imkanlar
16
Oct 2017
Entropy Guided Visualization and Analysis of Multivariate Spatio-Temporal Data Generated by Physically Based Simulation

Speaker: Dr. Selçuk Sümengen
Date: October 23, 2017, 13:00
Location: EEB 1302
Title: "Entropy Guided Visualization and Analysis of Multivariate Spatio-Temporal Data Generated by Physically Based Simulation"

Abstract
: Flow fields produced by physically based simulations are subsets of multivariate spatio-temporal data. It’s difficult to extract representative views for the interpretation of fluid behavior due to data complexity. In this talk, utilization of information theory to find entropy maps for vector fields, and use of entropy maps to aid visualization of the flow fields will be presented. Principal component analysis (PCA) is used to find a projection that has the maximal directional variation in polar coordinates, producing results with fewer artifacts than the traditional methods.
Entropy guided visualizations of different data sets are presented to evaluate proposed method for the generation of entropy maps. In addition to using data sets available for research purposes, a fluid simulation framework using Smoothed-particle hydrodynamics (SPH) is also developed to produce flow fields. SPH is used to generate data sets which are difficult to interpret with direct visualization techniques. A moderate improvement for the performance and stability of SPH implementations is also proposed with the use of fractional derivatives, which are known to be useful for approximating particle behavior immersed in fluids.
 
Bio: Selcuk Sumengen received his BS, MS, and Ph.D. degrees in Computer Science and Engineering from Sabanci University in 2004, 2006 and 2013, respectively. During his graduate studies, he visited Computer Vision Laboratory at EPFL, and Computer Graphics Laboratory at UC Merced. His research focuses on physically based simulations and visualization of multivariate spatio-temporal datasets.
In 2010, he started to work at VisioThink, a spin-off from Sabancı University, as a team lead developing LBS applications for Turkcell. In the end of 2011, he joined to HAVELSAN's image processing team as a research engineer, and also worked as a project technical manager. In 2014, he was working at SAP Development Center in Istanbul as a senior software developer focusing on in-memory database, big-data, predictive and visual analytics, and data visualization. Since May 2015, he works as a chief researcher and project technical lead at TUBITAK Space Technologies Research Institute.
 
Dr. Sumengen has published three journal articles, seven conference publications and one book chapter. His main research interests are computer graphics, physically based simulations, parallel & distributed computing, and information visualization.