Principles of summarization of large amounts of data: features selection of language means

Abstract

Natural language is the main resource of people to communicate and understand phenomena and processes. The accelerated introduction of information and communication technologies in all fields of society promotes the generation of large data volumes which contains words, codes, symbols, and numbers. However, originally only a small part of this data can be understood and used by people. There are computational approaches that, using soft computing techniques and natural language, construct summaries in form of sentences which describe large volumes of data. The purpose of this paper is to briefly describe a method used in linguistic summaries   construction,   emphasizing   the    use    of    natural    language    to offer    relevant and comprehensive information about data. First, a background on the linguistic summarization of data is presented. Then two examples of the application of this technique to different databases are analyzed. In both cases, short sentences in natural language are obtained that offer summarized information about the data. In the second example, the beneficiaries showed a satisfaction index of 0.7 according to Iadov's logical framework.