Notice: Undefined index: linkPowrot in C:\wwwroot\wwwroot\publikacje\publikacje.php on line 1275
[30605] Artykuł: Spectral clustering based on analysis of eigenvector propertiesCzasopismo: Proceedings of Computer Information Systems and Industrial Management Applications, Lecture Notes in Computer Science, Springer Verlag Tom: 1, Strony: 43-54ISSN: 0302-9743 ISBN: 978-3-662-45237-0 Wydawca: SPRINGER-VERLAG BERLIN, HEIDELBERGER PLATZ 3, D-14197 BERLIN, GERMANY Opublikowano: Listopad 2014 Seria wydawnicza: Lecture Notes in Computer Science Liczba arkuszy wydawniczych: 0.60 Autorzy / Redaktorzy / Twórcy
Grupa MNiSW: Materiały z konferencji międzynarodowej (zarejestrowane w Web of Science) Punkty MNiSW: 15 Klasyfikacja Web of Science: Proceedings Paper Pełny tekst Web of Science Keywords: spectra clustering  nearest neighbor graph  |
In this paper we propose a new method for choosing the number of clusters and the most appropriate eigenvectors, that allow to obtain the optimal clustering. To accomplish the task we suggest to examine carefully properties of adjacency matrix eigenvectors: their weak localization as well as the sign of their values. The algorithm has only one parameter - the number of mutual neighbors. We compare our method to several clustering solutions using different types of datasets. The experiments demonstrate that our method outperforms in most cases many other clustering algorithms.