Monday, September 16, 2019

Simulating Urban Patterns of Life: A Geo-Social Data Generation Framework


Data generators have been heavily used in creating massive trajectory datasets to address common challenges of real-world datasets, including privacy, cost of data collection, and data quality. However, such generators often overlook social and physiological characteristics of individuals and as such their results are often limited to simple movement patterns. To address these shortcomings, we propose an agent-based simulation framework that facilitates the development of behavioral models in which agents correspond to individuals that act based on personal preferences, goals, and needs within a realistic geographical environment. Researchers can use a drag-and-drop interface to design and control their own world including the geospatial and social (i.e. geo-social) properties. The framework is capable of generating and streaming very large data that captures the basic patterns of life in urban areas. Streaming data from the simulation can be accessed in real time through a dedicated API.





Our framework allows us to define and combine each of these four concepts to create unique simulation and data generators.

  • Trigger is a mechanism that is initiated by a change or event in internal or external factors. Internal factors include one's needs, beliefs, and characteristics, while external factors include the environment or other individuals. For example, a trigger called hunger can be defined as an agent having a food level less than a specified threshold.
  • Behavior is a construct that is directly initiated by a trigger. For instance, we assume that eating is a behavior because it is directly initiated by the trigger of hunger. Inside a behavior, one can define multiple actions that make up the lifetime of the behavior.
  • Action is the process of performing certain step(s) that directly produce an output once it reaches the defined goal. Sticking with the eating behavior example, each process leading towards becoming full is considered an action. For instance, deciding whether eating at home or outside is a one-step action that generates a decision. Similarly, the process of grocery shopping is an action with multiple steps. 
  • Goal is simply the end condition for an action, such as an agent reaching a food level of 100%.


The demo will be presented at ACM SIGSPATIAL'19. For more information (sample data, figures), please visit my demo web site http://sigspatial19demo.joonseok.org 

J.-S. Kim, H Kavak, U. Manzoor, A. Crooks, D. Pfoser, C. Wenk, and A. Z├╝fle, Simulating Urban Patterns of Life: A Geo-Social Data Generation Framework, In Proceedings of ACM SIGSPATIAL GIS ’19, November 5-8, 2019. Chicago, IL, USA (to appear)