We use these methodologies to solve application level DDoS attack problems in various application types with accuracy close to accurate performance. We are have testing the limit of AI and will release this feature for enterprise businesses in the near future on top of our network layer protection.
In this research, we did a thorough analysis of the logs generated during a DDOS attack, used supervised and unsupervised techniques for detection of threat, and finally used deep learning to achieve over 96% accuracy for classification of different types of DDoS threats along with the safe connection.
Processing the data was one of the first challenges faced by us. The data had 88 attributes or features. Processing such huge data within limited RAM memory was a really challenging task for us. So we downgraded the data type of the attributes, and hence reducing the memory usage of the data frame.
Even if you have very little labelled data compared to unlabeled data in a real-life scenario, there are techniques like semi-supervised learning and self-supervised learning to achieve remarkable performance. Model fairness indicator is also one of the TensorFlow tools that could also be used for better model evaluation and performance scaling.