Deep Learning in Cyber-security

Proposed methodology there are many deep-learning methodologies that can promote cyber-security. One such methodology is use of the neural networks. This network copies the functioning of the human brain to work. the system worked in similar way to process the information as the human brain works. the neural network has features such as image recognition language processing that can be used to protect the information systems of the organizations. it helps to prevent the various cyberattacks. It helps to develop smarter prevention techniques and also provide automation in cyber-security. NAME DRASHTI RATHOD (11677786) PROJECT TITLE DEEP LEARNING IN CYBER-SECURITY WEEK NUMBER 11 DATE 27/5/2020 PLANNING                             MILESTONE PLANNED ACTUAL COMMENT To conduct experimental analysis to...

Issue 5: Denial-of-service attacks The denial-service attackers are continuously increasing. they affect the normal functioning of the organization. in this type of attack, the attackers flood the network and servers of the organization with the false requests and make them reach their limitation resulting in server becoming unavailable. Deep-learning methods make the use of ANN that shows promising results. It analyzes the HTTP traffic of the network to detect any malicious activities. It helps to prevent DOS attacks. Moreover, familiar techniques can also prevent SQL injection attacks. NAME DRASHTI RATHOD (11677786) PROJECT TITLE DEEP LEARNING IN CYBER-SECURITY WEEK NUMBER 10 DATE 24/5/2020 PLANNING                             MILESTON...

Issue 4: Spam detection Spam attacks have become more and more popular. the attackers sent the spam emails to thousands of random users by laying a trap and which help them get hold of the confidential information of the users. If any of the users fall in this trap they are further blackmailed by the hackers to get money. Spam detection has become very difficult and the existing system sare reaching their limitations. The Natural language processing concept which is based on the deep-learning algorithms can be used to detect and deal with spam. the NLP learns the form of communication by learning the language patterns and use the statistical model to  detect and block the spam NAME DRASHTI RATHOD (11677786) PROJECT TITLE DEEP LEARNING IN CYBER-SECURITY WEEK NUMBER 9 DATE 17/5/2020 PLANNING               ...

Issue -3 Intrusion detection All the organization is prone to intrusion detection. It has become very difficult to protect the system from the intrusion attacks. there are existing methods that provide intrusion detection. These traditional use ML algorithms.these algorithms produces many false alarms. This, in turn, increases the load on the security teams. deep-learning is the best solution for this issue. Convolutional neural networks and recurrent neural networks of deep-learning technology can be sued to create smarter intrusion detection systems. These help to analyze the traffic with more accuracy. These systems detect malicious network activities and prevent intruders from accessing the systems and alerts the user. It will help to reduce false alarms and reduce the burden of the security teams. NAME DRASHTI RATHOD (11677786) PROJECT TITLE DEEP LEARNING IN CYBER-SECURITY WEEK NUMBER 8 DATE 10/...

Issue 2: Malware attacks Malware can be defined as the attack that can be caused by hackers by creating malicious software that harms the system that gets installed without the knowledge of the users.it is intended to steal the confidential information  and causing damage to the system. The existing methods use the firewalls to detect the malware. These attacks are detected by signature-based detection systems. but the nature of the threats keeps changing and the system can be affected by new threats. Deep-learning technology provides the algorithms that have the potential of identifying advanced threats and do not rely on known signature and pattern of the attack. the recognize the suspicious activity by learning the system.  NAME DRASHTI RATHOD (11677786) PROJECT TITLE DEEP LEARNING IN CYBER-SECURITY WEEK NUMBER 7 DATE 3/5/2020 PLANNING     ...

Issue 1: Vulnerability detection In recent times it has become very difficult to identify the vulnerability of the system. The existing methods when detecting malicious activity in the network generate the alarms. Then it requires human interaction to define the issue and take action. The human tends to miss these weaknesses and this can, in turn, lead to several consequences. Various researches are done to show that the deep-learning methods can solve the vulnerability issue in the best possible way. it reduces the number of false alarms and provides automation. NAME DRASHTI RATHOD (11677786) PROJECT TITLE DEEP LEARNING IN CYBER-SECURITY WEEK NUMBER 6 DATE 26/4/2020 PLANNING                             MILESTONE PLANNED ...

Deep-learning is an emerging technology and can be used in almost all fields today. It can be used in healthcare, cyber-security, car automation, and almost all sectors. But it has several applications of deep-learning in cyber-security. It can be used as an intrusion detection system for conducting network traffic analysis. It helps to provide the best accuracy than the other methods. It also helps to reduce the number of false alarms. It can be used to detect malware attacks. the existing system fails to identify the attacks when the behavior of the attack changes. deep-learning can identify such changing behavior as it does not work on a pre-defined pattern of attacks. It can be used in the detection of spam and social engineering. Deep-learning natural processing language technology can significantly help to prevent spams. NAME DRASHTI RATHOD (11677786) PROJECT TITLE DEEP LEARNING IN CYBER-SECURITY WEEK NUMBER 5...