Predictive Analytics for Cyber Security

By Avirag Jain, EVP & CTO, R Systems International


Avirag Jain, EVP & CTO, R Systems International

In the era of interconnected / digital world, most of our time is spent online being either on corporate networks or on world wide web doing office work, reading articles, news, education, doing online shopping, communicating with people.  Huge amount of data is gathered and transferred over networks by businesses, governments, banks, educational institutes and many more for their day-to-day operations. Data may include confidential, private and sensitive information and may have severe business / personnel impact if got into wrong hands.

Cyber Crimes increase with Cyberspace expansion

Majority of people are not even aware of tactics being deployed by cyber criminals and those who are aware; never take it seriously hence often become prey to intelligence of cyber criminals.  Cyberspace is expanding and so the techniques & sophistication of cyber criminals. They are embracing technology much faster than folks responsible for making systems secure. Mobile devices like smartphones, wearables & tabs etc. and Internet of things (IoT) devices are becoming beloved for cyber criminals due to increased penetration and availability of unsecured data on them.   Connected world with everything online is providing a superlative platform to cyber criminals and has opened up new avenues to create a bigger industry for malware-preneurs and has started offering CAAS i.e. cybercrime-as-a-service.

With commercialization of cybercrime via CAAS, it is becoming relatively easy to become a cybercriminal, becoming more organized.

Some of the most common cyber-crimes are:

• Data breach

• Identify theft

• Transaction / payment card fraud

• Cyber intrusion

• Cyber Stalking

• Hacking

Off late, there have been incidents of cyber-intrusions at U.S. Office of Personnel Management (OPM). In two separate incidents, intrusion “may have compromised the personal information of [approximately 4.2 million] current and former Federal employees" and  "compromised its databases housing background investigation records and resulted in the theft of sensitive information of 21.5 million individuals".

It is pertinent to ensure data privacy and protection in our interest but how? Organizations need to have appropriate policies & infrastructure for controlling infrastructure, networks and devices. Individuals are required to be more aware of cyber.. But is it sufficient?

Do we have a foolproof solution or can there be a foolproof solution? The straight answer is NO!!!

There cannot be one solution that will completely eliminate all issues related to cybercrime and the only way to circumvent is to bring awareness, constant monitoring and adaptability.

It is not practical to bring everyone at the same level in terms of awareness and are we not monitoring & adapting to the situations as an industry?

Yes, we are but due to such a large variety of touch points, it practically becomes impossible to track, monitor and analyze each attribute with traditional methodologies and is becoming very difficult to reduce the number of false positives.

Then what is the solution?

Gone those days where traditional technologies were not that sophisticated and were not able detect such attacks. Based upon recent technology stack, it is very much feasible to have one package solution that integrates various components that:

• Process peta bytes & zetta bytes of structured and / or unstructured real-time data in short span of time

• Constantly discover and analyze traffic for unusual and / or recurring patterns

• Evolve / Define / Apply KPIs constantly to cope up with ever changing patterns

• Derive meaningful real-time insights, trends and forensics based upon behavioral and predictive analytics

• Provide interactive visualizations and dashboards on variety of devices including handheld devices

• Integrate with other operational systems that will take feed from analytics for automated / manual preventive actions

An intelligent predictive analytic system will not only provide a platform that will safeguard enterprises across tiers based upon existing patterns but will also provision adaptability to include new ones as well.

Current Issue