A classification framework and computational methods for human interaction analysis using movement data

Abstract

Interaction analysis for moving individuals in space and time can contribute to understanding urban dynamics and human social networks. Recent advancements in trajectory analytics have created methods to identify and extract spatiotemporal patterns of interaction using movement tracking data. However, existing definitions and classifications of interaction between moving individuals are isolated.

ORTEGA v1.0: an open-source Python package for context-aware interaction analysis using movement data

“…Dynamic interaction [ 8 , 9 ] between a pair of moving entities can be an encounter (i.e., a brief contact in space and time), or it can be either a concurrent interaction (synchronous movement in proximity over a certain time interval) or delayed interaction (or indirect/asynchronous, i.e., visiting the same location with a time lag). Traditional techniques to quantify dynamic interactions primarily rely on the proximity between two moving entities, often determined by user-defined spatial and temporal thresholds [ 8 , 10 , 11 ].…”

“…Traditional techniques to quantify dynamic interactions primarily rely on the proximity between two moving entities, often determined by user-defined spatial and temporal thresholds [ 8 , 10 , 11 ]. However, the effectiveness of the proximity-based approaches decreases when interacting individuals’ paths are not tracked simultaneously due to varied sampling rates, signal loss or imperfect tracking, or when the interactions are delayed (e.g., two animals visit the same location at different times) [ 4 , 9 ]. In contrast, the time-geographic-based approaches provide a more robust framework to identify potential encounters as well as concurrent and delayed interactions between individuals [ 6 , 9 , 12 – 15 ].…”

“…However, the effectiveness of the proximity-based approaches decreases when interacting individuals’ paths are not tracked simultaneously due to varied sampling rates, signal loss or imperfect tracking, or when the interactions are delayed (e.g., two animals visit the same location at different times) [ 4 , 9 ]. In contrast, the time-geographic-based approaches provide a more robust framework to identify potential encounters as well as concurrent and delayed interactions between individuals [ 6 , 9 , 12 – 15 ]. This is mainly because the time-geographic-based approaches incorporate the uncertainty of positioning and gaps in movement data by considering potential locations accessible to moving individuals between consecutive tracking points.…”

“…ORTEGA (Object-oRiented TimE-Geographic Analytical approach) is an emerging time-geographic-based approach which can be used to identify various types of interactions and their duration at a reasonable computation cost, as we demonstrate later in the “ Results ” section [ 6 , 9 , 15 ]. This paper implements and evaluates a new extension of the tool, ORTEGA version 1.0, as an open-source Python package 1 , 2 for analyzing, contextualizing, and mapping interactions between entities based on their movement observations (i.e., movement tracking data).…”

“…This paper implements and evaluates a new extension of the tool, ORTEGA version 1.0, as an open-source Python package 1 , 2 for analyzing, contextualizing, and mapping interactions between entities based on their movement observations (i.e., movement tracking data). The Python package ORTEGA v1.0 is built on top of the existing theoretical developments of ORTEGA, originally introduced in [ 6 ], and further extended in [ 9 ]. The developed package also offers an extension of ORTEGA for incorporating context-aware capacities to inform interaction analysis.…”