Detection Ratio Of Cyber Attack Detection

Cyber-security demands are ever increasing in the field of Information Technology with the globalization of the internet. Disruptions due to cyber-attacks are affecting the economy, costing companies billions of dollars each year in lost revenue. To counter this problem

Cyber-attacks intruding network systems is a threat the economy, national and public safety. Computer system pirates are the possible persons who can form groups that can attack the nation and cause the negative impact to the country's attack. More research should be done in the networking area to monitor systems performance and detect any malice attacks as soon as possible with correct advanced measures taken to solve the problems.

In recent times to keep up the network security is a foremost and the network is hacked by the unofficial persons. There are various strategies to extend the safety similar to encryption and firewall. However these strategies are failed to detect the intrusions. For that a new technology is Intrusion detection system. The Intrusion detection is the problem of identifying unauthorized use, misuse and abuse of computer systems. Outside attackers are not only the problem, the threat of authorized users misusing and abusing their privileges is an equally pressing concern. The intrusion detection system used data mining strategies for the network safety, as a result of to guard the network from numerous assaults and malicious site visitors

of unique attack sources per day. To take a closer look at this phenomenon, we present

1)After the brief overview of our research in building predictive models for learning from rare classes, the paper gives a comparative study of several anomaly detection schemes for identifying novel network intrusions.

Most important security challenge on the Internet is the existence of the large number of compromised machines. Project main aim is to detect the compromised machines

Introduction Cyber threats are currently of large enough significance to enough to constitute a full blown national security priority in the West and the United States. In fact, we may be at zero hour. In order to better understand the challenges that cyber infrastructures are facing, it is necessary to understand the level of the threat that is posed, where the potential threats may emanate from, the techniques which help in identifying emerging cyber security threats posed by those who perpetrate computer-based crimes and attacks and the level of protection required. While computer hacking has traditionally been looked upon as a criminal enterprise, the high-tech nature of today's warfare, especially in warfare in and the command of the "high ground" of space, cyber warfare has become a critical battle that must be won for the West. While the Internet may be a "super-highway" for information and commerce, like any physical super-highway, it can also be a quick entry into the heartland of a nation. In this case, that is its cyber heartland.

The root of the attack tree is the goal of the attack, like in case of household security the root will be the valuable asset. The branches of the tree will represent different method for achieving the goal, like in case of household security, branches will be breaking the window of house and creating the duplicate key. With this branching representation different ways to attack the value asset , it is possible to evaluate the threats by assigning values of difficulties to various threats. It helps to determine the weak point in the system. The attack trees model can help expert to set priorities to the attack for improving the security. The attack trees are the graphical representation with the goal of the attacker at the root of the attack trees and the possible attacks techniques from attacker represent the leaf or subgoal below the root of attack tree.

To perform adaptive cyber-security analytics using a computer implemented method that includes receiving a report on a network activity, It works as follows, a scoring method is used wherein a score responsive to the network activity is calculated. The score indicates a likelihood of security breach. The score is validated and the scoring model is automatically updated responsive to results of the validating. If the score falls within a threshold value of a security violation rule, the network activity is reported as suspicious. This basically relates generally to cyber-security monitoring, and more particularly to monitoring incoming security events to determine if any activity is violating

Almost all kind of large and small organizations might face increasing number of attacks into their network or intellectual property. This may lead to data disclosure, data destruction, and damage of organization’s reputation. There are numerous threats in the cyber space which might be capable of stealing, destroying or making use of out sensitive data for financial and non-financial gains. As the amount of computer, mobile and internet users increases, so does the number of exploiters.

Over the past decade computer networks have rapid growth, but the computer security became a critical problem for computer systems. Thus in the recent years various soft computing technique based methods were proposed to detects the growth of intrusion. Also many researchers have reported that the large set of pattern classifications and machine learning algorithms are trained and testing is made on the knowledge discovery data intrusion detection dataset is unsuccessful in finding the remote-to-local attacks and user-to root attacks. Moreover Hyperbolic Hopfield Neural Network(HHNN) based Intrusion Detection Systems (IDSs) detection stability, detection ratio, particularly low-frequent attacks are still required to be improved. Thus this paper proposes a new method known as K-Medoids-HNNN using the technique HHNN and K-Medoids clustering. This system achieves the higher intrusion detection rate, detection stability and less false positive rate. At first the proposed system implements the K-Medoids clustering technique on the various training subsets. Afterwards a mono HHNN model is trained using the different training subsets to detect the intrusion. The experimental results shows the K-Medoids-HHNN approach achieve better results rather than other framework.

According to the method of detection , IDS can be divided into: Anomaly and Misuse based detection. After the construction of the normal pattern the data is classified into attacked data if it doesn’t belong to normal data. Misused detection is used to detect the attacks in different ways. Anomaly based detection often has high false alarm rate due to the changeful attack patterns. Its accuracy rate is low as compared to the Misuse based detection .The accuracy rate of misuse based detection is high because the

This technique follows the same notion of how human brain works. It consists of collection of highly interconnected processing elements that convert a set of given inputs to a set of desired outputs. The widely used neural network for intrusion detection is Multilayer Perceptions (MLP).Neural Network based intrusion detection systems are intended to classify the normal and intrusion patterns and the type of the attack. For this purpose, the network is trained on various types of attacks and normal data. One input node is present for each element of the feature vector, and one output node for each class, to which a feature may be assigned, as shown in fig.4. The nodes in the hidden layer are connected to the input

In the 21st century, cyber-attacks happen on a daily basis, to be more accurate the attacks are happening every second of our lives. Many of these attacks have done little harm to our cyber network, but many have and will result in a major disaster. The cyberspace became popular to our society no more than 20 years ago. Before then, computers and the internet were all foreign ideas to most people. While the advance of computing technology and the internet aided our civilization in almost every aspect of life, it has also become an avenue for criminals to exercise their deception and malicious ways on other internet users.

Cyberspace which is an upshot for the evolution of computer and digital communication technologies, has in recent decades become part and parcel of our daily lives. It has become an ideal source for cyber criminals to remain active while preying on victims. Higher the number of cyberspace users’, higher is the opportunities for exploitation. So it is the need of the hour to protect our computers, networks, digital applications and our data from unintended or unauthorized source, change or destruction.