2018 will be remembered as the year of data breaches. With well-established and multi-billion dollar companies like Facebook and Marriott falling victims to cybersecurity and data breaches, the concern for data security has grown significantly. With the improvement of technology and its availability to a large number of consumers, the use of technology remains unmonitored. In an attempt to curtail data breaches Microsoft with its new competition on Kaggle, is challenging the Data Science community to use Artificial Intelligence to prevent cyber security attacks.The competition is organized by the Windows Defender ATP Research team, in collaboration with Northeastern University and Georgia Institute of Technology as academic partners.
Artificial Intelligence is already helping in detecting and stopping a malware at first sight.With this competition, Microsoft takes it a step further. The idea is to stop a malware before it is even seen. The objective is to use predictive analysis to predict if a machine will soon be hit with a malware.The competition presents a fresh opportunity for data science enthusiasts across the world to showcase their skills on a real world problem more real than any other dataset you may possibly find. The company intends to use the results from the competition to further improve their defences. The challenge is to identify the systems with higher risk of getting affected so that preemptive measures can be taken.
Here’s What You Should Know About The Data
Participants will build their predictive models using 9.4GB of data collected from over 16.8 million devices.The dataset has over 82 features or columns of data. The contestants have to build a machine learning model to uniquely identify a machine that is affected by a malware.Each machine is identified by a unique MachineIdentifier which is a column in the provided data set.The column or feature HasDetections is the dependent factor the contestants have to predict for each machine in the test set..
Important Dates and Prizes
The competition was launched in Kaggle on December 13 2018 and has already gained the attention of almost 700 data scientists.
- Entry deadline : March 6, 2019
- Team Merger deadline (participants can join or merge teams) : March 6, 2019
- Final submission deadline : March 13, 2019 .
Microsoft presents the winners with an overall $25000 cash prizes:
- 1st Place – $12,000
- 2nd Place – $7,000
- 3rd Place – $3,000
- 4th Place – $2,000
- 5th Place – $1,000
This is not the first time Microsoft has challenged the Data Science community. In 2015, Microsoft hosted a similar malware classification competition which was a great success in terms of helping Microsoft as well as a large community of data scientists. The datasets provided by Microsoft are used in a number of researches and continues to produce value for Microsoft and the data science community.