Skip to main content

Recent and Ongoing Projects

  • Zero-day Exploitation Detection and Mitigation
    • While there have been major advancements in the speed and performance of AI-based intrusion detection systems, the response is still mostly at human speed when it comes to explaining and interpreting a specific prediction or decision. The delay in detection/response increases the chance of zero-day (0-day) exploitation, whereby a previously unknown vulnerability is discovered by an attacker and exploited before the vendor can fix it. We are exploring the knowledge-infusion-based approach to detect and mitigate zero exploitation.
  • Illegal Financial Activity Detection
    • We are doing big data analytics on financial data and developing novel AI-based explainable techniques to detect and explain illegal financial activities in the U.S. market.
  • Driving Under the Influence (DUI)
    • We are investigating whether the elimination of certain societal biases can reduce DUI-related death.
  • Social Network Mining for Identifying Potential Shooters
    • We are developing a multi-modal machine learning system to identify potential school shooters.