Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

An adaptive clustering method for trajectory data parameters

A technology of trajectory data and clustering methods, applied in special data processing applications, geographic information databases, structured data retrieval, etc., to achieve the effect of ensuring effectiveness and accuracy and reducing the amount of processed data

Active Publication Date: 2022-07-29
QINGDAO UNIV OF SCI & TECH
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Aiming at the problems existing in the existing DBSCAN clustering method, the present invention provides a clustering method for adaptive trajectory data parameters

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • An adaptive clustering method for trajectory data parameters
  • An adaptive clustering method for trajectory data parameters
  • An adaptive clustering method for trajectory data parameters

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0051] combine Figure 1 to Figure 3 By analyzing the trajectory data generated by mobile phone signaling, the traffic distribution characteristics can be obtained, which can provide a valuable reference for urban intelligent planning and urban management.

[0052] The method for clustering the trajectory data generated by mobile phone signaling using the method of the present invention is as follows:

[0053] A clustering method for parameter adaptation of trajectory data, comprising the following steps:

[0054] Step 1: Input the total trajectory data, set the parameters T and α, where T is the time interval for taking the data, and α is the confidence coefficient.

[0055] The setting of T can be taken according to the multiple of the data positioning interval of the time series. For example, if the positioning interval between data points is 10s, T can be set as multiple positioning intervals. The value of a is greater than 0 and less than 0.5.

[0056] In this embodime...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses an adaptive clustering method of trajectory data parameters. First, parameters T and α are set, and then the intersection part of the data is taken as the sample point to be processed according to the time dimension, and confidence interval and distribution divergence are introduced as evaluation The standard ensures the validity and accuracy of the sample point data, thereby reducing the amount of processed data. After that, use the intersection data to introduce a method of measuring point density to solve the problem of eps value setting, and find out the maximum number of samples per unit and the minimum number of samples per unit according to the sliding window method, and the operation of taking the mean is used as MinPts setting value. The adaptive clustering method of trajectory data parameters provided by the present invention not only solves the shortcoming that the DBSCAN clustering algorithm itself can only handle a small amount of data, but also can adaptively set the eps and MinPts values ​​in the DBSCAN clustering algorithm, so as to solve the problem of existing problems. In the prior art, it can only be manually set based on experience, resulting in an unsatisfactory clustering effect.

Description

technical field [0001] The invention relates to the field of data clustering algorithms, in particular to a clustering method for adaptive trajectory data parameters. Background technique [0002] Today's data hotspot clustering algorithm mainly uses a DBSCAN algorithm. As a density clustering, the DBSCAN algorithm is widely used in data mining and analysis by scholars. The algorithm measures its density depending on the unit hypersphere. The number of samples can not only cluster clusters of different shapes, but also explore outliers. However, the DBSCAN algorithm must set two parameters EPS and MinPts before use. These two parameters are manually set according to their own experience, and the quality of the clustering effect directly depends on whether the set parameters are suitable for their own data sets. There is basically no basis for setting algorithm parameters without knowing the data scale and data distribution. In addition, the DBSCAN algorithm itself can only...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/62G06F16/29
CPCG06F16/29G06F18/2321G06F18/23
Inventor 徐文进
Owner QINGDAO UNIV OF SCI & TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com