Summer'17 Intern-Week 1

Expectations and A look at Space-Time Kernel Density Estimation

Posted by Tahmid Efaz on June 2, 2017

Goals and Expectations

Participation in the Summer Research at the University of North Carolina at Charlotte would give me first-hand exposure to research in the field of Computer Science. Through this opportunity, I intend to sharpen my analytical and problem-solving skills and learn research techniques which might help me in the future. This summer research would help me gain knowledge in the realms of computer science where I have not set foot yet. Such topics would include parallel computing, GPU programming, and many others.

A look at Parallel Space-Time Kernel Density Estimation (STKDE)

Inside the technological boom that we are living right now, the amount of data has increased at an incredible pace. The growth of complexity in datasets calls for more interactive data analysis methods. Oftentimes, the datasets in the Geographical Information System (GIS) represent events in space and time. It is at this point where STKDE comes in. Using Space-time kernel density estimation, one can compute the specifics for data visualization. The STKDE make use of some clever algorithms to represent data GIS. On this blog post, I will look into two such algorithms involved in the Space-Time Kernel Density Estimation. These two algorithms are: the voxel-based sequential algorithm and the point-based sequential algorithm.