Comprehensiveness of linguistic data summaries: A crucial role of protoforms

Abstract

We show first the essence of our approach to linguistic database summaries, equated with linguistically quantified propositions in Zadeh's sense and mined through the use of a fuzzy querying interface to a database. We recast the problem from the perspective of comprehensiveness of patterns derived by linguistic data summaries. Motivated by Michalski's [21] seminal approach to the comprehensiveness of data mining and machine learning results in which he advocates the use of natural language, we advocate the use of linguistic summaries which provide a new quality and an exceptional human consistency and comprehensiveness. We illustrate our analysis by two examples related to the linguistic summarization of both static and dynamic data in the area of analysis of innovativeness of companies and of Web server log files. © Springer-Verlag Berlin Heidelberg 2013.