How did you choose this program?
I went to UCSC for my undergraduate degree before starting the NLP program. I was a computer science and computational mathematics major, and I graduated this year. I had an idea that I wanted to do grad school afterwards, and one day I got an email about the program. To tell you the truth, I didn’t even know what NLP stood for at the time!
I was intrigued by the description and started looking into what it was and why there was a separate program for it, and reached out to the program director, Professor Marilyn Walker.
I went into the NLP program fairly blind. I hadn’t taken any courses as an undergraduate, but the math and computer science I took at UC Santa Cruz has really helped me prepare for the program. I think of it as a wide enough platform to give me the footing I need for this program. I’m about a month in so far, and it’s super fun.
We’ve just begun so we’ve mostly been doing introductions and review; but there have been a couple of seminars which have been fantastic. It’s amazing hearing from people who are working on NLP now -- for example the professor teaching my five o’clock class is teaching at five because he’s working with the technology during the rest of the day. I feel lucky to have these people teaching us.
It’s an approachable subject. It’s easy to get started with NLP, and there are tons of resources available. I also like that this is such a small program, so all 25 of us are taking the same classes together and even though we’re on Zoom 100% of the time, it feels like we’re all connected, so I’m getting to know people.
I’ve really appreciated the emphasis the school has had on keeping our cameras on during class and making sure we all speak up. It helps maintain a sense of community.
Tell me more about why you like NLP.
Natural language processing is still very new to me. It’s been fascinating. I love learning about linguistics and languages and how languages create meaning and the different properties languages have. The application of the technology has also been really cool. We’re already working on projects that are producing results.
I’ll give you an example: The first project we made was a label classifier. It takes an utterance (something someone would say) and breaks it down. For example if you asked it, “Who plays Luke in Star Wars: A New Hope?” it would break the question down into categories (such as movie.starring.actor). We have a database of about 3,000 utterances and a test that we don’t have the relations for and that we have to predict the correct answer. We’re all trying to create the most accurate predictions... It’s incredible what you can do with this technology. I’d love to build things like this for a living.
What are some fun facts about yourself?
I’m an avid cyclist. I was on the cycling team at UCSC. In my NLP introduction, my fun fact was that I biked from Santa Cruz to San Francisco and back and I just did that again. It took all day. It was 16 hours or something.
I also work part-time as a bicycle mechanic. I’m obsessed with bikes. I’ve thought about combining NLP and biking into some kind of app, but so far I haven’t been able to think of anything. Bike technology tends to run about ten years behind the rest of the world. I’ve tried using a voice assistant while I ride and it was one of the most frustrating things I’ve ever done. It’s like 2001: A Space Odyssey when the computer betrays everyone!