Naslov
Outlier detection in experimental data using a modified expectation maximization algorithm

Autori
Novoselac, Vedran ; Pavić, Zlatko

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Proceedings of 6th International Scientific and Expert Conference of the International TEAM Society / Ádámné Major, Andrea ; Kovács, Lóránt ; Csaba Johanyák, Zsolt ; Pap-Szigeti, Róbert - Kecskemét : Faculty of Mechanical Engineering and Automation, 2014, 112-115

ISBN
978-615-5192-22-7

Skup
Technique, Education, Agriculture and Management

Mjesto i datum
Kecskemét, Mađarska, 10.11.2014. - 11.11.2014

Vrsta sudjelovanja
Ostalo

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
E-M algorithm ; Mahalanobis distance ; data clustering ; outliers

Sažetak
The paper studies the problem of clustering data sets with the E-M (Expectation Maximization) algorithm. Within the E-M algorithm is implemented procedure omitting data that have a low probability of belonging to a Gaussian mixture model components. For this purpose, the threshold is determined by the rejection of data, which are considered as outliers. For this procedure is used Mahalanobis distance of the observed data to the expectations component model that describes a particular cluster. Mahalanobis distance in this situation proved to be a good choice for Gaussian mixture models that describe clusters.

Izvorni jezik
Engleski

Znanstvena područja
Matematika, Računarstvo

POVEZANOST RADA