Deep Learning in Cyber-security

The existing cybersecurity methods are signature-based and mostly rely on the predefined behavior of cyber threats. They fail when cyber threats change their behavior. They also fail when there is a large amount of data. But in deep-learning technology works like the human brain, they process the data and can easily predict the malicious activities which can be missed by the existing methods. Moreover, existing methods just generate alarms when threats are identified. then it requires human experts to analyze the situation and take action. Sometimes, humans might miss vulnerabilities. Many of the times, the traditional methods generate false alarms. It makes the process tedious for the security team. On the other hand, cyber-security methods do not generate false alarms that are more accurate and are capable of taking decisions on their own. NAME DRASHTI RATHOD (11677786) PROJECT TITLE DEEP LEARNING IN CYBER-SECURITY WEEK ...

Deep learning is a subset of machine learning. Deep -learning is part of artificial intelligence. In this technique, the programs mimic the function of the human neurons to process the information and making decisions. the human brain makes decisions with a lot of experiences. In a similar way, deep-learning programs need a large amount of data to train them. Once they have enough data experience, this technique will be able to work accurately in real-time work which has a large amount of data. It is also capable of making its own decision. The deep -learning algorithms will analyze the problem and comes with the most effective solutions. This technique is best in today's era. NAME DRASHTI RATHOD PROJECT TITLE DEEP LEARNING IN CYBER-SECURITY WEEK NUMBER 3 DATE 18/03/2020 PLANNING                 ...

Today, everyone is dependent on the internet. We use our personal information in numerous ways. it is important to protect this information. There are various existing methods that help to protect the system from cyber threats. These techniques are used to predict and prevent cyber-threats. But in the real world data, these techniques have their limitations. The traditional methods of cybersecurity rely on signatures. This means the methods will only detect malicious activities if they follow predefined behavior. If the behavior of there's changes it will not be identified and the system would not be protected. it is also difficult for the existing methods to work in a large amount of real-time data. Moreover, the exiting system depends upon human experts. this often leads to missing vulnerabilities.

INTRODUCTION TO DEEP LEARNING IN CYBER-SECURITY Cyber-security is a very crucial issue in today's fast-growing digital world. Almost everything we do rely on the internet. There are numerous attacks that have been very common threats in a cyber society. Some of these are ransomware, DOS (Denial-of-service) attack, malware, viruses, etc. The existing methods used for cyber-security are largely signature oriented and it becomes difficult to identify and stop these attacks if they change their behaviors. I order to prevent these attacks it is also essential to detect them on a real-time basis. Due to such reasons identify malicious malware is becoming a challenging task. To solve this problem effectively, we use deep learning technology to identify attacks on real data. Deep learning is an AI (Artificial intelligence) methodology. It is a subset of machine learning. Deep learning makes the use of Artificial Neural Networks (ANNs). It has been proposed to use the me...