Use of artificial intelligence in Cybersecurity

Notably, physical devices like detectors and sensors are not effective enough to monitor and offer protection of the infrastructures. Therefore, there is a need for exceedingly sophisticated Information Technology that can create normal behaviors as well as detect, assess and evaluate abnormal behavior. There is much evidence showing the significance of Artificial intelligence in combating cybercrimes. The evidence suggests that Artificial Intelligence provides leaning and flexibility capabilities to software which aid human in detecting and preventing cybercrimes. What is cyber security? According to Goutam (2015), cyber security can be defined as the range of technologies and processes which are constructed purposely for protection of computers, their hardware, data, and software as well as computer networks for any access that is unauthorized and vulnerabilities that are passed through the internet by hackers, cybercriminals or terrorist groups.

According to Sattikar & Kulkarni (2012), AI is a vital means through which subjectivity reduction, as well as the human involvement in the assessment of security, is done. This is done through AI techniques such as the Natural Language processors which are the close semblance of human brains. AI is used in the detection of intrusion behavior through the misuse detection method and the anomaly detection. The tasks of solving any intrusion detection can also be achieved through the AI. Moreover, Sattikar & Kulkarnia (2012) indicates that AI through techniques such as the artificial neural network may be imperative in the classifying a system intrusion detection. Additionally, they stated that the existence of the closed source code is a vital attribute of the malevolent AI.

The risks that can be caused by AI systems may range from simple phishing to catastrophic risks. Through combining phishing with intelligent and automated technology, AI could be utilized to make influences on the next generation of phishing techniques that are more sophisticated. According to Yampolskiy (2015) failures of the already established cybersecurity systems is often curable since it mainly causes privacy or monetary risks. However, the failure of SAI (Super intelligent Artificial System) has the ability to cause random risky events which may result in large-scale losses and damage to the welfare of human beings. Artificial Intelligence as a Cyber Security Tool The advantage of utilizing AI over initial cybersecurity techniques is that the initial techniques approached the issues like fixing plumbing as noted by Morel (Morel, 2011).

AI plays a larger role in making a paradigm shift from initial methods to the more comprehensive framework of cybersecurity. Additionally, AI technology could also aid software’s improve their own cyber securities potentials, hence improving the effectiveness as well as supplementing the shortfall in trained individuals. Dilek et al. (2015) similarly state that AI techniques may aid in the explanation of complex cybersecurity policies to the users as well as detecting abnormalities in systems that might be easily noticed. There is also significant literature on BCI systems though authors tend to differ on how cybersecurity solutions could be offered. According to Bonaci et al. , an effective solution is the engineering solution in BCI anonymizer. This anonymizer does pre-process the numeral signals prior to their transmission or storage hence playing the role of a filter before any information gets to the processor.

On the hand, Victoria Tur states that policy instead of technology may be relied on to control the data collection as well as usage through BCI systems. In this meeting these demands, Nojeim indicates the government must take precaution to avoid infringement of innovation, privacy, and liberty which must be bestowed upon individual above anything else. To achieve this, policy implementation by the government is considered effective. Klein (2016) indicates that the need for internet freedom in the globe is characterized by competing values such as the determination to promote liberty and privacy while on the other hand serious crimes and protection of the country against terrorism. Therefore, Klein (2016) provides legal analysis on whether such penetration by the government to private information through the Fourth and First Amendment are necessary and lawful.

Joyce (2015) argues whether as a result of the increased correlation between cybersecurity and the internet, freedom of the internet should be considered as a basic human right. Muller (2015) further provides working examples of the CCB by Microsoft’s cybersecurity global assessment steps which have been designed to create classification systems and models that are based on the data that they own such as the Linking Cyber Security Policy Performance. This approach is designed to provide a system of comparative classification which compares the different cyber-capacity of nations across the globe. Lango (2016) on the other hand provides the CCB form that is generally applied in many nations of the world. This is the national computer emergency response team (CERT) which is divided into two to include the rescue and radar models.

However, to endure the risks of cyber-attacks, an integrative method whereby there is the inclusion of several actors is highly recommended instead of having a single actor (Lango, 2016). Erica Fraser (2016) describes Artificial Intelligence strategies like artificial neural networks, genetic programming as well as robot scientist that are utilized to create inventions. She acknowledges that patents were initially granted for inventions making use of AI and the creation method of the invention does not factor into patent granting processes. To point out the existence of the inventive step in patents, it is of great significance to point out the idea of an individual or ordinary skill in the art. Since Artificial Intelligence does raise the level of skills effective for ordinary inventors, this idea must be changed in the light of a contemporary inventor as well as the technology used.

Likewise, the rich information that Artificial Intelligence technologies entail, have to be considered into an account when evaluating obviousness. Through personal interviews and literature survey, Lasse makes the conclusion that ambiguity with respect to the rights and use of inputting data utilized in the machine. Lasse maintains that this is a major IP management issues, particularly in cases whether an input’s owner also possesses rights to an output from an MLS. The presence of multiple owners of input data offers restriction on the Later MLS which utilizes billions of inputs which could have corresponding IP rights. Even in the case where MLS may utilize data without any permission, copyright owners would still get it difficult proving the same since they lack understanding of how and what data the data was used for by MLS.

Application of Artificial Intelligence to defense against cyber crimes The automation of defense against cyber-attacks is considered by Tyugu & Branch as important since the required speed in processing a certain amount of data cannot be effective when handled by humans (2010). Wang et al. (2008) allege that in future, antivirus detection technology is applicable in Heuristic Technology meaning that the skills and knowledge that utilize some techniques to analyze and determine codes to detect any unknown virus through various rules while scanning. To understand the application of AI, there is a need to show the various ways it is being applied as shown by the existing literature. Expert systems application Anwar and Hassan (2017) Expert system is one of the common AI tools in curbing cyber-security.

The expert system as a tool is used to help the users to explore answers to any inquiries that are presented either by software or by a client. Likewise, Wu (2009) also presented a neural technique of rule-based processing as well as backpropagation neural networks for filtering spam. According to Anwar & Hassan (2017), the neural nets are used for purposes of detecting and preventing the intrusions. Moreover, their primary use in the cyber defense system is based on its high speed especially when installing in either hardware or components of the graphics processors. Anwar and Hassan have hailed field Programmable Field Gate Arrays and the 3G neural nets for they permit quick improvement as well as their ability to conform with changing threat in cyberspace.

Intelligent Agent Application Evidence suggests that intelligent agents can be defined as autonomous forces that are computer generated made to communicate and cooperate with one another and share with each other to plan as well as implement effective responses are a case of any unexpected events. Their strategy is based on the intelligent multi-agent model and simulation, in which intelligent agent groups adjust and interact with their behavior and configuration with respect to the attack severity as well as network condition. Te scholars tested their techniques by investigating DoS distributed attacks and the defense mechanism and concluded that the ability and cooperation to adapt in the intelligent agent groups significantly increases the defense mechanism effectiveness. Artificial Immune System Applications AISs tend to be employed to aid in upholding stability in an environment that is changing.

The intrusion detection is made up of antigen and immunocytes detection simultaneously. The immune system creates antibodies that aid repels pathogens, as well as the intrusion intensity, could be measured through the antibody v concentration variation. Yiqian and Qiang (2010) proposed a network security situation evaluation model that makes quantitative and real-time security situation assessments as well as give the support needed to adjust in real time of defense measures. Wanbo and Rui (2010) proposed a self-learning intrusion design which recognizes and classifies unknown attacks. Their design entails a dynamic response decision-making technique that adjusts defensive techniques with respect to the environment as well as ensures the systems is safe at a low cost. The experiments exhibited that the design had qualities like self-adaptation, quantitative calculation, rationality as well as providing efficient intrusion responses.

Endy et al. The simulation findings exhibited that the proposed design improved the performance as well as decreased the database size, false alarm rate as well as time complexity. Mabu et al. (2011) described the novel fuzzy network intrusion detection method with respect to the class-association-rule mining in programming genetic network. The suggested method is efficient and flexible for both anomaly and misuse detection in networks and can deal with mixed databases containing both continuous and discrete attributes to mine vital class association rules required for advanced intrusion detection. The evaluation and experiments of the proposed method showed that the technique offers competitively high rates of competition with other techniques of machine learning. (2013) introduced a network IDS that is based on a fuzzy genetic algorithm.

Fuzzy rules are utilized in classifying network attack data while the genetic algorithm does optimize finding the appropriate fuzzy rule to attain an optimal solution. The assessment findings exhibited that the proposed IDS could detect network attacks in the real tome upon data’s arrival to a detecting system that has a detection rate exceeding 97. Padmadas et al. (2014) exhibited a layered genetic algorithm that is based on intrusion detecting systems for monitoring various activities in a provided environment in determining if they are malicious or legit with respect to the present information, confidentiality, and system integrity. Scholars have examined into the mappings, semantic, knowledge acquisition as well as the selection procedure of the natural languages. Resnik (1999) article on semantic similarity on taxonomy, he presents a semantic similarity measures in IS-A taxonomy based on the idea of shared information context.

The author shows algorithms that have an advantage over taxonomic similarity while resolving semantic and syntactic ambiguity. The projects offer a vivid understanding of the semantics concepts, hence improving problem solving perspective as well as work conceptualization in semantic. In accordance to Thompsons et al (2003) the authors focus on WOLFIE, (World Learning from Interpreted Examples). The researchers assumed that knowledge is either a set of prepositional formulae or prepositional models (interpretations)A interesting finding showed that formalism with similar time complexity did not belong to an equal space efficiency. However, in Di-Sciascio et al (2002) proposed a structured model to image retrieving problems as well as presenting the description logic that was devised for semantic indexing as well as retrieval of images entailing complex objects.

Likewise, Borgida and Kusters (2001) researches on functional relationships between various objects. The authors exhibited that despite determining sub Sumption between various concepts, descriptions have an equal complexity. Other AI Application There also exists literature on different AI hybrid applications in combatting cyber security. Elsadig et al (2010) explained the novel approach for self-healing system and bor-inspired intrusion prevention. They showed program which used intelligent multiagent system to point out abnormal behavior as well as detecting, healing dangerous or harmful events network system. Likewise, Zhou et al (2011) established a AIS based IDS used to combat virus with a virus. The authors implanted viruses and cloned variation of viruses into an immune IDS based on eLearning to advance system immunity as well as eliminating invasion or attacking behaviors.

Additionally, Meng (2011) studies the holistic intelligence of the Neuro Endocrine Immune system as well as presented the artificial homeostasis security coordination design. (2012) created an intrusion detecting system which utilizes detectors created by the genetic algorithm integrated with deterministic-crowding technique. The experiment attained an entire average of intrusion detecting rate of 81. Moreover, Patel et al (2013) suggested the GAAIS. That is, the dynamic intrusion detecting method for mobile ad hocs. This system is based on the genetic algorithm as well as artificial AIS. Algorithms should also be built with the ability to provide expert knowledge rather than relying on human efforts to identify threats such as the utilization of a Bayesian Belief Network (Veiga, 2018). According to Oracle (2018), automation is the only possible and potential way through which future security of data can be achieved.

The notion of fixing any vulnerability before an attack is launched is the only possible way to avoid data theft and security of data. The future of AI in cyber-security more specifically the machine learning, should include encompass the ability to detect and mitigate any automated threat, providing faster remediation after quick detection, ability to provide a continuous detection, analysis of risks through context and identity as well as the provision of more adaptive responses (Oracle, 2018). Anwar & Hassan (2017) the future of AI in combating attacks will be characterized by the use of expert systems that have extensive applications which will require a large investment and extended modular bases of knowledge. Advancing Cybersecurity Capacity Building. Retrieved from: http://www. gppi. net/fileadmin/user_upload/media/pub/2017/Hohmann__Pirang__Benner 2017_Advancing_Cybersecurity_Capacity_Building.

pdf. Wijesinghe, L. S. et al. Combating Cyber Crime Using Artificial Agent Systems. International Journal of Scientific and Research Publications 6(4), 2016. Fischer, E. A. Cybersecurity issues and challenges: in brief. Nojeim, G. T. , & Van Niekerk, J. From information security to cyber security.  computers & security, 38, 97-102. Anwar, A. , & Hassan, S. Allen, G. , & Chan, T.  Artificial intelligence and national security. Cambridge, MA: Belfer Center for Science and International Affairs. Sattikar, A. , Dey, A. , Pal, A. , & Roy, N. Applications of artificial intelligence in machine learning: review and prospect.  International Journal of Computer Applications, 115(9). Dilek, S. , Çakır, H. , & Aydın, M. Applications of artificial intelligence techniques to combating cyber-crimes: A review.  arXiv preprint arXiv:1502. pdf. Accessed on August 2, 2018. Patel, M. Taghavi, K. Bakhtiyari, J. Systematically Understanding the Cyber Attack Business: A Survey.

 ACM Computing Surveys, 51(3), 1-36. doi:10. 1145/3199674 Coulter, R. , & Pan, L. , & Moslehpour, S. Cyber Security Management: A Review.  Business Management Dynamics, 5(11), 16-39. Mitchell, R. , & Chen, I. 9, No. 5, pp. Sharma, S. Kumar, M. Kaur, (2014) “Recent Trend in Intrusion Detection using Fuzzy-Genetic Algorithm,” International Journal of Advanced Research in Computer and Communication Engineering, Vol. Hassan M (2013) “Network Intrusion Detection System Using Genetic Algorithm and Fuzzy Logic”, International Journal of Innovative Research in Computer and Communication Engineering, Vol. 1, No. Jongsuebsuk N. Wattanapongsakorn, C. Charnsripinyo, (2013) "Real-time intrusion detection with fuzzy genetic algorithm," 10th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON), pp. Mabu, C. Chen, L. Nannan, K. Shimada, K. Hirasawa, (2011) "An Intrusion-Detection Model Based on Fuzzy Class-Association-Rule Mining Using Genetic Network Programming," IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews, Vol.