Intrusion Detection Data Mining Thesis

Intrusion Detection Data Mining Thesis-40
Anomaly detection techniques demonstrate good accuracy in detecting network-level attacks such as the SYN flood, teardrop, and denial of service (DOS), etc.; but not in recognizing application level exploits such as Remote to Local (R2L) and User-to-Root (U2R).All of anomaly detection schemes consider only the packet header fields such as flags, port numbers, and IP addresses, etc.; therefore, they work well if an attack involves only the related fields at the network level.Intrusion prevention systems can be classified into two categories: network-based intrusion prevention systems and host-based intrusion prevention systems [4].

Tags: Ethical Dilemma In Health Care EssaysBusiness Floor PlansHow To Write A Research Paper PdfThesis Style FilesWhy We Need Critical ThinkingPrawn Farming Business PlanWriting Medical Papers A Practical GuideOthello Research Paper Topics

However, if the attack pattern is slightly altered, this method will not be able to identify the changed versions of the attack [7].

An anomaly detection system has a profile of normal behaviour patterns about the defence system.

Current challenges of these methods in intrusion detection are also introduced. In the area of cyber security, effective and efficient situational awareness often requires knowledge of current and historical cyber (i.e.

Cite this paper: Lidong Wang , Randy Jones , Big Data Analytics for Network Intrusion Detection: A Survey, International Journal of Networks and Communications, Vol. host or network) activities to detect and respond to threatening behaviours [2].

After a new attack is launched, the attack pattern or signature is defined which can be targeted resources during an attack, the way that the resources are targeted in, or a name (in characters) within the body of the attack code.

Network security specialists can design a defence against a new assault after the attack signature is studied.

Intrusion detection systems (IDSs) can be categorized into three types: a network-based intrusion detection system (NIDS), a host-based intrusion detection system (HIDS), and a hybrid-based intrusion detection system (hybrid IDS).

An HIDS detects malicious activities on a single computer while an NIDS identifies intrusions by monitoring multiple hosts and examining network traffic.

Stream data mining involves dynamic changes and efficient discovery of general patterns within the stream data.

People are interested in identifying intrusions based on the anomaly of message flow that can be discovered by dynamically constructing stream models and clustering stream data, or comparing the current frequent patterns with those at specific previous times.


Comments Intrusion Detection Data Mining Thesis

The Latest from ©