Method for searching high-dimensional vector combining clustering and city block distances

Abstract

The invention discloses a method for searching a high-dimensional vector combining clustering and city block distances. In the method, an index structure CBlockB-tree combining clustering and the city block distances is provided; firstly, cluster partition is performed on high dimensional vector sets by adopting a clustering algorithm, and then BlockB-tree is constructed for each cluster data to form the CBlockB-tree. When the index structure performs searching, a part of cluster data which are disjointed with a query region can be filtered through clustering; the operation amount matched with final vector similarity can be further reduced by comparing Key values transformed from a high dimension to one dimension, and therefore the searching speed of the high-dimensional vector can be increased; meanwhile, the index structure is capable of effectively supporting the simple and efficient city block distances for matching searching.

Description

technical field [0001] The invention belongs to the fields of data processing such as multimedia information retrieval, intelligent information processing, and data mining, and specifically relates to a high-dimensional vector search method combining clustering and block distance. Background technique [0002] With the development of computer and information technology, a large amount of multimedia data has been produced. How to quickly find the required information in the massive multimedia database is a key issue in the field of multimedia database research. The traditional method is to manually mark the multimedia data, and then realize the multimedia information retrieval through text retrieval. However, manual labeling has the disadvantages of heavy workload and strong subjectivity. For the explosive growth of multimedia data, manual labeling is impossible. Therefore, it is necessary to study content-based multimedia information retrieval technology. [0003] The techn...

AI Technical Summary

Problems solved by technology

[0007] Pyramid technology, NB-tree, iDistance, iMinMax and other high-dimensional to one-dimensional conversion index structures achieve filtering and pruning through simple comparison of a single key value. Although complex distance calculations are not required and high retrieval efficiency is high, high-dimensional The process of converting to one-dimensional can cause a large amount of data information loss, resulting in different vectors may have the same one-dimensional key value, so that only a small proportion of data can be filtered out through a single key value, resulting in the operation of the final likelihood matching process The amount is still large, and the query overhead is still not small

Images