Sentiment Analysis on Indonesian Political Hoaxes
- 10.2991/assehr.k.200729.004How to use a DOI?
- sentiment analysis, hoax, appraisal theory, emotion in discourse
Distinguishing between hoaxes and real news from a linguistic perspective requires further identification than can be provided by structural analysis. The study of emotions and sentiments contained in the text is also important, since these can indicate the author’s mental state, rhetorical position, attitude, judgment, and relationship with an object or event. This study aimed to analyze how emotions and sentiments emerge and play a role in hoaxes, employing qualitative methods and the appraisal theory framework of Martin and White . Data were limited to five hoax texts, with five news texts from official sources used for comparative analysis. All texts were political in nature. Analysis was conducted using qualitative methods, such as annotation, description, interpretation, and comparison between kinds of text. The study found that (1) hoax texts are dominated by negative sentiments with strong semantics, (2) hoax texts tend to be affective and judgmental, and (3) hoax authors try to write texts as similar as possible to real news, often using a heterogloss voice to convey statements. When a monogloss voice is used, an attribute embedding process is dominant. These findings indicate that emotions and sentiments play a significant role in hoax claims and that appraisal theory can address deeper and broader aspects of sentiment analysis in texts.
- © 2020, the Authors. Published by Atlantis Press.
- Open Access
- This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
Cite this article
TY - CONF AU - Aghnia Salsabila AU - Totok Suhardijanto PY - 2020 DA - 2020/07/30 TI - Sentiment Analysis on Indonesian Political Hoaxes BT - Proceedings of the International University Symposium on Humanities and Arts (INUSHARTS 2019) PB - Atlantis Press SP - 15 EP - 21 SN - 2352-5398 UR - https://doi.org/10.2991/assehr.k.200729.004 DO - 10.2991/assehr.k.200729.004 ID - Salsabila2020 ER -