Identification of Network Traffic Intrusion Using Decision Tree

Network technology plays an increasingly important role in all aspects of social life. The Internet has brought a new round industrial revolution and upgrading. arrival the “Internet” era is accompanied by large-scale increase network applications number netizens. At same time, severity cyberattacks continue to increase. Therefore, intrusion detection systems (IDSs) have become part current security infrastructure various industries. Anomaly traffic data effective method for protection. In order better realize anomalies, several algorithms been successfully applied. Most them come from artificial intelligence (AI), but there general problem excessive model execution processing time low rates. And through lot research, it found that most models do not pay enough attention early stage. this paper, we optimize normalization process series experiments combine PCA feature selection propose optimized MaxAbs-DT classifier model. To train measure performance model, used NSL-KDD dataset, which benchmark dataset anomaly models. experimental results show outperforms other existing validates effectiveness method. addition, its greatly reduced compared many