Protection Against Information in eSociety: Using Data Mining Methods to Counteract Unwanted and Malicious Data

Abstract

Despite the positive aspects of usage of the Internet and social networks within the concept of eSociety, huge data collections available for viewing and analysis to the user of the Internet can contain information which can be unwanted or malicious. The paper considers the problem of protection of users in the “electronic society” infrastructure against such information. The paper discusses the nature of the problem and possible approaches to its solution. To solve the problem it is proposed to use modular approach to construction of automated systems of protection against information, based on application of Data Mining methods. We consider the implementation of the system of protection against unwanted and harmful content, based on the classifier with three-level hierarchical architecture. Its experimental evaluation, which confirmed high efficiency of functioning of the system for most of the analyzed categories of web sites, are also discussed.