Vendors Betting Big on Machine Learning in Cybersecurity: Can it Deliver?


Amazon Web Services recently launched its security service called Macie which uses machine learning to automate the process of identifying, classifying and protecting data in AWs. It uses the context of the information to understand what’s sensitive and how. Thus, it recognizes sensitive data such as personally identifiable information and intellectual property and monitors access to the data.

 

The service draws on the technology from Harvest.ai, an AI startup that AWS acquired earlier in the year. This just goes on to re-affirm the growing clout of machine learning within cybersecurity.

 

Feelers around this trend are already rife in the market with security vendors investing on sprucing up their portfolio with machine learning capabilities either through acquisitions or building them in-hose. Notable among the acquisitions is Sophos spending $ 100 million to acquire Invincia, an endpoint security software company which uses machine learning and deep learning neural-network algorithms to detect unknown malware without the use of signatures.

 

Meanwhile, Radware has placed its bet on Seculert, a cloud based cyber security technology company. Seculert’s acquisition will give Radware access to heightened machine learning technology and big data analytics tools to conduct advanced threat analysis, read a report.

 

And, the list goes on. Then, there are other vendors who are building the capabilities in-house.

 

The reason for machine learning becoming a key component of the newer cybersecurity offerings is the capabilities it brings to the table. Here’s taking a look at some of those capabilities.

 

-       Businesses today generate far too much data for any human to comprehend and process successfully. Machine learning can handle huge data sets and scale as the data volumes grow. Considering that the fuel for running machine learning is data, the higher the data volumes, the better the system gets.

-       Ability to analyze and discern patterns in network activities at a scale and speed impossible for humans to match.

-       Accelerate incident detection and incident response. On an average it takes an organization several months to detect a breach. Machine learning can help bring this down to minutes and hours.

-       Alert on true anomalies and eliminate ‘false positives’ among the several hundreds and thousands security alerts received in a day.


(Image Courtesy: Pixabay.com)

Categories: Technology

About Author

Write a Comment

Your e-mail address will not be published.
Required fields are marked*

*

Recent Comments