IVML  
  about | r&d | publications | courses | people | links
   

C. S. Kyrkou, Ph. Mylonas, E. Spyrou, S. Sioutas, A. Rapti, D. Tsolis
Evaluating Clustering Algorithms on a Multimedia Metadata Dataset
20th Pan-Hellenic Conference on Informatics (PCI 2016), Patras, Greece, 10-12 November 2016
ABSTRACT
In the framework of informatics, data analysis always played a crucial role in understanding various phenomena. When it comes to multimedia content, this task is getting more difficult to tackle in an efficient manner. In principle, cluster analysis, i.e., primitive data exploration with little or no prior knowledge, consists of research developed across a wide variety of approaches. The aim of this paper is to present a comparative survey of five clustering algorithms, namely k means, EM, DBSCAN, Mean Shift and KVQ applied on a real-life multimedia metadata dataset derived from Flickr social network.
10 November , 2016
C. S. Kyrkou, Ph. Mylonas, E. Spyrou, S. Sioutas, A. Rapti, D. Tsolis, "Evaluating Clustering Algorithms on a Multimedia Metadata Dataset", 20th Pan-Hellenic Conference on Informatics (PCI 2016), Patras, Greece, 10-12 November 2016
[ save PDF] [ BibTex] [ Print] [ Back]

© 00 The Image, Video and Multimedia Systems Laboratory - v1.12