HCRNNIDS: Hybrid Convolutional Recurrent Neural Network-Based Network Intrusion Detection System

Nowadays, network attacks are the most crucial problem of modern society. All networks, from small to large, vulnerable threats. An intrusion detection (ID) system is critical for mitigating and identifying malicious threats in networks. Currently, deep learning (DL) machine (ML) being applied different domains, especially information security, developing effective ID systems. These systems capable detecting automatically on time. However, occurring changing continuously, so requires a very advanced security solution. Thus, creating an smart massive research problem. Various datasets publicly available research. Due complex nature with constantly attack mechanism, existing must be modified systematically regular basis. So, this paper, convolutional recurrent neural (CRNN) used create DL-based hybrid framework that predicts classifies cyberattacks network. In HCRNNIDS, (CNN) performs convolution capture local features, (RNN) captures temporal features improve system’s performance prediction. To assess efficacy (HCRNNIDS), experiments were done data, specifically realistic CSE-CIC-DS2018 data. The simulation outcomes prove proposed HCRNNIDS substantially outperforms current methodologies, attaining high rate accuracy up 97.75% CSE-CIC-IDS2018 data 10-fold cross-validation.