Sentiment Analysis of Polish Online News Covering Controversial Topics : Comparison Between Lexicon and Statistical Approaches

Szwoch, Joanna; Staszkow, Mateusz; Rzepka, Rafal et al.

Human Language Technologies as a Challenge for Computer Science and Linguistics, 2024, 2023, 277-281

Permalink : https://hdl.handle.net/2115/95945

このアイテムのアクセス数:107件（2026-08-10 22:56 集計）

閲覧可能ファイル 

ファイル フォーマット サイズ 閲覧回数 説明
Proc. Hum. Lang. Technol. Chall. Comput. Sci. Linguist. 2023_277-281 pdf 901 KB 97

論文情報

ファイル出力 EndNote Basic出力 Mendeley出力

アクセス権
open access
URI
https://hdl.handle.net/2115/95945
タイトル
Sentiment Analysis of Polish Online News Covering Controversial Topics : Comparison Between Lexicon and Statistical Approaches
著者
Szwoch, Joanna ; Staszkow, Mateusz ; Rzepka, Rafal ; Araki, Kenji
著者名
Szwoch, Joanna
著者名
Staszkow, Mateusz
e-Rad_Researcher
80396316
著者名
Rzepka, Rafal
e-Rad_Researcher
50202742
著者名
Araki, Kenji
言語
英語
キーワード
Polish language
online news
sentiment analysis
発行日
2024-02-06
収録物名
Human Language Technologies as a Challenge for Computer Science and Linguistics
巻(号)
2023
ページ
277-281
抄録
In this paper we present results of a sentiment analysis performed on a part of the Polish Online News Corpus which consists of articles covering controversial topics in Poland. We perform both lexicon-based and statistical experiments and compare the obtained results for this task. The main goal of our comparative work is to discover whether supposedly objective online news contain any emotive content and how different the results are, depending on the approach. Our study found no meaningful differences in emotional load at the word level, but the multilingual language model trained with the Natural Language Inference objective has partially confirmed its existence. We believe this work can be used for further research on media polarization in Polish language.
資源タイプ
会議発表論文
出版タイプ
Version of Record
権利情報
Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
関連情報
ISBN
978-83-232-4177-5
関連情報 (isIdenticalTo)
URI
https://researchportal.amu.edu.pl/info/book/UAM5f68a911d2e44a1e877c916692331b7b/