Artificial Intelligence and Data Mining Approaches in Security Frameworks. Группа авторов

captcha.

      2 Whenever it does so a machine learning model based on its ability to recognize patterns from the past would detect the presence of bots through active monitoring and predictive analysis.

      3 If detected, it would terminate the current process and send out an alert.

      4 If the bot is not present then it would continue the process and run the anti-virus software, in order to remove any other malicious files.

      5 The Disaster recovery plan in the end would ensure that any important data is not lost and is backed up.

      1.5.1 System Architecture

Figure 1.2 System architecture [11].

      1.5.2 Future Scope

      While we are embracing new ways of digital interaction and more of our critical infrastructure is going digital, the parameters of the transformation underway are not understood by most of us. A better understanding of the global cyberspace architecture is required.

1.6 Conclusion

AI finds its applications in almost every field of science and engineering. AI models need precise safeguards in digital security and new technologies to battle antagonistic machine learning, retain confidentiality, and secure organized learning, and so on. In this chapter, the authors examined specific approaches in AI that are promising and proposed a system of preventing certain types of cyber-security attacks.

References

      1. Alex Roney Mathew, Aayad Al Hajj, Khalil Al Ruqeishi (2010), Cyber Crimes: Threats and Protection, International Conference on Networking and Information Technology, Manila.

      2. Cerli and D. Ramamoorthy (2015), Intrusion Detection System by Combining Fuzzy Logic with Genetic Algorithm, Global Journal of Pure and Applied Mathematics (GJPAM), vol. 11, no. 1.

      3. L.S. Wijesinghe, L.N.B. De Silva, G.T.A. Abhayaratne, P. Krithika, S.M.D.R. Priyashan, DhishanDhammearatchi (2016), Combating Cyber Using Artificial Intelligence System, International Journal of Scientific and Research Publications, vol. 6, no. 4.

      4. Naveen Kumar, Prakarti Triwedi, Pramod Singh Rathore, “An Adaptive Approach for image adaptive watermarking using Elliptical curve cryptography (ECC)”, First International Conference on Information Technology and Knowledge Management pp. 89–92, ISSN 2300-5963 ACSIS, Vol. 14 DOI: 10.15439/2018KM19

      5. Pramod Singh Rathore, “An adaptive method for Edge Preserving Denoising,” International Conference on Communication and Electronics Systems, Institute of Electrical and Electronics Engineers & PPG Institute of Technology (2017). Proceedings of the 2nd International Conference on Communication and Electronics Systems (ICCES 2017): 19-20 October, 2017.

      6. R. Hill (2010), Dealing with cyber security threats: International cooperation, ITU, and WCIT, 7th International Conference on Cyber Conflict: Architectures in Cyberspace.

      7. Onashoga, S. Adebukola, Ajayi, O. Bamidele and A. Taofik (2013), “A Simulated Multiagent-Based Architecture for Intrusion Detection System”, (IJARAI) International Journal of Advanced Research in Artificial Intelligence, vol. 2, no. 4.

      8. S. Dilek, H. Çakır and M. Aydın (2015), Application of Artificial Intelligence Techniques to Combating Cyber Crimes: A Review, International Journal of Artificial Intelligence & Applications (IJAIA), vol. 6, no. 1.

      9. S. Singh and S. Silakari (2009), A Survey of Cyber Attack Detection Systems, IJCSNS International Journal of Computer Science and Network Security, vol. 9, no. 5

10. Singh Rathore, P., Kumar, A., & Gracia-Diaz, V. (2020). A Holistic Methodology for Improved RFID Network Lifetime by Advanced Cluster Head Selection using Dragonfly Algorithm. International Journal of Interactive Multimedia and Artificial Intelligence, 6 (Regular Issue), 8. http://doi.org/10.9781/ijimai.2020.05.003

      11. Dr. Ritu Bhargava, Pramod Singh Rathore, Rameshwar Sangwa, February 18 Volume 4 Issue 2, “An Contemplated Approach for Criminality Data using Mining Algorithm”, International Journal on Future Revolution in Computer Science & Communication Engineering (IJFRSCE), pp. 236–240.

      1 *Corresponding author: [email protected]

      2

      Privacy Preserving Using Data Mining

       Chitra Jalota* and Dr. Rashmi Agrawal

       Manav Rachna International Institute of Research and Studies, Faridabad, India

       Abstract

      On the one hand, data mining techniques are useful to extract hidden knowledge from a large pool of data but on the other hand a number of privacy threats can be introduced by these techniques. The main aim of this chapter is to discuss a few of these issues along with a comprehensive discussion on various data mining techniques and their applications for providing security. An effective classification technique is helpful to categorize the users as normal users or criminals on the basis of the actions which they perform on social networks. It guides users to distinguish among a normal website and a phishing website. It is the task of a classification technique to always alert users from implementing malicious codes by labelling them as malicious. Intrusion detection is the most important application of data mining by applying different data mining techniques to detect it effectively and report the same in actual time so that essential and required arrangements can be made to stop the efforts made by the trespasser.

      Keywords: Data mining, security, intrusion detection, anamoly detection, outlier detection, classification, privacy preserving data mining

2.1 Introduction

A computer system has the ability to protect its valuable information, raw data along with its resources in terms of privacy, veracity and authenticity; this ability is known as computer security. A third party cannot read or edit the contents of a database by using the parameters i.e., Privacy/confidentiality and integrity. By using the parameter authenticity, an unauthorised person is not allowed to modify, use or view the contents of a database. When one or more resources of a computer compromises the availability, integrity or confidentiality by an action, it is known as intrusion. These types of attacks can be prevented by using firewall and filtering router policies. Intrusions can happen even in the most secure systems and therefore it is advisable to detect the same in the beginning. By employing data mining techniques, patterns of features of a system can be detected by an intrusion detection system (IDS) so that anomalies can be detected with the help of an appropriate set of classifiers. For easy detection of intrusion, some important data mining techniques such as classification and clustering are helpful.

Test data could be analysed and labelled into known type of classes with the help of classification techniques. For objects grouping into a set of clusters, clustering methods are used. These methods are used in such a way that a cluster has all similar objects. There could be some security challenges for mining of underlying knowledge from large volumes of data as well as extraction of hidden patterns by using data mining techniques (Ardenas et al., 2014). To solve this issue, Privacy Preserving Data Mining (PPDM) is used, which aims to derive important and useful information from an unwanted or informal database