Amit Kumar
Computer Science
March 2023
It has become harder to identify and stop cyber attacks using conventional security measures as their frequency and sophistication have increased. A subset of machine learning called deep learning has demonstrated considerable promise for enhancing cyber threat identification and response. In this article, we outline a method for detecting sophisticated cyber attacks that uses Deep Learning algorithms. The proposed strategy extricates includes and sorts network traffic information utilizing convolutional brain organizations (CNNs) and intermittent brain organizations (RNNs). The RNN is used to capture temporal dependencies while the CNN is used to extract spatial characteristics from the network data. To choose the most pertinent features for categorization, the proposed approach additionally incorporates a feature selection stage. Utilizing an assortment of datasets, we survey the exhibition of the proposed system and show that it beats signature-based and exemplary AI strategies with regards to exactness, accuracy, review, and F1-score. The proposed method is a powerful tool for enhancing cyber security because it can identify zero-day and previously undiscovered assaults. Overall, our work show how Deep Learning may be used to tackle the problems associated with cyber security and offers a viable path for further study in this field.
241- 248
© Airo Journals 2021
