Personalized recommendation method for nearest neighbor query based on big trajectory data

Abstract

The invention discloses a personalized recommendation method for a nearest neighbor query based on big trajectory data. In the method, an efficient storage and index structure is designed for the nearest neighbor query based on the big trajectory data to process the big trajectory data. The method comprises the steps of firstly carrying out extraction, noise reduction, conversion and storage processing on the big trajectory data; then establishing an overall R-tree index and a local R-tree index for the stored trajectory data; establishing a trajectory serial number set-based index and a trajectory number-based index for each sub-region; and when a user submits a query, visiting the index structure, carrying out the nearest neighbor query based on the big trajectory data so as to provide personalized recommendation service. The method satisfies the demand of the nearest neighbor query of a trajectory in a big data environment well, greatly improves the processing efficiency of the nearest neighbor query based on the big trajectory data and provides the optimal performance.

Description

technical field [0001] The invention relates to an indexing and query technology in the field of computer spatial databases, in particular to a personalized recommendation method based on nearest neighbor query of trajectory big data. Background technique [0002] With the explosive growth and widespread popularization of devices with GPS, spatio-temporal trajectory data (such as people, vehicles and animals, etc.) Applications and services in many fields, including animal behavior research. [0003] In the current era of big data, query algorithms for trajectory big data are also receiving more and more attention. Among them, the nearest neighbor query based on trajectory big data is an important query method and has important practical application value. The nearest neighbor query for trajectory big data refers to finding the trajectory with the closest distance to a specified query object from the trajectory dataset. The nearest neighbor query based on trajectory big d...

AI Technical Summary

Problems solved by technology

[0004] However, designing such an approach faces two challenges

First, the existing scale of trajectory data is very large and the growth rate is very fast, but most of the most effective trajectory processing algorithms are based on a centralized system architecture that is not easy to expand. Due to the large amount of data, the processing efficiency of a single machine is too low not even available

Second, the distributed frameworks designed and built for multi-dimensional data in recent years are not fully suitable for processing large-scale trajectory data.

Without modifications to these systems, it is impossible to integrate optimizations for trajectory data to efficiently support nearest neighbor query algorithms in trajectory big data environments

Images