Toward Constructing a Balanced Intrusion Detection Dataset

Several Intrusion Detection Systems (IDS) have been proposed in the current decade. Most datasets which associate with intrusion detection dataset suffer from an imbalance class problem. This problem limits performance of classifier for minority classes. paper has presented a novel processing technology large scale multiclass dataset, referred to as BMCD. Our algorithm is based on adapting Synthetic Minority Over-Sampling Technique (SMOTE) improve rate classes while ensuring efficiency. In this work we combined five individual CICIDS2017 create one contains several types attacks. To prove efficiency our algorithm, machine learning algorithms applied and without using BMCD algorithm. The experimental results concluded that provides effective solution imbalanced outperforms state-of-the-art methods.