Nestriktas klasterizācijas metodes balstītas uz F-transformētiem datiem
Loading...
Date
Authors
Advisor
Journal Title
Journal ISSN
Volume Title
Publisher
Latvijas Universitāte
Language
N/A
Abstract
Darbs ir veltīts F-transformācijām un to pielietojumam datu klasterizācijas uzde-
vumā. Darbā tika apskatītas nultās un pirmās pakāpes F-transformācijas, nestriktas
klasterizācijas metodes, kā arī validācijas indeksi un nestriktas klasifikācijas metodes.
Apskatītas datu transformācijas un klasterizācijas metodes tika izmēģinātas uz uzģenerē-
tām laikrindām, kas apraksta tīkla trafika plūsmu. Tika veikta iegūto rezultātu analīze
un interpretācija.
The Paper is devoted to F-transform and its implementation in data clustering problem. Zero and first degree F-transforms, fuzzy clustering methods as well as valida- tion index and fuzzy classification methods were reviewed in this Paper. Reviewed data transforming and clustering methods have been used on generated time series, which si- mulate network traffic flow. The analysis and interpretation of the obtained results have been made.
The Paper is devoted to F-transform and its implementation in data clustering problem. Zero and first degree F-transforms, fuzzy clustering methods as well as valida- tion index and fuzzy classification methods were reviewed in this Paper. Reviewed data transforming and clustering methods have been used on generated time series, which si- mulate network traffic flow. The analysis and interpretation of the obtained results have been made.