Thesis topic proposal
László Aszalós
Practical correlation clustering


Institute: University of Debrecen
computer sciences
Doctoral School of Informatics

Thesis supervisor: László Aszalós
Location of studies (in Hungarian): University of Debrecen Doctoral School of Informatics
Abbreviation of location of studies: ITDI

Description of the research topic:

In the majority of non-supervised learning algorithms, clustering is based on the distance between objects, however correlation clustering uses the similarity of objects. This allows us to introduce machine learning in new areas where earlier clustering methods did not allow for it. Our research focuses on the discovery of these areas and the construction of the tolerance correlation suitable for these cases. Emphasis is placed on solving various practical problems, taking into account the issue of efficiency.

[1] Bansal, Nikhil, Avrim Blum, and Shuchi Chawla. "Correlation clustering." Machine learning 56.1-3 (2004): 89-113.
[2] Kim, Sungwoong, et al. "Image segmentation using higher-order correlation clustering." IEEE transactions on pattern analysis and machine intelligence 36.9 (2014): 1761-1774.
[3] Samal, Mamata, V. Vijaya Saradhi, and Sukumar Nandi. "Scalability of correlation clustering." Pattern Analysis and Applications 21.3 (2018): 703-719.
[4] Keuper, Margret, et al. "Motion segmentation & multiple object tracking by correlation co-clustering." IEEE transactions on pattern analysis and machine intelligence (2018).

Deadline for application: 2022-11-15

2022. X. 05.
ODT ülés
Az ODT következő ülésére 2022. december 2-án 10.00 órakor kerül sor a Semmelweis Egyetem Szenátusi termében (Bp. Üllői út 26. I. emelet).

All rights reserved © 2007, Hungarian Doctoral Council. Doctoral Council registration number at commissioner for data protection: 02003/0001. Program version: 2.2358 ( 2017. X. 31. )