C SVM Classification and KNN Techniques for Cyber Crime Detection

Abstract

In the digital age, cybercrime is spreading its root widely. Internet evolution has turned out to a boon as well as curse for those confronting the issues of privacy, national security, social decency, IP rights, child protection, fighting, detecting, and prosecuting cybercrime. Hence, there arises a need to detect the cybercriminal. Cybercrime identification utilizes dataset that is taken from CBS open dataset. For identifying the cybercriminal, support vector machine (SVM) in the C SVM classification and K -nearest neighbor (KNN) models is utilized for determining the cybercrime information. The evaluation of the performance is done taking the following metrics into consideration: true positive, false positive, true negative and false negative, false alarm rate, detection rate, accuracy, recall, precision, specificity, sensitivity, classification rate, and Fowlkes-Mallows Scores. Expectation maximization (EM) calculation is utilized for evaluating the presentation of the Gaussian mixture model. The performance of classifier’s presentation is also done. Accuracy is accomplished in the event of grouping by means of SVM classifier as 89% in the supervised method.