Visualization and analysis of geographic information: algorithms and data structures

Abstract

In recent years, Geographic Information Systems have witnessed a large increase in data availability. There is a need to process a large amount of data before it can be managed and analysed. This project aims to develop an application operating through a Web platform in order to allow for a low cost and simpli ed integration, management and manipulation of georeferenced information. Special emphasis is given to the implementation of e cient clustering algorithms for nding a representative set of points in a map. In the thesis, this representation problem is formulated as two classic optimisation problems: the k-center and the geometric disk cover. The approaches covered in this thesis include exact algorithms for solving the k-centre problem, as well as approximation algorithms and heuristic methods to solve the geometric disk cover problem. The algorithms are experimentally evaluated in a wide range of scenarios.