LVA-Index: An Efficient Way to Determine Nearest Neighbors
- Piotr Paweł Lasek
In this paper we present our new LVA-Index for indexing multidimensional data. The LVA-Index has a layered structure improving performance when searching for nearest neighbors. The index combines some features of the VAFile and the NBC algorithm, namely: the idea of approximation of the data vectors and the idea of layers. The crucial advantage of the LVA-Index is that it stores n neighbor layers for each cell. For this reason, contrary to the VA-File, the LVA-Index does not require scanning of the entire approximation file. Our experiments proved that searching using the LVA-Index is faster than searching using the VA-File which was designed to effectively handle multidimensional data.
- Record ID
- Cyran Krzysztof A, Krzysztof A Cyran Kozielski Stanisław, Stanisław Kozielski Peters James F James F Peters [et al.] (eds.): Man-Machine Interactions, Advances in Intelligent and Soft Computing , vol. 59, 2009, Springer-Verlag Berlin Heidelberg, Springer, 690 p., ISBN 978-3-642-00562-6
- Keywords in English
- Applications of Mathematics, Appl.Mathematics/Computational Methods of Engineering, Artificial Intelligence (incl. Robotics), indexing, index structure, nearest neighbor search, VA-File
- DOI:10.1007/978-3-642-00563-3_65 Opening in a new tab
- http://link.springer.com/chapter/10.1007/978-3-642-00563-3_65 Opening in a new tab
- (en) English
- Score (nominal)
- Citation count
- Uniform Resource Identifier
* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or PerishOpening in a new tab system.