GeoAI and Deep Learning Symposium: Spatial-Temporal Modeling and Data Mining I

Type: Paper
Theme: Geography, GIScience and Health: Building an International Geospatial Health Research Network (IGHRN)
Sponsor Groups: Spatial Analysis and Modeling Specialty Group, Geographic Information Science and Systems Specialty Group, Cyberinfrastructure Specialty Group
Poster #:
Day: 4/4/2019
Start / End Time: 3:05 PM / 4:45 PM
Room: Capitol Room, Omni, East
Organizers: Yi Qiang, Yihong Yuan, Ying Song
Chairs: Yi Qiang

Call for Submissions

The modeling of time and space is fundamental to GIScience and critical for research of human dynamics. Currently, time- and geo-tagged data are generated at an unprecedented speed, which creates ample opportunities to investigate human dynamics from multiple perspectives at nearly real-time. In the meanwhile, the complexity, heterogeneity, and uncertainty of the data posed challenges for the GIS community to develop advanced analytical tools and modeling frameworks to convert the data into useful information and knowledge. The development of artificial intelligence (AI) and deep learning (DL) have facilitated spatial modeling tasks such as spatial feature detection and image classification. However, we argue that the potential of these techniques in modeling the dynamic aspects of geospatial phenomena has not been fully exploited. This session would call abstracts in both fundamental research of space-time modeling in the era of Big Data and applications of cutting-edge AI and DL techniques in spatio-temporal data mining and analytics.

Example topics include but are not limited to:

(1) Conceptual and computational framework of space and time
(2) Novel representations, data models, and analytical tools for spatio-temporal data
(3) Applications of AI and/or DL in spatio-temporal data mining and modeling
(4) The effect of scale in spatio-temporal modeling
(5) Machine learning of dynamic processes (e.g.  transition/behavior rule modeling, Markov decision processes)
(6) Dynamic modeling and simulation (e.g. agent-based modeling and cellular automata)

To present your paper in the session, please submit your abstract to the AAG annual meeting website, and then send the title, abstract and your PIN to Yi Qiang (yiqiang@hawaii.edu) by Wednesday October 25th or the extended deadline.

Organizing committee:
Yihong Yuan, Texas State University, y_y18@txstate.edu
Ying Song, University of Minnesota, yingsong@umn.edu
Yi Qiang, University of Hawaii – Manoa, yiqiang@hawaii.edu


Description

Abstract of Chaired Symposium
GeoAI and Deep Learning Symposium: Spatial-Temporal Modeling and Data Mining I
2019 AAG Annual Meeting, Washington DC, April 3-7, 2019


The modeling of time and space is fundamental to GIScience and critical for research of human dynamics. Currently, time- and geo-tagged data are generated at an unprecedented speed, which creates ample opportunities to investigate human dynamics from multiple perspectives at nearly real-time. In the meanwhile, the complexity, heterogeneity, and uncertainty of the data posed challenges for the GIS community to develop advanced analytical tools and modeling frameworks to convert the data into useful information and knowledge. The development of artificial intelligence (AI) and deep learning (DL) have facilitated spatial modeling tasks such as spatial feature detection and image classification. However, we argue that the potential of these techniques in modeling the dynamic aspects of geospatial phenomena has not been fully exploited. This session would call abstracts in both fundamental research of space-time modeling in the era of Big Data and applications of cutting-edge AI and DL techniques in spatio-temporal data mining and analytics.

Example topics include but are not limited to:

(1) Conceptual and computational framework of space and time
(2) Novel representations, data models, and analytical tools for spatio-temporal data
(3) Applications of AI and/or DL in spatio-temporal data mining and modeling
(4) The effect of scale in spatio-temporal modeling
(5) Machine learning of dynamic processes (e.g.  transition/behavior rule modeling, Markov decision processes)
(6) Dynamic modeling and simulation (e.g. agent-based modeling and cellular automata)

To present your paper in the session, please submit your abstract to the AAG annual meeting website, and then send the title, abstract and your PIN to Yi Qiang (yiqiang@hawaii.edu) by Wednesday October 25th or the extended deadline.

Organizing committee:
Yihong Yuan, Texas State University, y_y18@txstate.edu
Ying Song, University of Minnesota, yingsong@umn.edu
Yi Qiang (chair), University of Hawaii – Manoa, yiqiang@hawaii.edu


Agenda

Type Details Minutes Start Time
Presenter Jeon-Young Kang*, CyberGIS Center for Advanced Digital and Spatial Studies, University of Illinois at Urbana-Champaign, Jared Aldstadt, SUNY-Buffalo, Using Multiple Scale Space-Time Patterns in Sensitivity Analysis for Spatially Explicit Agent-Based Models 20 3:05 PM
Presenter Fei Wang*, , Nina Lam, Louisiana State University, The Spatio-Temporal Processes of Land Cover Land Use Change and the Process-Based Prediction of Urban Growth in the Texas Triangle 20 3:25 PM
Presenter Yang Zhang*, University College London, Yibin Ren, SpaceTimeLab, University College London, Tao Cheng, SpaceTimeLab, University College London, A Spatial-Temporal Deep Learning Framework for Network-based Traffic Flow Prediction 20 3:45 PM
Presenter Jesus Martinez-Manso*, Planet Labs, Spatio-temporal deep learning with daily medium-res satellite imagery 20 4:05 PM
Presenter Alexander Hohl*, Utica College, Eric Delmelle, University of North Carolina at Charlotte, Wenwu Tang, University of North Carolina at Charlotte, Xun Shi, Dartmouth College, Detecting Space-Time Patterns Under Non-Stationary Background Population 20 4:25 PM

To access contact information login