HYBRID ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING TECHNOLOGIES Hybrid optimization scheme for intrusion detection using considerable feature selection S. Velliangiri 1 • P. Karthikeyan 2 Received: 17 December 2018 / Accepted: 29 August 2019 Ó Springer-Verlag London Ltd., part of Springer Nature 2019 Abstract The intrusion detection is an essential section in network security because of its immense volume of threats which bothers the computing systems. The real-time intrusion detection dataset comprises redundant or irrelevant features. The duplicate features make it quite challenging to locate the patterns for intrusion detection. Hybrid optimization scheme (HOS) is designed for combining adaptive artificial bee colony (AABC) with adaptive particle swarm optimization (APSO) for detecting intrusive activities. The schemes are aggregated for locating improved optimization-based outcomes, and the precision during categorization is acquired using tenfold cross-validation scheme. The main objective of the proposed method is to improve the rate of precision in intrusion activities in internetwork by choosing the relevant features. Effectiveness of the hybrid categorization scheme is accessed using an NSL-KDD dataset. Single feature selection method and random feature selection method are used to assess the proposed HOS intrusion detection approaches. The effec- tiveness of the designed scheme is evaluated with existing machine learning schemes such as Naive Bayes, AABC, APSO, and support vector machine, which outperform the HOS. Keywords Intrusion detection  AABC  APSO  Support vector machine  Hybrid optimization scheme 1 Introduction Intrusion detection is a novel security method for discov- ering, checking, and resisting unauthorized access to a computer network or internetwork. Due to the remarkable advancements in the domain of data technology, there prevail crucial disputes in network confidentiality. There- fore, an intrusion detection system (IDS) is vitally needed for a network to assure safety. IDS can be categorized into different methods. The main categorizations are active and passive IDS. An active IDS is also recognized as an intrusion detection and prevention system (IDPS). IDPS is designed to spontaneously block mistrusted attacks without any intrusion required by an operator. IDPS has the benefit of providing concurrent remedial action in reply to an attack. A passive IDS is a system that is designed only to observe and examine network traffic activity and alert an operator to possible weaknesses and attacks. A passive IDS does not perform any protective or remedial functions [1–3]. The conventional active IDS is incapable of resolving novel occurring threats. The fundamental inten- tion of IDS is to locate and differentiate between the usual and unusual network connections as one of the significant problems in locating intrusions because of its immense volume of elements and feature. IDS is broadly applied for finding the manner and place of intrusion occurrence. The investigators carried out an in-depth analysis of different schemes accomplishing element choice for achieving real- time intrusion detection. Minimizing the count of feature based on the choice of the essential feature is an excellent dispute for enhancing the precision and speed of catego- rization schemes [4, 5]. Therefore, the choice of distin- guishing feature and modeling optimal categorizer model regarding improved precision and rate of the finding are the key features. The analysis of machine learning or & S. Velliangiri velliangiris@cmritonline.ac.in P. Karthikeyan karthikeyanp@presidencyuniversity.in 1 Department of Computer Science and Engineering, CMR Institute of Technology, Hyderabad, Telangana, India 2 Department of Computer Science and Engineering, Presidency University, Bengaluru 560064, India 123 Neural Computing and Applications https://doi.org/10.1007/s00521-019-04477-2