skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


Title: Undergraduate Research on Physics-Informed Graph Attention Networks for COVID-19 Prediction
The COVID-19 pandemic has significantly impacted most countries in the world. Analyzing COVID-19 data from these countries together is a prominent challenge. Under the sponsorship of NSF REU, this paper describes our experience with a ten-week project that aims to guide an REU scholar to develop a physics-guided graph attention network to predict the global COVID- 19 Pandemics. We mainly presented the preparation, implementation, and dissemination of the addressed project. The COVID-19 situation in a country could be dramatically different from that of others, which suggests that COVID-19 pandemic data are generated based on different mechanisms, making COVID-19 data in different countries follow different probability distributions. Learning more than one hundred underlying probability distributions for countries in the world from large scale COVID- 19 data is beyond a single machine learning model. To address this challenge, we proposed two team-learning frameworks for predicting the COVID-19 pandemic trends: peer learning and layered ensemble learning framework. This addressed framework assigns an adaptive physics-guided graph attention network (GAT) to each learning agent. All the learning agents are fabricated in a hierarchical architecture, which enables agents to collaborate with each other in peer-to-peer and cross-layer way. This layered architecture shares the burden of large-scale data processing on machine learning models of all units. Experiments are run to verify the effectiveness of our approaches. The results indicate the proposed ensemble outperforms baseline methods. Besides being documented on GitHub, this work has resulted in two journal papers.  more » « less
Award ID(s):
1924278
PAR ID:
10469680
Author(s) / Creator(s):
;
Publisher / Repository:
The International Institute of Informatics and Cybernetics
Date Published:
Journal Name:
Journal of Systemics, Cybernetics and Informatics
Edition / Version:
1
Volume:
20
Issue:
5
ISSN:
1690-4524
Page Range / eLocation ID:
148 to 159
Subject(s) / Keyword(s):
Layered Ensemble Learning Physics-guided Learning COVID-19 Graph Attention Network
Format(s):
Medium: X Other: xls
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; ; (, International Journal of Environmental Research and Public Health)
    The ongoing highly contagious coronavirus disease 2019 (COVID-19) pandemic, which started in Wuhan, China, in December 2019, has now become a global public health problem. Using publicly available data from the COVID-19 data repository of Our World in Data, we aimed to investigate the influences of spatial socio-economic vulnerabilities and neighbourliness on the COVID-19 burden in African countries. We analyzed the first wave (January–September 2020) and second wave (October 2020 to May 2021) of the COVID-19 pandemic using spatial statistics regression models. As of 31 May 2021, there was a total of 4,748,948 confirmed COVID-19 cases, with an average, median, and range per country of 101,041, 26,963, and 2191 to 1,665,617, respectively. We found that COVID-19 prevalence in an Africa country was highly dependent on those of neighbouring Africa countries as well as its economic wealth, transparency, and proportion of the population aged 65 or older (p-value < 0.05). Our finding regarding the high COVID-19 burden in countries with better transparency and higher economic wealth is surprising and counterintuitive. We believe this is a reflection on the differences in COVID-19 testing capacity, which is mostly higher in more developed countries, or data modification by less transparent governments. Country-wide integrated COVID suppression strategies such as limiting human mobility from more urbanized to less urbanized countries, as well as an understanding of a county’s social-economic characteristics, could prepare a country to promptly and effectively respond to future outbreaks of highly contagious viral infections such as COVID-19. 
    more » « less
  2. ; (, International Supply Chain Technology Journal)
    Due to the COVID-19 pandemic, a number of employees transitioned into virtually distributed supply chain project team environments for the first time in their careers due to shelter-in-place and social distancing mandates that impacted industries around the world. With project managers implementing processes and procedures remotely to employees fielding calls from home while caring for children, the world entered into an unprecedented time and an unknown unknown. As the pandemic required organizations to implement agile practices, project managers within various industries had to adjust operational processes, while at the same time had to ensure project team needs were addressed. In an effort to mitigate the risk of transitioning back to conventional management approaches, a qualitative survey method was performed in this study to understand supply chain project team member experiences during this transition period. The results of the survey uncovered that the creation of a supply chain uncertainty management plan, demonstrating the benefits of virtual supply chain project teams, and the continuous integration of collaborative technology tools may contribute positive outcomes from the COVID-19 pandemic. 
    more » « less
  3. ; ; (, Emerging Adulthood)
    Using a longitudinal design, we examined changes in parent and peer attachment among college students, and their predictive and protective roles in relation to mental health problems during the COVID-19 pandemic. College students ( N = 106) rated their parent and peer attachment, and self-reported anxiety, loneliness, and depressive symptoms 1 year before and during the pandemic. Participants also rated the impact of COVID-19 related stressors (CRS). Results demonstrate significant increases in loneliness and depression and a decrease in peer attachment security during the pandemic. Increases in peer attachment security were negatively correlated with loneliness during the pandemic. Parent attachment buffered the relationship between the impact of CRS and mental health problems during the pandemic. Guided by the integrated framework from attachment theory and life course theory, the current study discussed findings and practices regarding the important role of attachment for college students and their adjustment to the pandemic. 
    more » « less
  4. ; ; (, Proceedings of the Association for Information Science and Technology)
    Abstract As the impact of the COVID‐19 pandemic grew in 2020, uncertainty surrounding its origins and nature led to widespread conspiracy‐related theories (CRT). Use of technological platforms enabled the rapid and exponential dissemination of COVID‐19 CRT. This study applies social contagion theory to examine how Google Autocomplete (GA) propagates and perpetuates these CRT. An in‐house software program, Autocomplete Search Logging Tool (ASLT) captured a snapshot of GA COVID‐19 related searches early in the pandemic (from March to May 2020) across 76 randomly‐selected countries to gain insight into search behaviors around the world. Analysis identified 15 keywords relating to COVID‐19 CRT predictions and demonstrate how searches across different countries received varying degrees of GA predictions. When grouped with similar keywords, two major categories were identified “Man‐Made Biological Weapon” (42%, n = 2,111), and “Questioning Reality/Severity of COVID‐19” (44%, n = 2,224). This investigation is also among the first to apply social contagion theory to autocomplete applications and can be used in future research to explain and perhaps mitigate the spread of CRT. 
    more » « less
  5. ; ; ; ; ; ; ; ; ; ; et al (, Data and Information Management)
    null (Ed.)
    Abstract The COVID-19 outbreak is a global pandemic declared by the World Health Organization, with rapidly increasing cases in most countries. A wide range of research is urgently needed for understanding the COVID-19 pandemic, such as transmissibility, geographic spreading, risk factors for infections, and economic impacts. Reliable data archive and sharing are essential to jump-start innovative research to combat COVID-19. This research is a collaborative and innovative effort in building such an archive, including the collection of various data resources relevant to COVID-19 research, such as daily cases, social media, population mobility, health facilities, climate, socioeconomic data, research articles, policy and regulation, and global news. Due to the heterogeneity between data sources, our effort also includes processing and integrating different datasets based on GIS (Geographic Information System) base maps to make them relatable and comparable. To keep the data files permanent, we published all open data to the Harvard Dataverse ( https://dataverse.harvard.edu/dataverse/2019ncov ), an online data management and sharing platform with a permanent Digital Object Identifier number for each dataset. Finally, preliminary studies are conducted based on the shared COVID-19 datasets and revealed different spatial transmission patterns among mainland China, Italy, and the United States. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email