Classification of phases based on a Principal Component Analysis for Intrusion Detection Methods

The impacts of an intrusion can be dangerous for the information system of an organizational being. According to the Quebec office of the French language, an intrusion is an operation which consists in accessing, without authorization, the data of a computer system or a network, bypassing or defusing the security devices put in place. The detection of an intrusion is not the end in itself, but also the optimization of the reaction time, that is to say minimize the time between detection and reaction, for this reason we use the of experts to assess the effectiveness of a method and its phases. In this work we will propose an approach based on principal component analysis allowing the development of the typology of an intrusion detection method based on the opinion of experts in the field. Certainly the matrix of the example used is very small, but the aim is to propose a classification approach based on a Principal Component Analysis (PCA). The profiles are drawn up on the basis of a detailed study of the variables and individuals in relation to the axes generated by the PCA.