INTERACTION OF DISCOURSE ANALYSIS AND SENTIMENT ANALYSIS TO IDENTIFY EMOTIONAL STATE IN TEXT COMMUNICATION
DOI:
https://doi.org/10.34185/1991-7848.itmm.2024.01.061Keywords:
emotion identification, emotion recognition, sentiment analysis, discourse analysis, neuro-linguistic programming.Abstract
This paper is devoted to the study of two fundamental tasks of neuro-linguistic programming: discourse analysis and sentiment analysis, which are two fundamental tasks of natural language processing. The paper examined some models and algorithms for improving data processing in the process of textual communication between users. The study showed that the use of a joint model of discourse analysis and sentiment analysis is mutually beneficial. The results show that the information obtained from the discourse can help in determining the mood, and sentiment analysis and knowledge of two text fragments can help determine the discourse relations between them.
References
Dmytriieva I.S., Bimalov D.V. Identyfikatsiia emotsiinykh sliv u tekstovomu spilkuvanni / Informatsiini tekhnolohii v metalurhii ta mashynobuduvanni. ITMM2023: tezy dopovidei mizhnarodnoi naukovo-praktychnoi konferentsii (Dnipro, 22 bereznia 2023 r.) – Dnipro: UDUNT, 2023. – s. 290-293.
Dmytriieva I.S., Bimalov D.V. Rozrobka prohramnoho moduliu dlia identyfikatsiia emotsiinoho stanu korystuvacha // Systemni tekhnolohii. Rehionalnyi mizhvuzivskyi zbirnyk naukovykh robit. - Vyp. 4(147). - Dnipro: UDUNT, 2023. – s. 29-34
DOI 0.34185/1562-9945-4-147-2023-03
Lazaridou A., Titov I., Sporleder C. A bayesian model for joint unsupervised induction of sentiment, aspect and discourse representations. In ACL (1), pages 1630–1639, 2013.
