Title: Scalable  Methods for Big  Spatial Data Processing

Abstract: The rapid growth of data in recent years has posed significant challenges for complex and data-intensive operations, as many traditional data management techniques are inapplicable to such large data volumes, creating the need for scalable approaches. In this talk, we focus on two examples of such data-intensive operations that are typically used in large-scale data analysis, namely correlation measurement (via Kendall’s Tau) and spatial joins, and present scalable algorithms for efficient execution on large datasets. Specifically, for correlation measurement, we address the problem via a geometric approach that allows the computation of correlation information en masse. For spatial joins, we propose an adaptive parallelized approach for minimizing data replication that improves the performance of query processing, and excels over state-of-the-art algorithms especially on skewed datasets.