change detection and clustering in real time

Graduate Thesis uoadl:1324086 468 Read counter

Unit:
Τομέας Υπολογιστικών Συστημάτων και Εφαρμογών
Library of the School of Science
Deposit date:
2015-05-28
Year:
2015
Author:
Σκαράκης Γεράσιμος
Supervisors info:
Ευστάθιος Χατζηευθημιάδης
Original Title:
ανίχνευση αλλαγών και συσταδοποίηση σε πραγματικό χρόνο
Languages:
Greek
Translated title:
change detection and clustering in real time
Summary:
While the volume of information that is available increases with an exponential
rate over time, the need to export the useful knowledge that is accumulated in
the information. The study of the techniques through with which this is
achieved, is an important field of research.
The purpose of this project is to look at the ways in which I can find changes
in datasets with numerical data, using change detection and clustering
algorithms. With experimental studies I analyze the pros and cons of each
algorithm and I conclude to the most effective metrics for this purpose.
Keywords:
change detection, clustering, algorithm
Index:
Yes
Number of index pages:
5
Contains images:
Yes
Number of references:
21
Number of pages:
53
document.pdf (691 KB) Open in new window