YouTube, etc. Our method also provides stimulus for future research on properties and applications of fractals in big data and complex networks.

In the second part of this talk, some of our recent research in big data analytics domain will be presented. In this part, I will introduce some big databases that my group has created in different fields such as document analysis & recognition, check processing, screening of children with learning disabilities, and recognition of minerals in thin sections of rocks.

 

 

 

BIO

Javad Sadri is an assistant professor (LTA) at the Computer Science and Software Engineering Department of Concordia University. He obtained his Ph.D. (in 2007) in the field of Pattern Recognition and Machine Learning at CENPARMI (Center for Pattern Recognition & Machine Intelligence) at Concordia University. Then he spent one year of his postdoctoral research at CENPARMI and two years on big data in bioinformatics at McGill Center for Bioinformatics (MCB). He was an assistant professor at the Computer Engineering Department of University of Birjand in Iran and Chair of his department from 2010 to 2013. Then, he was a visiting professor at McGill center for Bioinformatics, McGill University in 2013-2014.

He has supervised and co-supervised more than 30 graduate students at the Ph.D. and Master level to the successful completion of their thesis. Javad has published more than 20 publications in journals such as IEEE/ACM Transactions on Computational Biology and Bioinformatics, Pattern Recognition, Journal of Bioinformatics, Image and Vision Computing, Pattern Analysis and Applications and contributed to a book chapter  published by John Wiley & Sons Inc. He has also published and presented more than 30 conference papers at conferences such as ICPR, IWFHR (Best paper award recipient), IEEE WCCI, HGM, ICPRAI. Javad also serves as a reviewer for several international journals, committee member and organizer in several related conferences. He was the recipient of the FQRNT, graduate fellowship, and best paper awards. His current research interests are applications of pattern recognition and machine learning techniques in big data analytics, Bioinformatics, affective computing, and document analysis.

 




Back to top

© Concordia University