April 24, SMU Professors Fred Chang and Jo Guldi, team up to tell Texas that SMU has joined the fight against COVID-19 by deploying the resources of a newly created Artificial Intelligence Lab. Chang’s expertise in science/engineering and Guldi’s in text mining are shining examples of the interdisciplinary initiatives that elevate SMU research efforts. Published in the Dallas Morning News: https://bit.ly/353XC2A
What if university computer scientists, biologists and historians collaborated to use modern artificial intelligence and machine learning to examine a massive trove of infectious disease research papers, text mining for abstract patterns, elusive insights and hard-to-spot trends related to COVID-19 and the coronavirus family of viruses?
Imagine the energy such a group could generate if their students, working remotely and cut off from the normal distractions of student life, jumped in to volunteer for the project? Welcome to the nascent Southern Methodist University Artificial Intelligence Lab. . .
By Fred Chang and Jo Guldi
What if university computer scientists, biologists and historians collaborated to use modern artificial intelligence and machine learning to examine a massive trove of infectious disease research papers, text mining for abstract patterns, elusive insights and hard-to-spot trends related to COVID-19 and the coronavirus family of viruses?
Imagine the energy such a group could generate if their students, working remotely and cut off from the normal distractions of student life, jumped in to volunteer for the project? Welcome to the nascent Southern Methodist University Artificial Intelligence Lab.
COVID-19 has been called the greatest challenge since World War II. Artificial intelligence and machine learning technologies were still young during other recent outbreaks — SARS in 2002, H1N1 in 2009, MERS in 2012 and even Ebola in 2014. Increasingly more powerful processors, better algorithms and massive amounts of data have changed what we can do in 2020.
Our charge at the time of this pandemic is to deploy everything we now know about AI to discover as much as possible about what we do not know about COVID-19. The larger challenge, however, is to shape a university response that effectively trains students to grapple with a rapidly-changing and destabilized world.
More than a dozen SMU faculty members and students are volunteering their time to text-mine a large collection of scientific papers made available via the White House Office of Science and Technology Policy and a collection of research groups. These works are packaged as a challenge on the data science site Kaggle. We are using SMU’s supercomputer to yield insights from these papers that we hope will aid active infectious disease researchers in their search for a solution. As we discover trends, patterns and insights, it is our hope that our AI and machine learning research can assist in the goal to shorten the time required to discover and develop a vaccine or therapy.
Teams are meeting virtually, outside of hours scheduled for classes, working from apartments or family homes. By matching the COVID-19 research papers with teams of students directed by faculty, we are making use of two major resources universities have to contribute at this very unusual moment in history: brainpower and the gift of time for creative reflection.
Many of our students have hours on their hands after finishing their virtual classwork, and they are eager to serve the public good. They are telling us that they want to be able to look back on this time and remember that they were part of the fight. There is no more valuable gift that we could give our students than preparing them to face an uncertain world.
While we have placed our near-term priority on assisting in the search for a vaccine or therapy, there are many facets to a global pandemic that can and should be addressed. Here are a few ideas that we are contemplating in our lab:
- Mining social media with a list of symptoms and body parts to generate a profile of when and where symptoms appear.
- Tracking the spread of fake news stories over social media and developing an experimental bot that could refer users to official information.
- Tracking other potential medical conditions that are affected by COVID-19, for instance, postponed chemotherapy.
- Creating an emergency text-mining service to assist in future pandemics to perform real-time analysis to enhance situational awareness.
Tasks of this kind are inherently interdisciplinary. Executing them in an effective manner requires creative work at the juncture between the disciplines. Students can only build these domains if they have knowledge of the technical skills they learn in computer science, as well as textual analysis skills they learn from the social sciences and humanities. Responding to coronavirus thus requires our faculty to think creatively about teaching, for instance, in the form of interdisciplinary labs such as the one we have recently launched.
Even while health-related restrictions are forcing students and teachers to limit their interactions to online spaces, SMU’s faculty is experimenting with new ways to connect, bringing the immediacy and liveliness of an engaged, intellectual community to meet the challenge of the pandemic. Even more importantly, we are modeling for our students what it looks like to respond in real time to real world challenges, regrouping and refocusing our research on the pandemic, and inviting them to make discoveries.
At the end of a disrupted semester, these SMU students will have had an educational experience that is enhanced, rather than diminished, at least in terms of opportunities for research.
Frederick R. Chang is a cyber security professor, chair of the Department of Computer Science at Southern Methodist University and the founding director of the SMU AI Lab.
Jo Guldi is an associate professor of history at SMU, where she teaches text mining and is a founding member of the SMU AI Lab.
They wrote this column for The Dallas Morning News.