Text-mining in Terms of Methodology and Development

Text-mining is gaining popularity and has attained the status of cross-field technology, which brings together soft and hard sciences. This happens by virtue of its applicability in various fields, for instance, Linguistics, for language units' identification and categorization, History, for ancient texts reconstruction, Sociology, for sentiment analysis, and others. In this paper, we make a deep analysis of the methodological background of text-mining. We prove that text-mining has accumulated several methodologies, which make it flexible for various purposes and fuzzy enough to bring into proximity of human-like text comprehension and knowledge extraction. We also make a retrospective journey into the history of text-mining, providing the description of technical means that underpin its appearance and the methods which draw it nearer natural cognitive processes. The abovesaid makes the study unique, for no papers focused on the methodologic and historical background of this rapidly developing technology are available.