Classification of Network Security Attack using KNN (K-Nearest Neighbour) and Comparison of different Attacks through different Machine Learning Techniques

• Published in: 2024 3rd International Conference for Innovation in Technology (INOCON) pp. 1 - 7
• Main Authors: Agarwal, Muskan, Gill, Kanwarpartap Singh, Chauhan, Rahul, Kapruwan, Akanksha, Banerjee, Deepak
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.03.2024
• Subjects: Artificial Intelligence; Classification algorithms; Computer networks; Cybersecurity; Intrusion Detection; KNN Classifier; Machine learning; Machine learning algorithms; Network security; Systematics; Technological innovation
• DOI: 10.1109/INOCON60754.2024.10512250

Network security refers to the systematic measures used to safeguard computer networks and the sensitive information they transmit against potential harm, unauthorised entry, and breaches. The concept encompasses a range of strategies and protocols designed to ensure the continued accessibility, secrecy, and reliability of network resources. The technique of categorising and assigning labels to different types of detrimental actions and vulnerabilities that specifically target computer networks is sometimes referred to as network security attack categorization. The categorization discussed herein plays a vital role in the field of cybersecurity by enabling the understanding, identification, and management of security concerns. In the field of cybersecurity, a valuable strategy involves the classification of network security attacks via the use of the K-Nearest Neighbour (KNN) algorithm. Additionally, the comparison of numerous attacks may be conducted by using various machine learning methodologies. In conclusion, the categorising of network security attacks is a fundamental aspect of cybersecurity that plays a crucial role in safeguarding computer networks and the sensitive information they contain. This strategy enables organisations and security specialists to effectively analyse, identify, and mitigate a wide range of threats. In summary, the K-Nearest Neighbour (KNN) algorithm demonstrates a high level of accuracy, reaching 99.5%, when fine-tuned with various classification parameters. This result is derived from the methodology given in the present work.