MB-tree: A quick indexing structure for multi-dimensional space
Hai Ying Huang, Ying Hou, Liu Li-xiong, Fucai Chen
Abstract
Hai Ying Huang, Ying Hou, Liu Li-xiong, Fucai Chen
Abstract
Efficient indexing structure is the key of multi-dimension retrieval issue. Lots of indexing structures are failed in the parallelism. This paper presents multi-branch indexing tree (MB-tree), using the nearest neighbor criterion to realize the multi space division of the data sets. The experimental results indicate that the structure is suitable for parallel computing and can significantly improve the speed of similarity query.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Efficient indexing structure is the key of multi-dimension retrieval issue. Lots of indexing structures are failed in the parallelism. This paper presents multi-branch indexing tree (MB-tree), using the nearest neighbor criterion to realize the multi space division of the data sets. The experimental results indicate that the structure is suitable for parallel computing and can significantly improve the speed of similarity query.
Key concepts: Search engine indexing, Computer science, Data structure, Nearest neighbor search, Tree (set theory), Dimension (graph theory), Tree structure, k-nearest neighbors algorithm