This article describes an improvement for K-means algorithm and its application in the form of a system that clusters search results retrieved from Wikipedia. The proposed algorithm eliminates K-means isadvantages and allows one to create a cluster hierarchy. The main contributions of this paper include the ollowing: (1) The concept of an improved K-means algorithm and its application for hierarchical clustering. (2) Description of the WikiClusterSearch system that employs the proposed algorithm to organize Wikipedia search results into clusters.
Authors
Additional information
- DOI
- Digital Object Identifier link open in new tab 10.1109/inista.2011.5946070
- Category
- Aktywność konferencyjna
- Type
- publikacja w wydawnictwie zbiorowym recenzowanym (także w materiałach konferencyjnych)
- Language
- angielski
- Publication year
- 2011
