tv.um.es

Nowadays corpora do not exist without statistics and quantitative approaches to linguistic material as well as colloquial speech doesn`t exist without the expression of psychoemotional states of speaker in it. That is why it seems appropriate to study the statistics of psycholinguistic correlates of various features of oral speech exclusively on corpus material.

In this paper, attention was focused on the search for correlations (primary dependencies) between the use of pragmatic markers (PM) of various types[1] (hesitatives, metacommunicatives, reflexives, discursive / delimiters, substituents, xeno-indicators, etc.) and the psychological characteristics of the speaker (extraversion / introversion grade). The study of the mentioned connection by statistical methods was based on the sound corpus of the Russian language of everyday communication “One Day of Speech” (ORD) in text annotations and audio recordings (to learn more, see: [2],[5],[6]).

The analysis of data was carried out using a set of Data Mining methods [4], the essence of which is the process of detecting in “raw” data new interpretations of knowledge, using methods of mathematical statistics and important for making decisions in various spheres of human activity. The dataset in this study was collected using all labeled PM and metadata of each informant of the sample. In this case, the main task was to search for associative rules [4] - private dependencies (or associations) between the PM and the psychological type of informant.

The psychological type of informant was automatically determined based on a comparison of the results of two psychological tests [3]. To analyze the material (3661 units in contexts, 300,000 tokens), the Python 3.6 programming language, the Jupyter Notebook development environment, and the Pandas and Numpy libraries were used.

An analysis of correlations revealed a number of signs associated with the use of pragmatic markers, which, with further expansion of the material, can become significant (diagnostic) for determining a person’s psychotype by the characteristics of the speech.
Acknowledgements
The presented research was supported by the Russian Science Foundation, project #18-18-00242 “Pragmatic Markers in Russian Everyday Speech”.