A study of geovisual analytics for exploring event anomalies over multiple geospatial datasets

Ul Hasan, Md Monjur (2015) A study of geovisual analytics for exploring event anomalies over multiple geospatial datasets. Masters thesis, Memorial University of Newfoundland.

[English] PDF - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. Download (13MB)

Abstract

Events performed by moving entities (e.g., shoppers purchasing products from stores, tourists visiting historical places, and fishing vessels trawling) are often described by geospatial data sets. When such events are independently collected in multiple data sets, comparing the same events for positional discrepancies with their spatial and temporal contexts may reveal important insights, such as data entry, instrumental, intentional, and/or processing errors. In this work, two independently collected data sets are considered: geospatial point data describing event locations and movement data describing movement activities of the entities that performed these events. For analyzing the anomalies within these data sets a geovisual analytics approach is presented, which extracts events, identifies event anomalies, represents these anomalies on a map, and allows analysts to perform exploratory analysis to make sense of the data. This approach makes extensive use of spatial and temporal filtering, geovisualization, colour encoding, and anomaly threshold filtering. It is highly interactive, supporting analytical reasoning and knowledge discovery through visual exploration and analysis of the data sets. This approach has been implemented as a prototype system for analyzing event anomalies within two real world data sets related to fishing activities. Field trials were performed with five expert fisheries analysts to evaluate the system. Results from this study confirm the value of the approach and its potential for supporting geospatial anomaly analysis of commercial fishing events.

Actions (login required)

View Item