Cloud Attack Dataset For Building Machine Learning and Deep Learning Models
Description
The cloud attacks dataset is prepared for evaluating the performance of machine learning and deep learning models which are built and designed to detect network attacks. The traffic dataset is provided with training and testing sets. The training set consists of 2500 Benign (normal) traffic instances and 2500 DDoS attack traffic instances in Binary visualization form suitable for building deep learning models. The test set consists of 584 Benign traffic instances and 584 DDoS attack traffic instances in image representation. Thus, the dataset consists of traffic instances categorized into two classes (i) Benign traffic and (ii) DDoS Attack traffic. The dataset is a balanced dataset. This dataset is obtained by considering the traffic instances from the CICIDS 2019 dataset which provides various categories of modern network attacks along with Benign traffic instances. Interested researchers can also refer to the IoT DoS DDoS Attack dataset which is an imbalanced dataset generated using CICIDS 2019 dataset. The dataset provided here can be used to test the performance of advanced machine learning algorithms and deep learning models aimed at modern network attacks detection in Cloud and IoT environments.
Files
Steps to reproduce
This dataset is obtained by considering the traffic instances available originally from the CICIDS 2019 dataset in pcap and CSV form which provides various categories of modern network attacks along with Benign traffic instances.