ANALISA CLUSTERING PHISING DENGAN K-MEANS DALAM MENINGKATKAN KEAMANAN KOMPUTER

Abstract

Almost the crime in cyber is a condition of criminal activity using computers or computer networks as tools and also as a target. Fraud in academic websites the most at risk. The action of Phishing is on the rise. Recorded globally, the number of fraudulent mode phishing 42% of the mode in addition to phishing which is stated in the website Anti-Phishing Working Group (APWG) in its monthly report, noting there 12.845 e-mail new and unique as well as 2.560 a fake site that is used as a means of phishing, in Addition to increase the quantity, the quality of the attacks is also increasing, the need for the work done by the network administrator in improving surveillance in monitoring activity on the network, in the action of data theft will perform the action of manipulating someone with the appearance of a particular web site. In this study a set of datasets will be clustering using k-means, K-Means algorithm will classify the dataset, resulted in the identification of phishing that is accurate and certifiable. With the results of this research iteration=10, the K-Fold=2 the of the Bouldin Davis index = 0.119.

Keywords

• clustering
• phishing
• k-means clustering
• computer security
• data mining