Detect network anomalies using a linear SVM

According to the advent of science and the increasing volume of communications, the need for secure access to the network and information is felt beforehand, networks are a very good platform for attacking and transmitting data while transferring or storing data. Data transfer causes different anomalies in the data. Therefore, data analysis is very important for detecting overwhelming data and problem data, hence the researchers have been focusing a lot on discovering the intrusion and detecting anomalies in the network. The most popular methods are the use of data mining for the detection of pertinent data. Further studies on the use of data mining methods, combined methods to cover the weaknesses of other methods Have been developed.In this research, using a hybrid method to address the weaknesses of the abnormal discovery method is used by deep motor learning support algorithm to improve the selection of features. The data used in this study is KDD standard data. Finally, the proposed method is compared with other methods in terms of speed and accuracy.