DOI: 10.17587/prin.9.76-81

Scientific Text Analysis and New World Trends

A. Bernadotte, alexandra.bernadotte@gmail.com, Lomonosov Moscow State University, Moscow, 119991, Russian Federation; Department of Medical Biochemistry and Biophysics, Karolinska Institutet, S-171 77, Stockholm, Sweden

Corresponding author: Bernadotte Alexandra, MD, PhD in Medicine, PhD candidate in Mathematics, Lomonosov Moscow State University, 119991, Moscow, Russian Federation; Adjunct Research Assistant Professor at Department of Medical Biochemistry and Biophysics, Karolinska Institutet, Stockholm, S-171 77, Sweden, E-mail: alexandra.bernadotte@gmail.com

Many of us of different fields and branches who are diving deep in science can notice slight or even quite sharp changes in the scientific world. The changes touch upon every part of science including semantics, methodology, axiomatic base, and objectives. From our experience one of the current scientific trends is a shift from theoretical science research to empirical research. The second trend is directly connected with the first one; there is a transformation from the basic science into an applied one, with a rapid transition to the commercial area. Being scientists, we also can notice marks of politicization and globalization in the scientific world. To transform our vague senses and feelings into the scientifically-recognizable form, we analyzed scientific papers published during the period of 20 years in frames of the top scientific journal — Science. To reach theory-practice equilibrium shift and dominance of commercialization and politicization we analyzed text data, assuming that the loss of interest of theoretical and basic knowledge correlates with the usage of the certain words and phrases in scientific papers, as well as changes in their meaning and usages. Indeed, recent work has demonstrated a 20-years transformation of scientific semantics and scientific interests. This paper has shown the growing commercialization and politicization of science, and reflected the current primacy of the application of knowledge upon basic science. This work also introduces a new word-classes model of document clustering algorithms based on preliminarily words clustering with a usage of ontology-based term similarity, which can be helpful for text data mining and recommendation system.

Keywords: text analysis, semantic analysis, tf-idf, scientific text, cognition, scientific trends, text clustering, data mining

pp. 76–81

For citation:
Bernadotte A. Scientific Text Analysis and New World Trends, Programmnaya Ingeneria, 2018, vol. 9, no. 2, pp. 76—81. DOI: 10.17587/prin.9.76-81