A Fast and Accurate Method for Estimating People Flow from Spatiotemporal Population Data

Abstract

Real-time spatiotemporal population data is attracting a great deal of attention for understanding crowd movements in cities.The data is the aggregation of personal location information and consists of just areas and the number of people in each area at certain time instants. Accordingly, it does not explicitly represent crowd movement. This paper proposes a probabilistic model based on collective graphical models that can estimate crowd movement from spatiotemporal population data. There are two technical challenges: (i) poor estimation accuracy as the traditional approach means the model would have too many degrees of freedom, (ii) excessive computation cost. Our key idea for overcoming these two difficulties is to model the transition probability between grid cells (cells hereafter) in a geospatial grid space by using three factors: departure probability of cells, gathering score of cells, and geographical distance between cells. These advances enable us to reduce the degrees of freedom of the model appropriately and derive an efficient estimation algorithm. To evaluate the performance of our method, we conduct experiments using real-world spatiotemporal population data. The results confirm the effectiveness of our method, both in estimation accuracy and computation cost.

Generating contextual trajectories from user profiles

“…The GPS data was collected from all the survey participants for four consecutive weeks using this mobile application. 1 https://github.com/jlinear/ConTraSim The collected survey responses are processed with the following steps to obtain the structured input (activity schedule and type file):…”

“…With the widespread use of location aware mobile devices, such as smartphones, GPS navigation devices, and smartwatches, an immense amount of spatial-temporal data can be collected to support the study of human mobility. Popular research problems in this field include location/destination prediction [6,16], people flow analysis [1,12], and location-based recommendation systems [9,13].…”

Neural Collective Graphical Models for Estimating Spatio-Temporal Population Flow from Aggregated Data

“…Although this paper focuses on population flow, our framework is applicable to other applications of collective graphical models, where auxiliary information is available to effectively estimate parameters. A number of methods have been proposed for estimating population flow from aggregated population data based on collective graphical models (Kumar, Sheldon, and Srivastava 2013;Du, Kumar, and Varakantham 2014;Sun, Sheldon, and Kumar 2015;Iwata et al 2017;Tanaka et al 2018;Akagi et al 2018). The collective flow diffusion model (Kumar, Sheldon, and Srivastava 2013) assumed static transition probabilities that do not change over time.…”

“…Iwata et al (2017) modeled the temporal dependence of transition probabilities by clustering time points using mixture models. Akagi et al (2018) modeled spatial dependence using the distance between the origins and destinations. However, these methods do not consider spatio-temporal dependence.…”

Exact and Efficient Inference for Collective Flow Diffusion Model via Minimum Convex Cost Flow Algorithm

“…(CGM) (Sheldon and Dietterich 2011), which enables us to conduct learning and inference with aggregated count data, was proposed. In particular, Collective Flow Diffusion Model (CFDM) (Kumar, Sheldon, and Srivastava 2013), which is a special case of CGM, has been proposed to infer people flows between the areas by modeling individual movements via a Markov chain approach; it has been applied to the analysis of the hidden movements behind observed count data in a traffic network (Kumar, Sheldon, and Srivastava 2013), urban space (Iwata et al 2017;Akagi et al 2018;Iwata and Shimizu 2019), amusement park (Du, Kumar, and Varakantham 2014) and exhibition halls (Tanaka et al 2018).…”