/novartis-hack-challenge

This was a small dataset where we had to predict if a cyberattack will happen or not for a customer. Main focus is on EDA and Data Visualisation. This challenge was sponspored by HackerEarth.

Primary LanguageJupyter Notebook

Novartis-hack-challenge

alt text

The Insights & Analytics team in Novartis works on 20+ brands catering to 50+ disease areas, using the latest techniques in data science. Our Insights & Analytics specialists are on a data and digital transformation journey, leveraging analytics to generate actionable insights for Novartis medicines impacting the lives of 10% of the world’s population. The team is poised to enable easier, faster, and reliable decisions for Novartis divisions across the globe.

The team works like a start-up, is agile and adopts an innovation-driven culture, experimenting with new technologies in building data and digital solutions to serve patients. The team is looking for data scientists and data engineers to be a part of a fast-growing group of inspired, curious Novartis associates who apply their digital expertise in creative ways.

This is not a Novartis website. Information on this site is solely for purposes of the hiring challenge and is hosted on the HackerEarth website.

Know more about the Hiring challenges by - https://www.instagram.com/hackerearth/

I used the LightGBM algorithm to get a recall score of 99.7. Screenshot is attached on the repo.

LightGBM is a gradient boosting framework that uses tree based learning algorithms. It is designed to be distributed and efficient with the following advantages:

Faster training speed and higher efficiency.

Lower memory usage.

Better accuracy.

Support of parallel and GPU learning.

Capable of handling large-scale data.