Imagine your favorite apps had a "digital twin" of your personality that actually grew up with you. Right now, most AI systems create a static snapshot of your interests. For example, a personal shopper who keeps recommending video games just because you bought one three years ago, even though you've long since moved on to hiking and cooking. To bridge this gap, Professor Dong Wang's team at the University of Illinois Urbana-Champaign is partnering with Snap Research to build a "Profile Agent."
Instead of writing a one-time bio of your likes and dislikes, this smart system constantly refines its notes to make sure it actually reflects who you are today, rather than who you were last year. The secret sauce is a "drift detection" system that acts like a real-time pulse check on your interests, said Wang.
"If you suddenly swap your late-night movie habit for early-morning workout routines, for instance, the Agent notices that its old suggestions aren't hitting the mark anymore," he said. "Rather than staying stuck in the past, it triggers an immediate update to your profile to capture these new 'interest trajectories.'"
According to Wang, the goal is to create a digital experience that feels less like a cold algorithm and more like a helpful assistant that can explain exactly why it's recommending something, making your time on social media or shopping apps feel intuitive and genuinely personal.
Wang's research includes social sensing, intelligence and computing, human-centered AI, and big data analytics. His work has been applied in a wide range of real-world applications such as social network analysis, crowdsourcing, disaster response, education, smart cities, synthetic biology, and environmental sustainability. He holds a PhD in computer science from the University of Illinois Urbana-Champaign.