Cybersecurity enhancement using conditional generative adversarial network with transformer-based conditional variational autoencoder

Since, Artificial Intelligence is highly developing and concatenating into several domains, cybersecurity is an important field of delivering both the advantages and disadvantages. In addition to this, Artificial Intelligence is applied in a wide variety of applications like healthcare sector, conte...

Full description

Saved in:
Bibliographic Details
Published in:Engineering applications of artificial intelligence Vol. 161; p. 112220
Main Authors: Singh, Prithvipal, Singh, Sandeep, Singh, Gurupdesh, Singh, Amritpal
Format: Journal Article
Language:English
Published: Elsevier Ltd 12.12.2025
Subjects:
Conditional generative adversarial network
Conditional hybrid network
Cybersecurity enhancement
Generative artificial intelligence
Spatial-temporal attention mechanism
Transformer-based conditional variational AutoEncoder
Conditional hybrid network
Transformer-based conditional variational AutoEncoder
Conditional generative adversarial network
Generative artificial intelligence
Spatial-temporal attention mechanism
Cybersecurity enhancement
ISSN:0952-1976
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Since, Artificial Intelligence is highly developing and concatenating into several domains, cybersecurity is an important field of delivering both the advantages and disadvantages. In addition to this, Artificial Intelligence is applied in a wide variety of applications like healthcare sector, content creation and entertainment and financial industries. Therefore, this work finds the efficiency of Artificial Intelligence -oriented cybersecurity metrics in succeeding the digital environment over elevating cyber threats. Here, the developed models consist of two different stages while implementing the model. In the first stage, the essential dataset is assembled from the benchmark data source. These datasets are assembled by using Generative Artificial Intelligence (Gen Artificial Intelligence networks). Consequently, the raw data is given as an input to Conditional Hybrid Network for cybersecurity enhancement. Further, the Transformer-based Conditional Variational Autoencoders with Spatial-temporal Attention are designed for feature extraction that is subjected to the Conditional Generative Adversarial Network for classifying the cyber attacks. Henceforth, the developed network is evaluated and designed with multiple measures. Comparing baseline models, the suggested network obtains higher performance for developing security over cyber networks.
ISSN:0952-1976
DOI:10.1016/j.engappai.2025.112220