DOI: 10.17586/2226-1494-2015-15-1-169-171


TEXTS SENTIMENT-ANALYSIS APPLICATION FOR PUBLIC OPINION ASSESSMENT

I. A. Bessmertny, R. V. Posevkin


Read the full article 
Article in Russian

For citation: Posevkin R.V., Bessmertny I.A. Texts sentiment-analysis application for public opinion assessment. Scientific and Technical Journal of Information Technologies, Mechanics and Optics, 2015, vol. 15, no. 1, pp. 169–171

Abstract

The paper describes an approach to the emotional tonality assessment of natural language texts based on special dictionaries. A method for an automatic assessment of public opinion by means of sentiment-analysis of reviews and discussions followed by published Web-documents is proposed. The method is based on statistics of words in the documents. A pilot model of the software system implementing the sentiment-analysis of natural language text in Russian based on a linear assessment scale is developed. A syntactic analysis and words lemmatization are used to identify terms more correctly. Tonality dictionaries are presented in editable format and are open for enhancing. The program system implementing a sentiment-analysis of the Russian texts based on open dictionaries of tonality is presented for the first time.


Keywords: text tonality processing, tonality, sentiment-analysis, natural language text

References

1. Bessmertnyi I.A., Dzhaliashvili Z.O., Maksimov V.V., Markin D.A. Lingvootsenochnoe upravlenie tekstom [Linguistic estimated text management]. Tezisy dokladov X Mezhdunarodnoi konferentsii «Primenenie novykh tekhnologii v obrazovanii» [Abstracts X Int. Conf. Application of new technologies in education]. Troitsk: Fond Novykh Tekhnologii v Obrazovanii «Baitik», 1999.

2. Nugumanova A., Bessmertnyi I. Applying the latent semantic analysis to the issue of automatic extraction of collocations from the domain texts. Communications in Computer and Information Science, 2013, vol. 394, pp. 92–101. doi: 10.1007/978-3-642-41360-5_8

3. Pozel'skaya A.G., Solov'ev A.N. Metod opredeleniya emotsii v tekstakh na russkom yazyke [A method of sentiment analysis in Russian texts]. Tezisy Dokladov Mezhdunarodnoi Konferentsii po Komp'yuternoi Lingvistike i Intellektual'nym Tekhnologiyam «Dialog 2011» [Proc. Int. Conf. Computational Linguistics and Intellectual Technologies]. Moscow, RGGU, 2011, pp. 510–522.

4. Cruz F.L., Troyano J.A., Pontes B., Ortega F.J. Building layered, multilingual sentiment lexicons at synset and lemma levels. Expert Systems with Applications, 2014, vol. 41, no. 13, pp. 5984–5994. doi: 10.1016/j.eswa.2014.04.005

5. Ermakov S.A., Ermakova L.M. Metody otsenki emotsional'noi okraski teksta [Overview of sentiment analysis methods]. Vestnik Permskogo Universiteta. Seriya: Matematika, Mekhanika, Informatika, 2012, no. 1, pp. 85–90.

6. Parau P., Stef A., Lemnaru C., Dinsoreanu M., Potolea R. Using community detection for sentiment analysis. Proc. IEEE 9th International Conference on Intelligent Computer Communication and Processing, ICCP 2013, 2013, pp. 51–54. doi: 10.1109/ICCP.2013.6646080

7. Chiru C.-G., Hadgu A.T. Sentiment-based text segmentation. Proc. 2nd International Conference on Systems and Computer Science, ICSCS 2013, 2013, pp. 234–239. doi: 10.1109/IcConSCS.2013.6632053

8. Minakov I.A. Analiz emotsional'noi tonal'nosti teksta i ego primenenie dlya povysheniya kachestva perekhodov po relevantnym ob"yavleniyam [Sentiment analysis: practical recommendations and its usage for improving user-reaction on relevant online advertisements]. Vestnik Samarskogo Gosudarstvennogo Tekhnicheskogo Universiteta. Seriya: Tekhnicheskie Nauki, 2013, no. 1 (37), pp. 58–63. 

Copyright 2001-2017 ©
Scientific and Technical Journal
of Information Technologies, Mechanics and Optics.
All rights reserved.

Яндекс.Метрика