According to a report on March 10, according to the BBC, there is a huge loophole in the company's digital defense system against cyber attacks. This loophole is the lack of skilled technicians around the world, and they play a vital role in ensuring that the secure hardware runs, analyzes threats, and counters intruders. Industry research institute ISC2 in the security field shows that the global security industry currently lacks one million trained technical personnel. And in the next five years, this gap will reach 1.8 million people.
Figure: Intelligent machines are helping companies defend cyberhackers
According to Ian Glover, head of Crest, the British ethical hacking technology certification body, the shortage of technicians has been widely recognized, which has also caused many other problems. Grove said: "The shortage of talent is driving costs up. There is no doubt that this will have a huge impact, because companies are trying to get scarce resources. In addition, this may also mean that companies cannot find competent people because they are always Look for people to fill the vacancies."
Grove warned that many countries are taking measures to attract people to the security industry, but these efforts are not enough to make up for the shortage of technicians. He said: "If you see an increase in the degree of automation of attack tools, then you also need to increase the degree of automation of defense tools."
In the data "Shuishui"
Peter Woollacott, founder and CEO of Huntsman Security, a security company headquartered in Sydney, Australia, said that we are moving toward higher automation, but this change may seem a little late. For a long time, innovation in the security field seems to be carried out in a "manual" slow manner.
Analysts believe that these defense companies are "inundated" in the data. These data are generated by firewalls, PCs, intrusion detection systems, and other devices, and they cause many problems. Oliver Tavakoli, chief technology officer at security company Vectra Networks, said that automation is not new and was used long ago to help anti-virus software discover emerging malicious programs.
Figure: In a complex network, humans are not always able to find abnormal activity
But now, machine learning is helping automation go further. Taracley said: "Compared with artificial intelligence, machine learning is easier to understand and simpler." But this does not mean that it can only handle simple problems. The analytical power of machine learning comes from the evolution of algorithms that can process large amounts of data and pick out abnormal or major trends from them. Increasing computing power also makes this possible.
These "deep learning" algorithms are different. Algorithms such as OpenAI are open to everyone. But most of the algorithms are owned by the companies that develop them. Larger security companies are scrambling to acquire smaller but more intelligent start-ups in order to quickly enhance their defense capabilities.
Not so smart
Simon McCalla, chief technology officer of UK-based domain registrar Nominet, said machine learning has been proven to be very useful in tools he developed called Turing. Turing can look for evidence of cyberattacks from a large number of queries the company handles on a daily basis, including information on the location of UK websites.
McCorra said that Turing had helped analyze what happened during the cyber attack on Lloyds Bank in January when thousands of customers could not use the bank’s services. Distributed Denial of Service (DDoS) attacks generate a large amount of data that needs to be processed. McLaura said: "Usually, we need to process 50,000 queries per second. In the Lloyds Banking Incident, we need to process more than 500,000 queries per second."
Figure: Stop spammers by allowing machine learning to analyze data traffic
Once the attack is over and the dust settles, Nominet can handle the equivalent of 1 day of traffic in 2 hours. Turing absorbs all the information from the Nominet server and then uses what he learned to issue early warnings, warning people to be prepared for longer-lasting attacks. It records the IP address of the hijacked machine that issued the query and checks if its email address is valid.
McCallula said: "In most cases, the defense system we see is not that smart." Without the help of machine learning, human analysts are unlikely to find an attack until the target is affected, such as the sudden collapse of a bank's website. Analysts believe that Turing can now help the British government maintain internal network security, such as preventing staff from accessing suspicious websites and avoid becoming a victim of malware.
Chaos and Order
There are also more ambitious plans to use machine learning analytics. At the 2016 Def Con Hacking Conference, Darpa, a military research organization under the Pentagon, organized a competition to allow seven smart computer programs to attack each other to see which program has the strongest self-defense capability. The winner is Mayhem, which is now being reworked so that bugs in the code can be discovered and fixed to prevent exploitation by malicious hackers.
Figure: Mayhem wins Darpa competition to find smart computers that can find bugs
Taracley said that machine learning can correlate data from different sources and provide analysts with comprehensive information to determine if a series of continual events will pose a threat. It can usually understand the flow of data in the company and its low tide, as well as the staff's daily schedule. Therefore, when a cyber hack attempts to peek into a network connection or enter a database, this malicious behavior is immediately discovered.
Of course, in a huge network, hackers are also very good at covering their tracks. For humans, these "compromise signs" may be difficult to find. Now, as long as security analysts sit quietly, they can let the machine learning system crunch all the data and find signs of serious attacks that haven't caught human attention. Tarlacley said: “This is like a doctor who is preparing for an operation. Although the patient is not yet in place, they are ready and are in a state of calm.â€
SMD Electric Power tansformer ,POE transformer ,EP7 SMD tansformer ,EP13SMD tansformer，EPD SMD tansformer
IHUA INDUSTRIES CO.,LTD. , https://www.ihua-sensor.com