Open Access

ARTICLE

Optimized Fuzzy Enabled Semi-Supervised Intrusion Detection System for Attack Prediction

Gautham Praveen Ramalingam¹, R. Arockia Xavier Annie¹, Shobana Gopalakrishnan^2,*

1 Department of Computer Science and Engineering, CEG, Anna University, Chennai, 600025, India
2 Department of Information Technology, Loyola ICAM College of Engineering and Technology, Chennai, 600034, India

* Corresponding Author: Shobana Gopalakrishnan. Email:

Intelligent Automation & Soft Computing 2022, 32(3), 1479-1492. https://doi.org/10.32604/iasc.2022.022211

Received 31 July 2021; Accepted 28 September 2021; Issue published 09 December 2021

Abstract

Detection of intrusion plays an important part in data protection. Intruders will carry out attacks from a compromised user account without being identified. The key technology is the effective detection of sundry threats inside the network. However, process automation is experiencing expanded use of information communication systems, due to high versatility of interoperability and ease off 34 administration. Traditional knowledge technology intrusion detection systems are not completely tailored to process automation. The combined use of fuzziness-based and RNN-IDS is therefore highly suited to high-precision classification, and its efficiency is better compared to that of conventional machine learning approaches. This model increases the accuracy of intrusion detection using Machine Learning Methodologies and fuzziness has been used to identify various categories of hazards, and a machine learning approach has been used to prevent intrusions. As a result, the hypothesis of security breaches is often observed by tracking system audit reports for suspicious trends of system use, and access controls for granting or limiting the degree of access to the network are often established as the result of an improvement in the detection accuracy of intrusions which is extremely effective.

---

Keywords

---