Dr. Jianhua Chen, On the Power of Watson's Mind
After getting her Ph.D. in computer science from Jilin University in Changchun, China, Jianhua Chen joined the faculty of the Computer Science Department at LSU in 1988. Moving from northeast China to Baton Rouge took a lot of adaptation, she says, but she’s become comfortable living here and now teaches several undergraduate and graduate courses, including data structures and algorithms, database systems, artificial intelligence, machine learning and data mining.
We talked with Chen recently about her research and its real-world applications.
ARTIFICIAL INTELLIGENCE AND BIG DATA
Chen’s research interests fall in the broad area of artificial intelligence and involve understanding the nature of intelligence and building intelligent systems as smart as humans. Specifically, she works on machine learning and data mining, which focus on developing algorithms that allow computer programs to learn patterns and knowledge useful for more intelligent decision-making by analyzing huge data sets.
“I was attracted to the study of AI since I was an undergraduate student,” she says. “My professor’s seminar on artificial intelligence opened my eyes to this exciting field. It’s so fascinating to build a smart computer system capable of learning new knowledge on its own and competing at the level of human experts in tasks such as playing a chess game.”
Chen says she’s also interested in scaling up data-mining algorithms. “In this big data era, we often encounter data sets that are really huge. The sheer size of the data renders traditional learning algorithms unfeasible and we have to build scalable algorithms that can easily handle big data.” She’s developed new methods of adaptive sampling for learning an ensemble of classifiers from a sampled subset of data, which are much more efficient and thus more scalable.
TAPPING INTO WATSON'S MIND
Chen offered a course last fall titled “Artificial Intelligence and Cognitive Computing.” Students learned several major topics in AI, including machine learning, natural language processing and heuristic search for problem-solving, which form the foundations of AI techniques used in IBM’s Watson, the famous cognitive computing system that appeared on Jeopardy.
As part of the course, students worked on the theories of cognitive computing and got hands-on experience working with Watson. They formed several project groups to build a “powered by Watson” mobile app using the cognitive computing concepts learned in class. “The application uses Watson’s powerful natural language processing capability and performs query-answering with users via natural language dialogue,” she says.
Chen says she focuses on courses that teach skills and techniques that are highly desirable in IT. Her database course trains students extensively on database programming and design, while the AI and cognitive computing course with Watson exposes them to real-life applications. “This experience gives the students a competitive edge in seeking jobs that require knowledge of AI and Watson, as well as in pursuing graduate studies with a concentration in AI.”
Chen says she hopes her teaching contributes to a solid foundation of computer science knowledge that students can use to advance to the next level of their careers. “I would be thrilled if some of the students get inspiration from me and become researchers in artificial intelligence and machine learning, just like I got inspired by my professor.”