Deep Learning Approaches for Intrusion Detection

Salih, Azar Abid; Ameen, Siddeeq Y.; Zeebaree, Subhi R. M.; Sadeeq, Mohammed A. M.; Kak, Shakir Fattah; Omar, Naaman; Ibrahim, Ibrahim Mahmood; Yasin, Hajar Maseeh; Rashid, Zryan Najat; Ageed, Zainab Salih

doi:10.9734/ajrcos/2021/v9i430229

Cited by 19 publications

(6 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These models classify data transfers based on the characteristics of various attacks. In addition, the authors gave an overview of intrusion detection system datasets [10]. Network Security Laboratory Knowledge Discovery in Databases (NSL-KDD), and Community Integrated Cyber Intelligence Data Sharing (CICIDS2017) are public datasets with the features of cyberattacks represented by various IDS datasets.…”

Section: Literature Reviewmentioning

confidence: 99%

A Novel Eccentric Intrusion Detection Model Based on Recurrent Neural Networks with Leveraging LSTM

Muthunambu,

Prabakaran,

Kavin

et al. 2024

CMC

View full text Add to dashboard Cite

The extensive utilization of the Internet in everyday life can be attributed to the substantial accessibility of online services and the growing significance of the data transmitted via the Internet. Regrettably, this development has expanded the potential targets that hackers might exploit. Without adequate safeguards, data transmitted on the internet is significantly more susceptible to unauthorized access, theft, or alteration. The identification of unauthorised access attempts is a critical component of cybersecurity as it aids in the detection and prevention of malicious attacks. This research paper introduces a novel intrusion detection framework that utilizes Recurrent Neural Networks (RNN) integrated with Long Short-Term Memory (LSTM) units. The proposed model can identify various types of cyberattacks, including conventional and distinctive forms. Recurrent networks, a specific kind of feedforward neural networks, possess an intrinsic memory component. Recurrent Neural Networks (RNNs) incorporating Long Short-Term Memory (LSTM) mechanisms have demonstrated greater capabilities in retaining and utilizing data dependencies over extended periods. Metrics such as data types, training duration, accuracy, number of false positives, and number of false negatives are among the parameters employed to assess the effectiveness of these models in identifying both common and unusual cyberattacks. RNNs are utilised in conjunction with LSTM to support human analysts in identifying possible intrusion events, hence enhancing their decision-making capabilities. A potential solution to address the limitations of Shallow learning is the introduction of the Eccentric Intrusion Detection Model. This model utilises Recurrent Neural Networks, specifically exploiting LSTM techniques. The proposed model achieves detection accuracy (99.5%), generalisation (99%), and falsepositive rate (0.72%), the parameters findings reveal that it is superior to state-of-the-art techniques.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

A Novel Eccentric Intrusion Detection Model Based on Recurrent Neural Networks with Leveraging LSTM

Muthunambu,

Prabakaran,

Kavin

et al. 2024

CMC

View full text Add to dashboard Cite

show abstract

“…On the face of it, the device monitoring seems to be done for the sake of speed; behind the scenes, though, its objective is to discover disturbances, issues, or dangers. Typical for web-servers, the control system is usually is mounted on a separate computer and is configured to keep track of other computers and software [37]. Today, the network is the only place to get to knowledge and records.…”

Section: Software As a Service (Saas)mentioning

confidence: 99%

Performance Analysis of Enterprise Cloud Computing: A Review

Malallah¹,

Qashi²,

Abdulrahman³

et al. 2023

JASTT

View full text Add to dashboard Cite

Cloud computing has swiftly established itself as the norm in its field as a result of the advantages described above. In an attempt to reduce the amount of time spent on infrastructure upkeep, an increasing number of businesses are moving their operations to the cloud. As a result, maintaining a cloud environment has proved to be exceedingly difficult. It is necessary to have an efficient cloud monitoring system in order to reduce the workload associated with administration and improve cloud operation. The cloud monitoring service is beneficial since it has the potential to improve performance and make administration easier. The administration of Quality of Service (QoS) parameters for cloud-hosted, virtualized, and physical services and applications is one of the most important responsibilities of cloud monitoring. As a result, cloud management software retains a record of both actions and services, and it also conducts dynamic setups of the cloud in order to increase operational effectiveness. The performance of businesses and businesses as a whole was examined in this article, as was the influence that cloud-ready programs and tools have on that performance, as well as the advantages that may be obtained from adopting such programs and products.

show abstract

“…Lastly, Salih et al [30] proposed a model for improving the performance of network intrusion detection systems. The author used an Anaconda Python open-source platform to deploy a set of algorithms for selecting features for authentication.…”

Section: Related Workmentioning

confidence: 99%

Grey Wolf-Based Method for an Implicit Authentication of Smartphone Users

Almazroi¹,

Eltoukhy²

2023

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

Smartphones have now become an integral part of our everyday lives. User authentication on smartphones is often accomplished by mechanisms (like face unlock, pattern, or pin password) that authenticate the user's identity. These technologies are simple, inexpensive, and fast for repeated logins. However, these technologies are still subject to assaults like smudge assaults and shoulder surfing. Users' touch behavior while using their cell phones might be used to authenticate them, which would solve the problem. The performance of the authentication process may be influenced by the attributes chosen (from these behaviors). The purpose of this study is to present an effective authentication technique that implicitly offers a better authentication method for smartphone usage while avoiding the cost of a particular device and considering the constrained capabilities of smartphones. We began by concentrating on feature selection methods utilizing the grey wolf optimization strategy. The random forest classifier is used to evaluate these tactics. The testing findings demonstrated that the grey wolf-based methodology works as a better optimum feature selection for building an implicit authentication mechanism for the smartphone environment when using a public dataset. It achieved a 97.89% accuracy rate while utilizing just 16 of the 53 characteristics like utilizing minimum mobile resources mainly; processing power of the device and memory to validate individuals. Simultaneously, the findings revealed that our approach has a lower equal error rate (EER) of 0.5104, a false acceptance rate (FAR) of 1.00, and a false rejection rate (FRR) of 0.0209 compared to the methods discussed in the literature. These promising results will be used to create a mobile application that enables implicit validation of authorized users yet avoids current identification concerns and requires fewer mobile resources.

show abstract

Deep Learning Approaches for Intrusion Detection

Cited by 19 publications

References 50 publications

A Novel Eccentric Intrusion Detection Model Based on Recurrent Neural Networks with Leveraging LSTM

A Novel Eccentric Intrusion Detection Model Based on Recurrent Neural Networks with Leveraging LSTM

Performance Analysis of Enterprise Cloud Computing: A Review

Grey Wolf-Based Method for an Implicit Authentication of Smartphone Users

Contact Info

Product

Resources

About