k-Nearest neighbor searching in hybrid spaces

作者:

Highlights:

• Developed algorithm for searching multi-dimensional hybrid data spaces.

• Introduced the method of improving search performance by examining search stages.

• Suggested new search heuristic to improve initial stage of searching by 33%.

• Derived theoretical model accurately predicts the performance of algorithm.

摘要

Highlights•Developed algorithm for searching multi-dimensional hybrid data spaces.•Introduced the method of improving search performance by examining search stages.•Suggested new search heuristic to improve initial stage of searching by 33%.•Derived theoretical model accurately predicts the performance of algorithm.

论文关键词:Hybrid data space,Nearest neighbor search,Similarity search,Spatial indexing,Information retrieval

论文评审过程:Received 18 September 2010, Revised 12 December 2013, Accepted 19 February 2014, Available online 26 February 2014.

论文官网地址:https://doi.org/10.1016/j.is.2014.02.004