Sparse auto encoder driven support vector regression based deep learning model for predicting network intrusions

摘要: The Network Intrusion Detection System (NIDS) assumes a prominent aspect in ensuring network security. It serves better than traditional security mechanisms, such as firewall systems. result of the NIDS indicates enhanced and efficient performance algorithms. is utilized to predict intrusions, it also has training times for In this paper, capable deep learning model using Sparse Auto Encoder (SAE) proposed. self-taught framework. Such competent unsupervised algorithm reconstructing new feature representation; thus, diminishes dimensionality. SAE requires minimum time substantially efficiently enhances prediction accuracy Support Vector Regression (SVR) related attacks. experiments are administered standard intrusion detection dataset NSL-KDD, therefore, implementations performed python tensor flow. proposed model’s effectiveness estimated with other models viz., PCA-SVR SVR applying metrics R2 score, Mean Squared Error (MSE), Absolute (MAE), Root (RMSE), time. Results validate that SAE-SVR accelerated edge over weighed terms metrics. improves rate by bringing down error rates yields pioneering research mechanism predicting intrusions.