Linguistic summarization of the contents of Web server logs via the Ordered Weighted Averaging (OWA) operators

Analyses of Web server logs may be very useful and are often needed for designers and analysts of computer networks who try to design and optimize the functioning of the systems, safeguard them against attacks, provide for the most effective and efficient access path for the users/customers, etc. Obviously, they are also an interesting research problem that can be viewed from many perspectives, ranging from just a general analyses to topic focused analyses that are aimed at just a specific aspect. In traditional approaches, various statistics are computed and used for analytic and design purposes. Sometimes, visualization tools are also employed. In this paper we present the use of verbalization of results of Web server log data analysis/mining through linguistic data summaries based on fuzzy logic with linguistic quantifiers. Linguistic summaries of both static and dynamic analyses are presented, with an emphasis on the latter. We extend our previous works by employing as an aggregation tool the Ordered Weighted Averaging (OWA) operators due to Yager [52]. We present some examples of potentially interesting linguistic summaries of Web server logs, and indicate their possible assignment to different classes.