An individual activity place identification method based on a space-time knowledge graph

By constructing a spatiotemporal knowledge graph, quantifying the spatiotemporal correlation strength of individual activities, and using graph clustering algorithms, the problem of misjudgment in location identification in existing technologies is solved, achieving more accurate activity location identification and parameter robustness.

CN118349867BActive Publication Date: 2026-05-19TONGJI UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
TONGJI UNIV
Filing Date
2024-04-12
Publication Date
2026-05-19

AI Technical Summary

Technical Problem

Existing technologies struggle to simultaneously consider temporal and spatial relationships when identifying individual activity locations, leading to misjudgments in location identification and poor parameter robustness, especially when dealing with atypical outliers.

Method used

By constructing an individual activity location identification method based on spatiotemporal knowledge graph, a spatiotemporal reference system is reconstructed using mobile communication data, the spatiotemporal correlation strength of individual activities is quantified, and a graph clustering algorithm is used to divide activity communities and map them onto geospace to determine activity locations.

Benefits of technology

It achieves the ability to limit the size of the venue within a reasonable range, distinguish between waypoints and activity points, improves parameter robustness, and can distinguish activity venues with different time characteristics. The identification results are consistent with the land use function type.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118349867B_ABST
    Figure CN118349867B_ABST
Patent Text Reader

Abstract

The application relates to an individual activity site identification method based on a space-time knowledge graph, which comprises the following steps: extracting multi-day mobile information of a user by studying mobile communication data in a research area, reconstructing a space-time reference system of the research area, and constructing a space-time knowledge graph of human mobile behavior based on the reconstructed space-time reference system; deducing spatial proximity and time co-occurrence between individual stays based on the space-time knowledge graph of human mobile behavior; constructing a space-time correlation graph of individual stays based on the deduced spatial proximity and time co-occurrence between individual stays, dividing the graph into densely connected subgraphs by using a graph clustering method, and finally mapping the subgraphs to a geographical space to determine activity sites. Compared with the prior art, the application has the advantages of adaptively limiting the scale of a site and the length of a single activity, high robustness, and the provision of site information with practical significance for subsequent individual mobile mode analysis.
Need to check novelty before this filing date? Find Prior Art