Spatial data mining approaches for GIS vector data processing

Abstract

Geographical Information Systems (GIS) are very useful system in capturing, storing, manipulating, analysing, managing, and presenting spatial data. GIS systems can be used for solving problem and making decision in various applications. Data mining is the automated process of discovering patterns in data. This thesis outlines the issues and challenges of GIS data to advance the use of data mining techniques in the context of GIS applications. This thesis focuses mainly on two domains of applications: first is the digital vector map copyright protection and second is the digital vector map partitioning. Further more, this thesis presents an efficient approach for identifying the resilient locations for embedding the watermark; improving the robustness of the watermarking approach against defined set of attacks; investigating the impact of clustering approaches on the application of vector map protection; defining an effective metric for measuring the topological distortion in the watermarked GIS maps; and developing a spatial clustering approach that takes into consideration the GIS map properties. The experimental results show the reliability of using data mining techniques in combination with GIS map properties in advancing the GIS applications with more focus on spatial data protection and partitioning.

Date of Award	Sept 2018
Original language	English
Awarding Institution	University of Portsmouth
Supervisor	Ella Haig (Supervisor)