Skip to main content
WiDS Posts | July 12, 2024

WiDS Datathon++ University Edition 2024 Winners

This year, we launched the WiDS Datathon++ University Edition, engaging 70+ universities in 23 countries including universities in underserved communities. Sponsored by Gilead Sciences, it focuses on equity in healthcare and is used by instructors across all levels and fields of study. Students gained real-world experience by working on the challenge; some students continuing to pursue further research.

Congratulations to all participating students on your outstanding performance! Meet the winning teams:

1st place: TUCN_AI_ErrorExorcists
Andreea-Maria Onaci & Petrut-Betuel Paul

How do you see the datathon impacting your university studies going forward?

This datathon has had a significant impact on our university studies. It provided us with practical, hands-on experience in applying theoretical concepts, enhancing our understanding of machine learning and data analysis. The skills we developed in model selection, hyperparameter tuning, and data preprocessing will be directly applicable to our course work and lab work. Additionally, the collaborative nature of the datathon improved our teamwork and problem-solving abilities. Moving forward, this experience will help us tackle academic challenges more effectively and prepare us for our future jobs.

What was your team’s biggest learning or insight participating in the challenge?

Participating in the challenge gave our team valuable insights into various regressors, hyperparameters, and scalers. We learned their strengths, weaknesses, and the importance of selecting the right combinations. This knowledge will help us make informed decisions when building and tuning models.

The experience has enhanced our technical skills and emphasized the importance of continuous learning. These insights will not only improve our future projects but also equip us with practical skills that are applicable in real-life scenarios, allowing us to solve complex problems more effectively.

2nd place: NMBENGALURU_ANALYTICS
Vadiraj G Moktali and Aditi Kulkarni

What motivated your team to join the datathon?

WiDS gave us an invaluable platform to not only contribute to a noble cause but also to deepen our understanding of the transformative power of data science. Through this experience, we gained insight into how effectively data science can be harnessed to address and solve pressing real-world problems. It was an eye-opening opportunity that highlighted the significant impact data-driven solutions can have on society.

What was your team’s biggest learning or insight participating in the challenge?

The learning curve has been steep and cannot be encapsulated in a few words. In our pursuit of improving our models and climbing the leaderboard, we learned numerous lessons, some of which proved successful while others did not. Our biggest takeaway was the importance of constant reinvention and the willingness to try new approaches.

3rd place: ASE_Data Mining_Fantasticians
Ana-Maria Oprea, Bianca Contolencu, Gianina-Maria Petrașcu, and Ioana Birlan

How do you see the datathon impacting your university studies going forward?

The datathon has already impacted our university studies in the sense that we have been more creative and innovative in our data science solutions. All of us are interested in research and we appreciate the fact that this experience has provided us with practical insights and inspired us to approach our studies with a more applied, hands-on perspective. It has also fueled our passion for continuous learning and experimentation in the field of data science.

What was your team’s biggest learning or insight participating in the challenge?

The biggest learning from the datathon was the importance of teamwork and diverse perspectives in tackling complex data problems. We realized that combining our unique strengths and insights led to more creative and effective solutions than working individually. Additionally, we recognized the need to balance the pursuit of achieving the best score with verifying the stability of our algorithms, while also learning to appreciate the critical role of data cleaning. Our greatest challenge was finding the most effective methods to clean the dataset while ensuring the relevance of variables remained intact.