By GREGORY ZELLER //
Millions of dollars in new federal grants are flowing to Stony Brook University’s Department of Computer Science, with cyber-defense and mobile data playing large.
The research couldn’t be more disparate: digital security via cryptographic codes and better understanding of alcoholism through mobile data.
But the nine-digit awards – one from the Defense Advanced Research Projects Agency, the other from the National Institutes of Health – both support ongoing efforts inside the busy and diverse Department of Computer Science, part of SBU’s College of Engineering and Applied Sciences.
The DARPA grant awards $1 million to Assistant Professor Omkant Pandey, who is investigating new cryptographic code methods to better secure data transmissions and otherwise enhance corporate computer security (banks are especially interested).
Pandey, an affiliate of SBU’s National Security Institute, shares the four-year grant with Assistant Professor Sanjam Garg of the University of California-Berkeley. The duo hopes to create more efficient “zero-knowledge proofs,” perchance to streamline digital verification procedures.
“Professor Pandey’s research … will advance the security of future computer systems more efficiently and significantly contribute to the nation’s cyber-defense,” noted SBU Computer Science Department Chairman Samir Das. “[He] is a key member of our highly visible cybersecurity group.”
The $2.5 million NIH grant supports a cross-disciplinary study incorporating “temporal methods” and artificial intelligence to determine if social media posts and mobile phone data can predict and quantify excessive alcohol consumption.
The project, led by Computer Science Department Assistant Professor H. Andrew Schwartz, will analyze social posts and mobile data (including texts) and build AI models that predict heavy alcohol use – defined by the National Institute on Alcohol Abuse and Alcoholism as more than four drinks on one day for men, more than three for women – and its effects on an individual.
The effort is part of Schwartz’s larger focus on better healthcare through AI. In this case, the predictor will be developed with a focus on frontline restaurant workers, including bartenders and servers, who have “among the highest rates of heavy alcohol use of all professions,” according to SBU.
The team will leverage the $2.5 million grant to create “temporal methods” that assess the association between unhealthy drinking and empathy among restaurant industry workers based on their social media language.
They’ll also create a mobile app good for “longitudinal tracking of drinking behavior and psychological health” – key to building predictive models – and develop methods to assess how community factors influence unhealthy drinking, “utilizing AI-based representations of the communities in which restaurant industry workers live and work,” according to SBU.
Schwartz, also a faculty member of SBU’s Institute for AI-Driven Discovery and Innovation, said the data collected by the study would advance existing sciences that focus on language as a doorway to the mind, providing a real-time treasure trove for the mental-health profession and a giant leap toward individualized treatment plans.
“We now know analysis of everyday language can cover a wide array of daily factors affecting individual health, but their use over timespans is limited,” Schwartz said. “The methods we will develop in this project should enable real-time study into how health plays out in each individuals’ own words – and bring about the possibility for personalized mental healthcare.”