STATISTICAL SOFTWARE R IN CORPUS-DRIVEN RESEARCH AND MACHINE LEARNING

The rapid development of computer software and network technologies has facilitated the intensive application specialized statistical not only in traditional information technology spheres (i.e., statistics, engineering, artificial intelligence) but also linguistics. R is one most popular analytical tools for processing a huge array digitalized language data, especially quantitative corpus linguistic studies Western Europe North America. This article discusses functionality package R, focusing on its advantages performing complex analyses data corpus-driven creating classifiers machine learning. With this mind, three-stage strategy computer-statistical analysis elaborated: 1) preparing to be subjected procedure, 2) utilizing hypothesis testing methods (MANOVA, ANOVA) Tukey post-hoc test, 3) developing model classifier analyzing effectiveness. implemented 11 000 tokens English detached nonfinite constructions with an explicit subject extracted from BNC-BYU corpus. indicates significant differences realization factors parameter “Part speech subject”. analyzed are employed build classification given constructions. Particular attention devoted methodological perspectives interdisciplinary research fields linguistics studies. potential elaborated case study training undergraduate, master, postgraduate students Applied Linguistics indicated. provides all codes written script comprehensive descriptions explanations. concluding part summarizes obtained results highlights issues further connected popularization raising awareness specialists system.