DDCATF: Deep Learning Approach for Detection of Cybercrime Activities Based on Temporal Features

The detection of cybercrime activities holds paramount importance in today's interconnected digital landscape. As technology becomes more integrated into every aspect of one's lives, the potential for cybercrime and its impacts have increased exponentially. In this context, the presented research employs a Deep Learning approach for Detection of Cybercrime Activities based on Temporal Features (DDCATF), specifically leveraging Convolutional Neural Networks (CNNs), to effectively detect and combat cybercrime activities. The significance of this approach lies in its potential to empower numerous organizations in identifying intricate and ever-evolving threats. By integrating temporal features into the detection process, these organizations can substantially strengthen their cybersecurity capabilities. To accomplish this objective, the research makes use of publicly accessible datasetnamed SIMARGL2022 and compared with other benchmark datasets: KddCup99 and CICIDS2017. These datasets serve as valuable resources for evaluating the CNN efficacy in identifying a diverse range of cybercrime activities. To determine the performance, a comprehensive set of evaluation metrics including accuracy, precision, recall, and f-measure are employed.