Where is the Ore in Your Big Data?
Speaker:Ömer Muhammet Soysal
Abstract:
In the last century, data-driven decision making is becoming more challenging due to production and processing of extremely huge amount of data acquired from a variety of sensors. The decision makers are often required to understand relations within the multi-dimensional space before taking an action, making a law, producing a product, setting up regulations, etc. In my talk, I will introduce my recently proposed algorithm Mostly Associated Sequential Patterns for extracting useful relations from big data will discuss measuring causality in association patterns.
Bio:
Dr. Soysal earned his Ph.D. degree in Computer Science at Louisiana State University in 2009. He joined the division of the Highway Safety Research Group in 2008 as a Research Associate at Louisiana State University. He is currently serving as a Research Assistant Professor at LSU. He has been supervising research and software development projects funded by the Department of Transportation and Development.
Dr. Soysal is currently conducting research in the fields of Bioinformatics, Computer Vision, Machine Learning, and Data Mining. Particularly, his research focus is on recognizing brain activities using different imaging modalities and deep learning, detection of lung cancer by means of hierarchical deep learning, causality analysis of big data by association rule mining, efficient and scalable data structures for data mining on a parallel computing framework, and mining concepts from traffic video streams.