AbstractsComputer Science

An Analysis of Hierarchical Clustering Algorithms for Hotspot Detection in Geographical Request Maps

by Paulos Kidane

Institution: KTH Royal Institute of Technology
Year: 2015
Keywords: Natural Sciences; Computer and Information Science; Computer Science; Naturvetenskap; Data- och informationsvetenskap; Datavetenskap (datalogi); Master of Science in Engineering - Computer Science and Technology; Civilingenjörsexamen - Datateknik; Datalogi; Computer Science
Record ID: 1358049
Full text PDF: http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-166420


The purpose of this thesis was to investigate the suitability of using different hierarchical clustering algorithms for locating hotspots of HTTP-requests in geographical maps. To do so, four different maps were modeled with pre-set hotspots. Nine algorithms were evaluated with a cost value to how well they could approximate the locations of the hotspots.  Four of the algorithms were found to perform better than the other ones overall. Although the results does show potential in using hierarchical clustering algorithms, the low number of models makes it hard to determine the possibility for arbitrary maps of HTTP-requests.