November 20, 2018

E FlemingSenior Eleanor Fleming talks about her path from humanities to data science and how data science opens up opportunities in a variety of fields. As an intern for the Division of Data Sciences, Eleanor made videos featuring data science people and programs. She will graduate this spring as a data science major with a domain emphasis in cognition.  

This interview has been edited for clarity and length. 

How did you first get interested in data science?

I first started doing things in the Division when I applied with URAP (the Undergraduate Research Apprentice Program), because I wanted to see how data related to cognitive science and computer science. I was originally thinking of being a cognitive science major with a computer science minor. I heard the Division was putting together a team to explore some sort of data science major, and I thought, “This sounds really cool!” So, I was on that team originally, doing marketing work to try to get the major off the ground, and that’s how I got involved with making videos for the Division.

Taking a step back, how did you get interested in cognitive science and computer science?

I went to a boarding school where I loved my French and history/English classes. My school was pretty humanities-focused, so when I first started at Cal, I hadn’t really been exposed to the tech world. I decided to try CS61A, which was really hard but I really liked it. While I found 61a and 61b super interesting, I felt like I didn’t want pursue straight computer sciences as a career, so I took CogSci 1 because I heard that it was a cool class, and I knew that there was a computational side to cognition.

What got you interested in cognition?

When I first started taking cognition/neuroscience classes, I thought that I’d just be learning things like what is memory? and how does the brain work? And while that was a big basis for my classes, there was also a fascinating theoretical aspect, like what does it mean to be conscious? and Can artificial intelligence truly replicate humans? Now, I’m in a cognitive neuroscience class, and I’ve also taken a computational cognitive science class. Both delved more into specific brain imaging techniques and how cognition can be represented through logic and neural networks.

With data science, if you have an understanding of the baseline, you can explore many different avenues without ever feeling like you’re veering off-course. 

Where does the data science major fit into all of this?

I just wanted to try something different, and I thought that it would be nice to join something that started from the ground up. I learned video skills and also got the opportunity to talk to all these amazing students and professors who were using data science in their respective fields. That was kind of my aha moment: With data science, if you have an understanding of the baseline, you can explore many different avenues without ever feeling like you’re veering off-course. When people come to Cal, it can feel like such a big and impersonal school. I was definitely stressed out by not knowing what I wanted to pursue right off the bat. Giving people the ability to have a foot in a couple of different things they’re interested in is really special for such a big school.

The major gives me the tools to be able to change disciplines; if you have the tools to analyze data, that makes you valuable whichever industry you end up in.

What industry are you headed for?

I’m going to be working at Blackrock, an asset management firm, starting this summer. I think data and technology is revolutionizing the finance sphere, and I find fintech super interesting. They have the Aladdin platform, which combines risk analytics with portfolio management, trading, and operations tools on a single platform. I will be learning the ins and outs of the platform and look forward to working with their data.

Was there anything particularly surprising or revealing to you over the course of your data science studies?

In my Computational Models of Cognition class, we learned a lot about Turing machines, semantic networks and probabilistic models of language. Thinking about cognition and language through the lens of these models was interesting because we got to see both the power of certain systems and where they fall short of the biological reality. Mapping the brain and its functions is still a work in progress, so approaching it from both the biological and computational side is really cool–essentially trying to figure out how we can fill in the gaps.

Are there any particular ethical aspects of data science that are important to you?

Data is really what you take from it, and it shouldn’t always be taken as fact. In my data ethics class, we talked a bit about the word itself.  “Data” from dare in Latin means to give vs. “capta” from capere means to take. Data implies something raw organic and natural–meaning it is “given,” when really it’s selectively taken. As students and scientists, we “take” meaning from the data. Everything you extract should be taken with a grain of salt, as there are different processes that produced this data in the first place.

Say a bit about your current project at Berkeley.

I’m currently in machine learning (ML) at Berkeley and working on a project to train an autonomous drone to avoid obstacles within a Unity simulation. I’m pretty new to the ML space and it is a great opportunity to learn more about reinforcement learning and how data collection is applied to solving a problem that industry faces in the real world.

Just because something is technical doesn’t mean it has to be ridiculously hard.

Anything else?

I think that students, especially like me who did not have a technical background entering Cal, need to realize that just because something is technical doesn’t mean it has to be ridiculously hard. Sure, anything you haven't practiced or seen before is going to seem harder, but there’s a stigma at Cal that you have to have a certain threshold of knowledge when you come in to be successful. I mean, we do this to ourselves, but also there’s a lot of pressure being at such a large, technically-strong school with grade cut-offs for majors, etc. You don’t have to be the best at it; you just have to be willing to try it and enjoy the exploration along the way, and you can still be successful.

News

February 21, 2022

Kelechi Nnebedum fell in love with science and technology at Los Medanos College in Pittsburg, Calif. But when she transferred to UC Berkeley to be a data science major and walked into classes with only two other girls and no other Black students, she felt intimidated.

February 14, 2022

Marti Hearst is UC Berkeley School of Information’s new head of school. She will manage the unit’s day-to-day operations and communicate its vision on and off campus. She aims to build on the work of the I School’s last head of school, Hany Farid; launch the unit’s first undergraduate minor; and increase resources for the school.

January 31, 2022

Meet Ashish Sahni, the new executive director of the Berkeley Institute for Data Science. Sahni joined the UC Berkeley research and training organization – and Division of Computing, Data Science, and Society affiliate – this month from UC San Francisco.

January 27, 2022

Deb Nolan, emeritus professor of statistics at UC Berkeley and former associate dean for undergraduate education at the Division of Computing, Data Science, and Society (CDSS), has returned to CDSS as Associate Dean for Faculty for the spring 2022 semester. 

January 26, 2022

The American Association for the Advancement of Science (AAAS) elected the Division of Computing, Data Science, and Society’s (CDSS) top communicator and three affiliate faculty members to be AAAS fellows, one of the scientific community’s highest honors, the group announced today.