CICIoT2023 Dataset

The CICIoT2023 dataset is a comprehensive benchmark for evaluating intrusion detection systems in Internet of Things (IoT) environments. It captures realistic IoT network traffic, including benign flows and multiple attack scenarios such as Distributed Denial of Service (DDoS), Denial of Service (DoS), reconnaissance, and data exfiltration. The dataset was generated in a controlled testbed using diverse IoT protocols and devices to reflect real-world heterogeneity. Each record contains detailed flow-level features (e.g., packet counts, byte rates, inter-arrival times) and is labeled for supervised learning tasks. CICIoT2023 addresses key challenges in IoT security research, including class imbalance, protocol diversity, and scalability, making it suitable for machine learning, deep learning, and explainable AI studies. The dataset is provided in CSV format with accompanying documentation for reproducibility and ease of integration into IDS research pipelines.