A review of intrusion detection using anomaly based detection

Kumari, Uma; Soni, Uma

doi:10.1109/cesys.2017.8321199

Cited by 12 publications

(2 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Important trainings of data features can decline IDS performance and level of accuracy. The results show that selection of features will substantially increase IDS' efficiency even if it improves the inaccuracy of the function [9]. The analysis of selecting features on the NSL-KDD dataset is performed that does not consider attacks of multiple types [10].…”

Section: Salih Et Almentioning

confidence: 99%

A Comprehensive Analysis on Numerous Learning Models for Intrusion Detection for Security Conservation

Sri¹,

Peddireddy²,

Madamanchi³

et al. 2022

RIA

View full text Add to dashboard Cite

The term intrusion refers to a series of behaviours that exposes computer networks and systems' security to compromises. Corrective action on the network cannot go on without intrusion detection. IDS and IDS is the framework used to detect network traffic intrusions, which is how the network control mechanism identifies potential intrusions. Security breaches are designed to undermine one or more of the network's three primary security goals: privacy, availability, and trust. To get access to a system, an attacker must follow a predetermined set of procedures. Once inside, they can begin gathering data such as the protocol being used and the network resources available. There are many ways for a hacker to find out what systems are available on the network and how vulnerable they are to attacks. The rapid advancement of network technology necessitated IDS to focus on the detection of assaults using contextual analysis from signature matching processes. Using machine learning to detect and prevent intrusions, the IDS is a critical part of protecting data systems. Network intrusion detection is the focus of this paper, which examines and shows various machine learning techniques.

show abstract

Section: Salih Et Almentioning

confidence: 99%

A Comprehensive Analysis on Numerous Learning Models for Intrusion Detection for Security Conservation

Sri¹,

Peddireddy²,

Madamanchi³

et al. 2022

RIA

View full text Add to dashboard Cite

show abstract

“…AIDS [2] uses statistical models trained on existing intrusion detection data to analyze and classify incoming traffic. With the development of machine learning (ML) and deep learning (DL) algorithms and high availability of data, AIDS shows great promise.…”

Section: Introductionmentioning

confidence: 99%

Temporal Weighted Averaging for Asynchronous Federated Intrusion Detection Systems

Agrawal

Chowdhuri

Sarkar

et al. 2021

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Federated learning (FL) is an emerging subdomain of machine learning (ML) in a distributed and heterogeneous setup. It provides efficient training architecture, sufficient data, and privacy-preserving communication for boosting the performance and feasibility of ML algorithms. In this environment, the resultant global model produced by averaging various trained client models is vital. During each round of FL, model parameters are transferred from each client device to the server while the server waits for all models before it can average them. In a realistic scenario, waiting for all clients to communicate their model parameters, where client models are trained on low-power Internet of Things (IoT) devices, can result in a deadlock. In this paper, a novel temporal model averaging algorithm is proposed for asynchronous federated learning (AFL). Our approach uses a dynamic expectation function that computes the number of client models expected in each round and a weighted averaging algorithm for continuous modification of the global model. This ensures that the federated architecture is not stuck in a deadlock all the while increasing the throughput of the server and clients. To implicate the importance of asynchronicity in cybersecurity, the proposed algorithm is tested using NSL-KDD intrusion detection system datasets. The performance accuracy of the global model is about 99.5% on the dataset, outperforming traditional FL models in anomaly detection. In terms of asynchronicity, we get an increased throughput of almost 10.17% for every 30 timesteps.

show abstract

Towards Data Science for Cybersecurity: Machine Learning Advances as Glowing Perspective

Mihăilescu

Nita

2022

Lecture Notes in Networks and Systems

View full text Add to dashboard Cite

A review of intrusion detection using anomaly based detection

Cited by 12 publications

References 9 publications

A Comprehensive Analysis on Numerous Learning Models for Intrusion Detection for Security Conservation

A Comprehensive Analysis on Numerous Learning Models for Intrusion Detection for Security Conservation

Temporal Weighted Averaging for Asynchronous Federated Intrusion Detection Systems

Towards Data Science for Cybersecurity: Machine Learning Advances as Glowing Perspective

Contact Info

Product

Resources

About