“Scalable Mining and Processing of Big Trajectory Data in Spark”, 11/12/2018, 13:00-14:00, University of Piraeus, Lecture room No. 338 (3rd floor)
– personal website
Tracking moving objects is now much easier thanks to the developments in positioning devices which are used in Location-Based Services and transportation. The extensive use of LBS applications imposes significant storage and processing challenges to application servers. Traditional data management systems are no longer able to analyze high volumes of trajectory data of individuals and the need for management, querying and mining of such data has opened new research directions in the field of mobility data analytics. Accordingly, most of the recent methods have been developed based on existing big data tools such as Hadoop and Spark. Due to complex properties of trajectory data, designing efficient analytics methods is not straightforward and there are a number of considerations that should be taken into account when partitioning, indexing and processing queries on these data.
In this talk, we will review the challenges we must adapt to, in management and analysis of big trajectory data, specially data partitioning, recent theoretical and practical successes in the field, and some of the targets we should aim for in future.