Identifying Anomalies in Location Data and Movement Pattern Analysis
AbstractDespite numerous technological advancements GPS data remains unreliable. Location services rely on this data and are so integrated into society that improvement in this field is imperative. As streams of GPS data are recorded and analysed constant erroneous values go undetected and are processed by applications with the remaining accurate information. This level of inaccuracy harms city planning, infrastructure development and user engagement with applications. Using SAS analytics to process the data these anomalies can be identified. The process of univariate analysis, time series line graph analysis and nearest neighbour diagrams have been used to visualise these issues. From these visualisations, it is apparent that clustering methods can highlight anomalies. However, due to the uncertain nature of human transport they cannot be applied to every sample.Through this process other observations have been made surrounding the context of the information collected. Using the basic location information and visualisations, trends showing type of transport used, road types and traffic patterns have been identified. This demonstrates the untapped potential of location data and the need for increased context during collection.
How to Cite
GEORGE, Mark Edward. Identifying Anomalies in Location Data and Movement Pattern Analysis. Discovery, Invention & Application, [S.l.], june 2017. Available at: <https://computing.derby.ac.uk/ojs/index.php/da/article/view/225>. Date accessed: 22 aug. 2019.