A SCALABLE FRAMEWORK FOR AIS DATA ANALYSIS AND VISUALISATION
Abstract
Maritime traffic data is an important resource to understand vessel activities. Several topics, such as maritime traffic anomaly detection, have been widely studied. With the increase in the number of vessels, location, transponders and satellite receiving statistics, the size of these data is huge and rapidly increasing. A single machine or a sequential algorithm may not be able to handle data at this scale. Due to the lack of scalable tools, many problems on marine data have only been studied on relatively small datasets. In this thesis, we propose a framework, AIS Data Explorer, to analyse and visualise marine trajectory data. The framework achieves the following goals: scalability, sup- port for big data visualisation and acceleration of large scale data analysis. In our experiments, using 5 worker in- stances for a dataset size of 100 million items, the framework visualised a global heat map in 20 seconds.