We're an independent, student-run newsroom.

As an independent student newspaper and the paper of record for the city of Berkeley, the Daily Cal has been communicating important updates during this pandemic. Your support is essential to maintaining this coverage.

donate here
  • With a massive amount of data online, computer programmers have recently taken to looking for ways to manage it all.

    Marti Hearst, a professor in the UC Berkeley School of Information, has been at the forefront of research on working with “Big Data” to make it searchable and accessible. Her 2009 book, “Search User Interfaces,” was the first formal investigation of the different ways computer users search through data and how the process can be improved.

    After completing her undergraduate and graduate work at UC Berkeley, Hearst has spent more than 15 years teaching in the School of Information, with interludes spent working at PARC, Xerox’s research and development company and with the federal government.

    Hearst sat down with The Daily Californian to discuss how Big Data at large as well as on campus has evolved over the years.

    Q: How did you become involved with computer science?
    A: I was not really a computer-type as a kid. I was more of a literary kid. I liked science but not really math. When I went to college, I wanted to study linguistics and biology and psychology. I became interested in artificial intelligence, so I decided to major in computer science. When it came to graduate school, I was still really interested in language — natural language processing — so I started thinking about language databases. No one was thinking about searching through that much data yet.

    Q: What was your experience like studying computer science at UC Berkeley?
    A: The computer science degree at Berkeley was hard then, and it’s still hard today. It certainly was for me because I didn’t have the background. It was hard to get in the classes; you had to work on shared computers late at night in Cory Hall because there weren’t (personal computers), so it was physically hard.

    Q: How has your academic work affected the research community?
    A: Everyone was trying to understand language the way people do. I did very early work on statistical trends in text, and then the Web happened. I was interested in search before the Web. Through my work at PARC, I became interested in human-computer interfaces for search, which no one was doing at the time, and I wrote the first book on the topic a few years ago.

    Q: How is the School of Information changing as new technology trends emerge?
    A: The I-School is a very interdisciplinary program. We’re training people to be the information professionals of the future. So I don’t know if our students need to change, but our program needs to continue to change. We’ve had a strong emphasis on human-computer interaction and information visualization, and the legal aspect, cyberlaw, has always been a big strength in our school. Now there’s a new emphasis on Big Data, and we’re starting to offer a lot more classes on it.

    Q: What are your students currently working on?
    A: We have about 15 lecturers from Twitter coming into one class to work with students on Big Data projects. The students will make presentations on campus and at the Twitter headquarters (in San Francisco), so it’s very exciting. They’re a very cool company, and they’re doing it entirely for free. They’re giving back to the programmer community.


    2 thoughts on “Five Questions: Marti Hearst, ‘Big Data’ pioneer

    Leave a Reply