错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Opinion Mining Hybrid Approach. An Application to Investigate the Users’ Political Positions in Disinformative Echo Chambers

  • Suania Acampa

摘要

Although the definition is not yet conceptually well defined, the image of echo chambers ((Quattrociocchi & Vicini. Misinformation. Guida alla società dell’informazione e della credulità. Franco Angeli, Roma, 2016); (Terren, L., & Borge-Bravo, R., Review of Communication Research 9:99–118, 2021)) aims to convey the idea that people who use social media platforms are exposed largely (or exclusively) to one type of pro-attitudinal content that polarizes their opinions to the point of extremes. This type of exposure raises much concern in the academic community concerning the information consumption, which plays a fundamental role in the construction of public opinion. Despite studies ((Mocanu et al. Computers in Human Behavior 51:1198–1204, 2015); (Bessi et al. Social determinants of content selection in the age of (mis)information. International conference on social informatics (pp. 259–268). Springer, Cham, 2014). (Bessi et al. Science vs conspiracy: Collective narratives in the age of misinformation. PLoS One, 10(2), 2015)) that have contributed to understanding the social dynamics of user behavior in social media in relation to disinformative content, an exclusively data-driven approach seems to reduce the complexity of the disinformation phenomenon to a simple true/false dichotomy and considers individuals and their complex social behaviors as a single subject with undifferentiated behavior. The aim of this work is to use an epistemological theory-driven approach and a hybrid opinion-mining approach to analyze opinions of Italian users who consume disinformative content, and to reconstruct their narratives and worldviews to identify any elements that guide users’ opinions and their experience in disinformative echo chambers. The papers contain part of the results of a broader analysis that was conducted on two sequential levels. The results of the first level will be mentioned, useful for understanding the results that will be seen here which refer to the second level of analysis.