21.11.2017 Views

SJ113 Master 2

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

5 Ways Machine Learning<br />

Can Be Used for Security<br />

During 2016, we had a record high of 1,093 breaches, according to<br />

the Identify Threat Resource Center. Don’t let your business become<br />

a statistic. Machine learning is being applied to five new areas with<br />

hopes that this capability can catch threats before they result in<br />

major security incidents.<br />

Do you enjoy being<br />

the new Grandmaster?<br />

Learning<br />

security moves.<br />

When a hacker’s system which is<br />

capable of machine learning goes up<br />

against an enterprise system, there’s<br />

a battle between security know-how.<br />

And, when data centre security is<br />

also learning defensive moves, there<br />

can be a chaotic and rapid exchange<br />

of strategic moves. That is, until<br />

someone wins.<br />

Briefly, machine learning is a series<br />

of techniques used to “train” a<br />

machine how to solve a problem.<br />

Although, there’s been a lot of “new”<br />

hype around machine learning, it’s<br />

currently being used now in several<br />

industries. Common applications<br />

of machine learning in technology<br />

include voice recognition, fraud<br />

detection, email spam filtering, search<br />

recommendations, and video analysis.<br />

So, chances are you’ve already seen<br />

machine learning in action.<br />

You’re most likely already using<br />

machine learning as part of your<br />

data centre security protection.<br />

Today’s defense solutions are<br />

optimised to prevent targeted attacks,<br />

advanced threats, and malware<br />

from embedding or spreading<br />

within a data centre or network. For<br />

example, at Zones we offer Trend<br />

Micro’s solution that can analyse<br />

unknown threats with machine<br />

learning algorithms to make real-time<br />

decisions on whether network traffic<br />

is malicious or benign.<br />

Zones also provides solutions that<br />

protect enterprises from attack<br />

methods that no one has ever<br />

seen before. We help you apply<br />

a unique combination of behavior<br />

analytics, data science techniques,<br />

and machine learning algorithms.<br />

This baseline, “normal” network,<br />

and endpoint behavior, helps identify<br />

attack indicators and minimise<br />

false positives.<br />

These examples of cross-generational<br />

techniques to defend networks<br />

against known, undisclosed, and<br />

unknown threats, demonstrate the<br />

value of today’s security solutions.<br />

Winning the<br />

defense game.<br />

With machine learning in play,<br />

security becomes the only game<br />

to win. Many advanced threats<br />

are designed specifically to bypass<br />

traditional security defenses. Today’s<br />

sophisticated threats require new<br />

ways to protect everyone and<br />

everything – users, networks, data,<br />

customers, organisations, and the list<br />

goes on.<br />

While no one can say that a single<br />

security solution is foolproof, machine<br />

learning can certainly help you<br />

go from vulnerable to victorious<br />

in one fell swoop. Today’s security<br />

solutions use machine learning –<br />

behavioral analysis, intrusion and<br />

exploit prevention, investigation<br />

and forensics, memory inspection,<br />

prevalence/maturity checks, and<br />

reputational intelligence – to protect<br />

against threats. In some cases,<br />

this occurs even months before a<br />

vulnerability can be exploited by a<br />

threat. Seeing ahead is a significant<br />

strength in the security game.<br />

Instead of tossing security solutions<br />

around without rhyme or reason<br />

like pieces on a chessboard, let<br />

Zones help you figure out your next<br />

move. For starters, it’s easy to see<br />

every piece that’s undefended. Once<br />

you know your undefended pieces,<br />

you can make a jump in security<br />

strengths. Zones Security Assessment<br />

services can help you make that jump.<br />

Whether it’s security assessments,<br />

solutions, or both, we’re ready<br />

to help you be the winner of<br />

the security match, and help you<br />

turn “check” into “checkmate.”<br />

1. Closing IoT vulnerabilities: You<br />

can leverage machine learning<br />

technology to analyse data and<br />

network interactions, determining<br />

safe device behavior. With this<br />

level of insight into general use<br />

patterns, it becomes easier to<br />

identify abnormal activity and<br />

block harmful actions.<br />

2. Preventing false alerts: When<br />

machine learning identifies<br />

something as suspicious, you<br />

can provide feedback by tagging<br />

legitimate threats. The solution<br />

then adjusts its monitoring filters,<br />

boosting its ability to find and<br />

mitigate real breaches before<br />

they cause damage.<br />

3. Stopping exploits: Criminal<br />

hackers often share ideas and<br />

information on forums and<br />

websites that aren’t indexed by<br />

search engines. Machine learning<br />

can be used to monitor traffic<br />

within these areas to identify<br />

data relating to zero-day exploits.<br />

Then, you can use this data to<br />

close vulnerabilities and stop<br />

patch exploits before they result<br />

in a data breach.<br />

4. Predicting threats: Cybersecurity<br />

solutions can easily integrate<br />

Source: Trend Micro blog, 16/3/17. http://blog.<br />

trendmicro.com/5-ways-machine-learning-canbe-used-for-security-today/<br />

with machine learning to detect<br />

network changes over time<br />

and revise behavioral profiles.<br />

This helps predict threats based<br />

on historical data and limit<br />

the number of false positives<br />

that appear.<br />

5. Providing holistic protection:<br />

Upgrade to the latest security<br />

solutions that combine proven<br />

threat detection techniques,<br />

advanced protection methods,<br />

and high-fidelity machine<br />

learning. This combination<br />

ensures that you’ll have a holistic,<br />

effective enterprise defense.<br />

A layered security<br />

infrastructure will protect<br />

your company against<br />

98 percent of the known<br />

threats, but there’s<br />

always the possibility of<br />

compromise through an<br />

unknown vulnerability.”<br />

22 Solutions by Zones Autumn 2017<br />

ukzones.com +44 (0) 207 608 7676 23

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!