We believe today’s university students are tomorrow’s leading data scientists. As such, we decided to launch Kaggle University Club — a virtual community and Slack channel for existing data science clubs who want to compete in Kaggle competitions together. As our end-of-year event, we hosted our first-ever University Hackathon!
18 total kernels were submitted and the three top scoring teams won exclusive Kaggle swag and an opportunity to be featured here, on No Free Hunch. Please enjoy this profile from one of the top scoring university teams, ‘Team NDL’ from Penn State!
To read more about the Hackathon and its grading criteria, see here: Winter ‘18 Hackathon and to read this team’s winning kernel, visit: Team NDL: Algorithms and Illnesses.
**MEET THE STUDENTS **
Neil Ashtekar
Major: Computer ScienceHometown: State College, PennsylvaniaAnticipated graduation: Spring 2020
What brought you to data science?
I had read a lot about machine learning/artificial intelligence in the news, and I wanted to see what all the hype was about. So, I decided to complete Andrew Ng’s machine learning class on Coursera. I learned a ton, and I really enjoyed the material. After finishing the class, I wanted to apply what I learned, so I turned to Kaggle. I started out with the basic competitions (Titanic, MNIST), then moved on to work with some more interesting datasets (Kobe Bryant Shot Selection, World Happiness Predictors).
What are your career aspirations after graduation?
I want to get a job as a Machine Learning Engineer (not sure where!).
William Wright
Major: Mathematics
Hometown: Dallas, TexasAnticipated graduation date: Spring 2019
What brought you to data science?**I originally wanted to become a math professor, but after reading Smart People Should Build Things by Andrew Yang and Zero to One by Peter Thiel, I became more interested in pursuing a career involving technology. In his book, Yang claims the decisions we make in the next decade will decide whether society moves towards the future of Mad Max or Star Trek. This comment really stuck with me and inspired me to start learning python and to join Nittany Data Labs (the Penn State data science club).
What are your career aspirations after graduation?
I’d love to work as a data scientist or machine learning engineer.
Izzi Oakes
Major: Integrative Arts
Anticipated graduation date: Fall 2020
What brought you to data science?
I went to my university’s first data science club meeting by random chance, and within five minutes I was hooked. This was about a year ago, and I had never programmed anything before and was in a completely unrelated major. I’ve spent the past year grabbing any and all resources online I could find related to data science and devouring them, as well as moving towards studying higher level math and statistics.
What are your career aspirations after graduation?
I’d like to be in a position where I do work related to some kind of intersection between machine learning and music / visual arts.
TEAM QUESTIONS
How familiar was your team with Kaggle competitions prior to the Hackathon?
A few of us had completed Kaggle competitions in the past, but they were mainly the beginner ones. This was our first time working on a competition as a team, as well as on a longer term project, as this competition lasted about a month.
How did your team work together on **your Kernel**?
We started out working individually to explore and understand the data. After a week of exploration on our own, we met up to talk about our findings and ideas moving forward. At this point, we created a shared kernel and implemented our ideas in code.
What was the most challenging part of the hackathon for you?
Working with text data! None of us had any experience with natural language processing, so understanding how to represent the written review data was challenging.
What surprised you most about the competition?
We were surprised by how well a very simple linear regression model worked with the problem. We had a long conversation about whether we should be using Neural Networks to solve the problem, and potentially why other approaches would work just as well.
What advice would you give another student who wanted to compete in a Kaggle competition or even a hackathon?
If you’re just starting, definitely start with one of the beginner challenges. Try to work your way through it as much as you can by googling things if you get stuck, then begin looking through existing kernels people have once you’re finished. These will give you great approaches to the problem, and you can begin on improving your own model.
Also, try to build your way up to this if you’re just starting. If you really don’t feel like you’re understanding anything you’re doing, there are many great free ML courses and books online!
Anything else?
Thanks a lot for featuring us!
You’re welcome!**
Team NDL from Penn State University (from left to right: Neil Ashtekar, Izzi Oakes, Suraj Dalsania, Will Wright, Ming Ju Li).