Urban human flow mode detection method based on deep neural network graph encoder

Abstract

The invention discloses a human flow mode detection method based on a deep neural network sparse graph encoder, and the method comprises the following steps: collecting city data which comprises humanactivity data, city interest point data and city geographic information data; dividing a city into a plurality of regions by combining an improved OPTICS algorithm and city data; according to a region division result, constructing human activity flow matrixes in multiple time periods, and constructing a human activity tensor matrix based on the human activity flow matrixes; solving structural data of the human activity tensor matrix based on a sparse graph encoder of the deep neural network; and processing the structured data by using a clustering algorithm to obtain a human flow mode in eachscene. The method can be applied to more complex data structures, more effective urban area selection is realized, the structural characteristics of the network are better reflected, the preferencesof the user to the POI in different space-time environments are deduced, the effective detection of the urban human flow mode is realized, and suggestions are provided for urban planning and POI construction.

Description

technical field [0001] The invention relates to the field of pattern analysis, in particular to a method for detecting urban human flow patterns based on a deep neural network graph encoder. Background technique [0002] With the rapid development of cities, problems such as traffic congestion, waste of resources, and dense population have arisen. On the other hand, with the popularization of positioning technology service (LTS) and the rapid improvement of computing power, human behavior in real society is digitized, forming massive data. The scale and multidimensionality of these data sets give researchers the opportunity to explore the behavioral laws of society, and to make effective predictions based on the laws to solve the problems brought about by rapid urbanization. [0003] Since the flow of people in a city is affected by time, temperature, environment, etc., different time and weather conditions will present different crowd flow patterns. If you can understand t...

AI Technical Summary

Problems solved by technology

[0004] There have been a lot of studies on exploring human flow patterns, and most scholars have discovered urban human flow patterns based on topic models. For example, B.Gao et al. discovered human flow patterns by creating topic models from digital footprints and social links, and Ziyatdinov et al. A multi-view data pattern extraction method using spectral clustering algorithm is proposed, but the research proves that the topic model is more effective for discovering potential motion patterns, but there are still some problems in the actual human flow pattern detection, mainly Because: First, urban data is very time-sensitive, and human mobility patterns are highly time-dependent, so it is difficult to be captured by topic models

Second, there is no corresponding evaluation measure in the topic model for the analysis of human mobility patterns

Finally, human traffic data is a graph structure, which increases the difficulty of the data processing process

Images