AdobeStock_268518669.jpeg

GEAR Lab's Presentations at AAG 2022!

A huge shout out to all the GEAR Lab members who have presented their respective research in this AAG 2022! These seminar recordings are viewable on the session gallery until August 29, 2022. GEAR lab delivered eight presentations at AAG2022.



 

Navigation List of All Presentations:


Social Media for Emergency Rescue: An Analysis of Rescue Requests on Twitter during Hurricane Harvey


Community awareness and sentiment inequalities during Winter Storm Uri


Revealing the Global Linguistic and Geographical Disparities of Public Awareness to Covid-19 Outbreak through Social Media


A modular machine learning Platform for Resilience Inference Measurement and Enhancement


Spatial-Temporal Changes of Flood Risk in the United States


Simulating land loss and land gain by integrating neighborhood effect and deep learning with cellular automata


A Community-Based Assessment of Flash-Flood Vulnerability in Texas and the Influence from Different Social Variables

 

Social Media for Emergency Rescue: An Analysis of Rescue Requests on Twitter during Hurricane Harvey

Presented by: - Dr. Lei Zou



Although social media have become a significant player in the disaster response procedure, effectively using it for rescue purposes is still challenging. This study thereby focuses on understanding the characteristics of rescue-request messages; revealing the spatial-temporal patterns of rescue requests; determining the social-geographical conditions of communities needing rescue and identifying the challenges of using social media for rescue and proposing improvement strategies. Inferentially, a framework is provided that embodies the steps and strategies needed to improve social media use for rescue operations.


| Learn More! | View the Session! | Meet Us! | Back to List! |



 


Community awareness and sentiment inequalities during Winter Storm Uri

Presented by: - Dr. Heng Cai



External stressors, like pandemics and natural disasters, have been demonstrated to cause emotional distress and depression. The Covid-19 pandemic and lockdown/social distancing policies, compounded by the Winter Storm Uri in early 2021, have caused tremendous infrastructural damages and disruptions to social networks and various degrees of mental stress and depression among different populations in Texas communities. Therefore, in this study, Twitter data was analyzed to develop community awareness and sentiment indices. As Social media data harvest users’ digital traces that rapidly reflect their experiences and subjective feelings at a low cost and reliable manner, this study identifies the happiness levels of communities; and if low, has elucidated the underlying reasons and arguments behind it.


| Learn More! | View the Session! | Meet Us! | Back to List! |


 


Revealing the Global Linguistic and Geographical Disparities of Public Awareness to Covid-19 Outbreak through Social Media

Presented by: - Binbin Lin



As the Covid-19 pandemic posed an unprecedented challenge to public health, the disparity in awareness of it amongst the residents of the myriad countries led them to suffer uneven health impacts. So, this study seeks to answer what the linguistic and geographical disparities of public awareness in the Covid-19 outbreak period are reflected on social media; and if the changing pandemic awareness predicts the Covid-19 outbreak. The results alongside predicting indexes are presented.


| Learn More! | View the Session! | Meet Us! | Back to List! |



 

Analysis of Linguistic and Geographic Disparities Across Hurricane Related Rescue Request Social Media Data

Presented by: - Bing Zhou



Since the burgeoning of social media platforms, social media data have infused themselves into the geospatial big data that had been widely leveraged among disaster-related research. Capturing the dance of rescue request data can unveil deeper underlying issues because it is representative of the more severe form of disaster responses. Therefore, this research seeks to certify if a universal rescue request classifier exists and to provide proper guidance to social media users on how to draft rescue messages so that the classifier can be reliably applied in future events. Alongside, the relationship between rescue request Twitter activities and the actual damage caused by hurricane events is delineated.


| Learn More! | View the Session! | Meet Us! | Back to List! |


 

A modular machine learning Platform for Resilience Inference Measurement and Enhancement

Presented by: - Debayan Mandal



Disasters have been affecting human life in myriad ways. Just in 2020, the global pandemic showed the vulnerability that humans face. It would be much easier to plan the building up of the community resilience to battle such vulnerability if resilience indexes were readily available to access. Hence, this project is aimed at building an ArcGIS Toolbox using the iRIM model to quantify such resilience accommodating various administrative boundaries. Comprising of a lot of flexibility, this tool will enable planners and researchers alike to mark out vulnerable communities at the county level and come up with proper strategies to counter disasters, increase community resilience.


| Learn More! | View the Session! | Meet Us! | Back to List! |


 

Spatial-Temporal Changes of Flood Risk in the United States

Presented by: - Joynal Abedin



Floods are the most recurrent and common disaster leading to significant fatalities and economic losses in the United States. Flood risk is defined as the product of flood hazard, exposure, and vulnerability. Previous studies have investigated the three elements separately, but estimations of flood risk considering all three factors are rare. This study evaluates both nationwide and county-based changes in flood risk in the US from 2001 to 2021.


| Learn More! | View the Session! | Meet Us! | Back to List! |

 

Simulating land loss and land gain by integrating neighborhood effect and deep learning with cellular automata

Presented by: - Mingzheng Yang



Dynamic land cover changes in coastal zones, e.g., land losses and land gains, severely disrupt regional ecosystem balance and affect coastal human communities in profound ways. There is an urgent need to simulate the coastal land change from a spatial perspective and identify areas prone to land loss or land gain in the future. ANN-CA, generally applied in land-use change simulation, is unable to retrieve the spatial features of the driving factors within a neighborhood and to discover their relationships with the land use type/change of the central cell, leading to lower simulation accuracy. This research offers a more applicable method for researchers and managers to investigate land change patterns in coastal regions.


| Learn More! | View the Session! | Meet Us! | Back to List! |


 

A Community-Based Assessment of Flash-Flood Vulnerability in Texas and the Influence from Different Social Variables

Presented by: - Mouxan Li



As one of the most dangerous natural disasters, flash floods account for 52% of economic losses and over 70% of fatalities and injuries caused by flood-related disasters. There is an urgent need to evaluate community-based flash flood vulnerability, identify its driving factors, and develop mitigation strategies in different communities to reduce damages from future events. This project develops a framework to assess vulnerability at the community level. The results would support further analysis of natural disaster risk assessment and monitoring and assist disaster mitigation and responding.


| Learn More! | View the Session! | Meet Us! | Back to List! |


 

Looking forward to seeing everyone again in AAG 2023, Denver!

GearLab-Logo-White.png

Geospatial Exploration and Resolution (GEAR) Lab