Article Complete
AIM
Meet The MachineHack Champion Who Cracked The ‘Cardiac Arrest Risk Prediction’ Hackathon
Meet The MachineHack Champion Who Cracked The ‘Cardiac Arrest Risk Prediction’ Hackathon

Meet The MachineHack Champion Who Cracked The ‘Cardiac Arrest Risk Prediction’ Hackathon

Published Updated

Key takeaways

  • The AI conference plugin showcased global talent and facilitated engaging discussions and workshops.
  • Karan Juneja, a passionate data scientist, excelled in the cardiac arrest risk prediction hackathon.
  • His approach involved straightforward data analysis and selecting optimal features for model stability.
  • Karan experimented with various models, including XGBoost and Random Forest, to enhance performance.
5 min

Last week, one of the biggest virtual AI conferences -- plugin -- was hosted by Analytics India Magazine. It brought in brilliant minds from around the world to the ones who sought, all in a well-integrated virtual setting unmatched by any.

Organised by Analytics India Magazine in collaboration with ADaSci, Altair, Analyttica Datalab, SAS and Kyvos Insights, the event was filled with highly informational talks and workshops. Apart from the talks, plugin also hosted a hackathon conducted by MachineHack solely for the Data Science enthusiasts looking for some fun.

In this article, we present to you the champion of the hackathon and his approach to the problem that drew the spotlight.

Karan Juneja

Karan is an Electronics and Telecommunication Engineer from PICT, Pune. His data science journey began out of his passion and curiosity for robotics. He has been acquiring new data science skills from free online resources as well as by participating in hackathons. 

Karan is a regular participant in MachineHack hackathons and has been in top 3 for several of the competitions.

Approach To Solving The Problem 

“The provided data was simple and straightforward, and from the initial EDA, I found out there wasn't much to do with preprocessing and model training. The most important part was just to make the model as stable as possible by selecting the best features,” said Karan.

“I also tried fitting different models such as XGBoost, Random Forest and Logistic Regression. The trick to getting the best score was to remove as much noise from the data as one possibly could,” he added.

Get the complete code here.

Check out for new hackathons here.