SENTIMENT ANALYSIS: PRAGMATIC SPECIFICITY

UDC 551. 81’33

 

Barkovich Aleksandr Arkad’yevich – DSc (Philology), Associate Professor, Head of the Department of Informatics and Applied Linguistics. Minsk State Linguistic University (21, Zakharova str., 220034, Minsk, Republic of Belarus). E-mail: barkovichaa@gmail.com

Petrova Natal’ya Sergeevna – Lecturer. Minsk State Linguistic University (21, Zakharova str., 220034, Minsk, Republic of Belarus). E-mail: natali040100@gmail.com

 

DOI: https://doi.org/ 10.52065/2520- 6729-2023-273-2-6 (In Russian).

 

Key words: sentiment analysis, pragmatics, methodology, text tonality, tone dictionary, machine learning.

For citation: Barkovich A. A., Petrova N. S. Sentiment analysis: pragmatic specificity. Proceedings of BSTU, issue 4, Print- and Mediatechnologies, 2023, no. 2 (273), pp. 40–46. DOI: 10.52065/2520- 6729-2023-273-2-6 (In Russian).

 

Abstract

The article considers the pragmatic aspects of the practical implementation of the main methods of sentiment analysis. The introduction outlines the specifics of sentiment analysis as an actual direction in the development of computational linguistics. The focus of the study is the pragmatics of implementing the methodological potential of the text tonality evaluation. In the section devoted to methods for the text tonality evaluation, a description is given of the priorities of sentiment analysis: expediency and effectiveness. The methods of sentiment analysis and the underlying models are described. The section devoted to the results of the study and their evaluation reflects the specifics of the use of methods of tone dictionaries, machine learning and machine learning using tone dictionaries. The illustrative material is presented by a fragment of the English-language media discourse, representative for a detailed study of the level specifics of sentiment analysis. The study was based on a generally accepted categorical foundation, including that covering the polarity, intensity and magnitude of the text tonality. In addition to the traditional lexical-grammatical paradigm of text research, due attention was paid to such innovative linguistic categories as entities and keywords. Such categories of metadata as brands, geopolitical realities, nationalities and a number of others were identified. The analysis is supported by detailed statistical data. In general, the novelty of this study is provided by the linguistic analysis of informational procedures and methodological reflection of interdisciplinary issues. Equally significant is the consideration of the pragmalinguistic aspects of the text tonality evaluation and the actualization of new knowledge in this regard. In particular, a factually substantiated comparison of the main methods of sentiment analysis made it possible to draw meaningful conclusions about their advantages and drawbacks. The most reliable metadata about the tonality of the studied document was obtained using machine learning techniques. The explication of relevant data was carried out by software tools specially designed for evaluating the text tonality. The results of this study will be relevant for optimizing the sentiment analysis procedure at all its stages and implementing the relevant metadata and knowledge in linguistic and informational practices.

 

Download

References

  1. Liu B., Hu M., Cheng J. Opinion Observer: Analyzing and Comparing Opinions on the Web. WWW– 2005. Proceedings of the 14th International World Wide Web conference, May 10–14, 2005, Chiba, Japan. New York. 2005, pp. 342–351.
  2. Pang B., Lee L. Opinion Mining and Sentiment Analysis. Foundations and Trends in Information Retrieval, 2008, vol. 2, no. 1–2, pp. 1–135.
  3. Feldman R. Techniques and applications for sentiment analysis. Communications of the ACM. 2013, vol. 56, no. 4, pp. 82–89.
  4. Handbook of Natural Language Processing / eds. N. Indurkhya, F. J. Damerau. Boca Raton: Chapman & Hall / CRC, 2010. 702 p.
  5. Barkovich A. Informational Linguistics: The New Communicational Reality Newcastle upon Tyne: Cambridge Scholars Publ., 2020. 271 p.
  6. Barkovich A. A. Internet-diskurs: komp’yuterno-oposredovannaya kommunikatsiya [Internet discourse: computer-mediated communication]. Moscow, FLINTA Publ.; Nauka Publ., 2015. 288 p. (In Russian).
  7. Reagan A. J., Danforth C. M., Tivnan B., Williams J. R., Dodds P. S. Sentiment analysis methods for understanding large-scale texts: a case for using continuum-scored words and word shift graphs. EPJ Data Science, 2017, 6.
  8. Poria S., Hazarika D., Majumder N., Mihalcea R. Beneath the tip of the iceberg: current challenges and new directions in sentiment analysis research. IEEE Transactions on Affective Computing, 2020, no. 14(1), pp. 108–132.
  9. Rana M. R. R., Nawaz A., Iqbal J. A survey on sentiment classification algorithms, challenges and applications. Acta Universitatis Sapientiae, Informatica, 2018, no. 10(1), pp. 58–72.
  10. Pazel’skaya A. G., Solov’ev A. N. A method for determining emotions in texts in Russian. Komp’yuternaya lingvistika i intellektual’nyye tekhnologii: ezhegodnaya Mezhdunarodnaya konferentsiya “Dialog” [Computational Linguistics and Intelligent Technologies: annual International Conference “Dialogue”]. Moscow, 2011, issue 10, pp. 510–522 (In Russian).
  11. Taboada M., Brooke J., Tofiloski M., Voll K., Stede M. Lexicon-based methods for sentiment analysis. Computational linguistics, 2011, no. 37(2), pp. 267–307.
  12. Kulagin D. I. Open Tonal Dictionary of the Russian Language KartaSlovSent. Komp’yuternaya lingvistika i intellektual’nyye tekhnologii: materialy ezhegodnoy Mezhdunarodnoy konferentsii “Dialog” [Computational Linguistics and Intelligent Technologies: materials of the annual International Conference “Dialogue”]. Moscow, 2021, issue 20, pp. 1106–1119 (In Russian).
  13. Barkovich A. A. Trends in the development of computer-mediated discourse: metalinguistic aspect Vestnik Tverskogo gosudarstvennogo universiteta [Bulletin of the Tver State University], series Philology, 2015, no. 4, pp. 108–114 (In Russian).
  14. Minsky M. The Emotion Machine: Commonsense Thinking, Artificial Intelligence, and the Future of the Human Mind. New York: Simon & Schuster, 2007. 400 p.
  15. Urban Dictionary. Available at: https://www.urbandictionary.com/define.php?term=Lettuce (accessed 13.05.2023).

26.06.2023