announces winners of its $15,000 Data Science for Good Kaggle Competition
3 min readMay 3, 2019

-- invited the Kaggle community to use their data science expertise to improve the algorithm that matches career questions from youth to working professionals.

Palo Alto, CA — This week announced the winners of its $15,000 Data Science for Good competition on the data science crowdsourcing platform Kaggle. The two-month-long competition was designed to to help the nonprofit use AI to match students to the perfect career advisors efficiently.

First place winner’s Data Overview
Above you can find the diagram with data and relations. Essential primary entities are “Professional”, “Student”, “Question” and “Answer”. The most useful secondary entity is “Tag.”

Primary and secondary entities are used to create so-called “content” features. Other data is divided into two main parts “activity” and “time” features.

“The results were amazing. These are some of the best data scientists in the world, and they submitted ideas using cutting-edge machine learning technologies to solve one of our biggest challenges: connecting the right students to the right volunteers at the right time. The five winning submissions are great examples of recommendation engines, but there’s a ton of depth in the submission field.” said Jared Chung, Executive Director of

The Data Science for Good Competition contained hundreds of kernel submissions, and over 1,300 individual kernel commits, and over 1,000 data analysis charts. Participating data scientists used a variety of statistical methods and machine learning tools to surface insights about recent trends, highlight areas where the matching algorithm could be improved, and generate open code that performs better. “The quality, depth, and documentation of these models and analyses are impressive,” said Jonathan Wang, a member of the Data Science Steering Committee, and one of the judges on the selection committee.

The top five winners hail from Ukraine, Morocco, Belgium, Germany, and the UK, and were selected by a panel of judges. Each winner gets a prize of up to $5,000 from a total prize pool of $15,000: Ivan Didur’s nn-based-recommender-engine in first place, Rod Hyde’s triage-recommender-with-cold-start in second place, Hamza El Bouatmani’s epsilon-greedy-latent-recommender in third place, Daniel Becker’s recommendation-engine in fourth place, and Sobrino Mario’s tag-based-recommender-system in fifth place. plans to extend the submitted ideas for use in its platform to better support its community of 3.5M learners and 30,000 professionals providing career advice. Chung noted: “This is just the beginning. This shows us that data science is an incredibly powerful tool for platforms with large-scale data like We are going to be expanding our data science work over the coming months and hopefully use that to better deliver life-changing career advice to underserved youth everywhere.”

About’s mission is to democratize access to career information and advice for underserved youth. We do this by crowdsourcing personalized career advice to students. Our open-access platform matches student questions to our volunteer corps of over 28,000 professionals, enabling our community of 3.5 million young people to access new information and plan their career path.

About Kaggle

Kaggle provides cutting-edge data science, faster and better than most people ever thought possible. They have a proven track record of solving real-world problems across a diverse array of industries including pharmaceuticals, financial services, energy, information technology, and retail. Kaggle offers public and private data science competitions and on-demand consulting by an elite global talent pool.


Press Release
For Immediate Release

May 3, 2019

Contact: Jordan T. Rivera (



We help underserved youth get the advice they need to get ready for exciting careers. Our volunteers answer questions quickly and accurately!