Genetic programming support vector machine model for a wireless intrusion detection system

Objectives. The rapid penetration of wireless communication technologies into the activities both humans and Internet Things (IoT) devices along with their widespread use by information consumers represents an epochal phenomenon. However, this is accompanied growing intensity successful attacks, involving bot attacks via IoT, which, network has reached a critical level. Under such circumstances, there increasing need for new technological approaches to developing intrusion detection systems based on latest achievements artificial intelligence. most important requirement system consists in its operation various unbalanced sets attack data, which different techniques. synthesis difficult task due lack universal methods detecting technologically attacks; moreover, consistent application known unacceptably long. aim present work eliminate scientific gap. Methods . Using intelligence fight against authors proposed method combination genetic programming support vector machine (GPSVM) model using CICIDS2017 dataset. Results. presented architecture offers possibility train dataset extracting objects. provides separation verifiable not elements, latter being added training set feedback. By improving GPSVM set, better accuracy ensured. operability flowchart demonstrated terms entry input data output after processing model. Numerical analysis results experiments selected quality indicators showed increase as compared SVM method. Conclusions. Computer have confirmed methodological correctness choosing effectiveness detection. A procedure forming feedback proposed. datasets shown create conditions collect sample increases theaccuracy provide improved performance