(IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 11, No. 6, 2020 Signature based Network Intrusion Detection System using Feature Selection on Android Onyedeke obinna Cyril 1 , Taoufik Elmissaoui 2 , Okoronkwo M.C ,3 , Ihedioha Uchechi .M 4 , Chikodili H.Ugwuishiwu 5 , Okwume .B. Onyebuchi 6 Department of Computer Science, University of Kairouan, Tunisia 1 Innov’Com , SUP’COM,University of Chartage & Higher Institute of Applied Mathematics and Computer Science, University of Kairouan, Tunisia 2 Department of Computer Science, University of Nigeria, Nsukka (UNN), MNCS, MCPN 3 Department of Computer Science, University of Nigeria, Nsukka (UNN) 4, 5 Department of Computer Science, University of Nigeria, Nsukka (UNN) 6 Abstract—This paper Smart Intrusion Detection System (IDS), is a contribution to efforts towards detecting intrusion and malicious activities on Android phone. The goal of this paper is to raise user’s awareness of the high rate of intrusions or malicious activities on Android phones and to provide counter measure system for more secured operations. The proposed system (SIDS) detects any intrusion or illegal activities on android and also takes a selfie of the intruder unknown to him/her and keep in the log for the view of the user. The object oriented analysis and design method (OOADM), was adopted in the development. This approach was used to model and develop the system using real intrusion features and processes to detect intrusions more flexibly and efficiently. Signature detection was also used to detect attacks by looking for specific patterns. The system detects intrusions and immediately sends an alert to the user to notify of an illegal or malicious attempt and the location of the intruder. Keywords—Signature Detection; Feature Selection; android phone; Smart Intrusion Detection System (SIDS) I. INTRODUCTION Smartphones, tablets, and other mobile platforms are rapidly emerging as popular appliances with progressively amazing computing, networking, and detecting abilities. Smartphones are currently the overwhelming individualized computing devices with so many features, and strength comparable to mini computers. Some of the attractive features of these smartphones include calls, short messages, multimedia, email, video calling, voice dictation, eservices, file exchange, internet browsing, services, etc. According to Pew Research Centre in 2015, about 43% of the global population uses a smartphone device [1]. Also, there were 5.11 billion interesting portable clients worldwide in 2019, and 2.71 billion of them use smart phones, it evaluated that there will be 2.87 billion smartphone clients worldwide in 2020, and 2.5 billion dynamic Android gadgets around the world, this value was based on Google’s Play store Statistics, and this implies that the number is higher. These numbers of Android devices and users additionally underscore the size of the fracture challenge and Google hopes to apply essential updates and security principles to all Android gadgets across various renditions, districts, and producers. Android was launched by Google and Open Handset Alliance in September 23, 2008. Android has experience a vast growth since its inception because of its user friendliness, open source, ease of developing and publishing applications. The ubiquitous usage of Android OS has induced the burst of mobile application market. Google Play is the largest app store followed by Apples App store. According to [2], The Android Applications are available for download through Google Play Store and third party agents. Though intrusion is not specific to android phones; most smart devices are used for e-businesses; which expose both private and financial data to public domain. Several techniques have been proposed and implemented to detect, prevent and reduce malicious intrusions on smartphones. Notwithstanding, intrusion is any unapproved action on a computer network. Much of the time, such undesirable action retains network assets expected for different utilizations, and about consistently compromises the security of the system as well as its information. Appropriately structuring and sending a system intrusion detection system will help obstruct the interlopers. Recognizing an intrusion relies upon the protectors having a clear understanding of how assaults work [3]. Intrusion activities seek to unsettle the confidentially, availability or integrity of a resource or the controlling applications. As a result of high prevalence, intrusion detection systems (IDS) are provided to checkmate intrusions. IDS is a sort of security measures use to alert the right owner of a device when a person or thing is attempting to bargain data framework through vindictive activities or through security approach encroachment. The Proposed system (SIDS) is focused at developing a model that will identify malicious intrusions on smart phones, through finger print and password validity and also takes a selfie of the intruder unknown to him/her and keep in the log for the view of the user. The techniques used for detecting intrusion can be arranged into Signature based location and Anomaly based recognition. Signature based detection is termed as misuse detection which helps in the detection of attacks by looking for specific patterns. Here, the dataset has number of occasions and each data must be named as typical or malevolent. In [4], AI calculations are utilized to prepare the informational collection as indicated by their name, and abuse identification strategy is made naturally. Contingent upon the 551 | Page www.ijacsa.thesai.org