Sentiment in Academic Texts

• Published in: Proceedings of the XXth Conference of Open Innovations Association FRUCT pp. 408 - 414
• Main Authors: Solovyev, Valery, Solnyshkina, Marina, Gafiyatova, Elzara, McNamara, Danielle, Ivanov, Vladimir
• Format: Conference Proceeding
• Language: English
• Published: FRUCT 01.04.2019
• Subjects: Data mining; Dictionaries; Education; History; Mood; Sentiment analysis
• ISSN: 2305-7254
• DOI: 10.23919/FRUCT.2019.8711900

The problem of sentiment analysis has been widely studied in the past several decades. The research in the area has been predominantly based on data collated from online messages, microblogs, reviews, etc. Significantly fewer studies have been conducted based on academic discourse and especially school textbooks. However, sentiment analysis of academic texts can help answer pressing issues relating the ways in which different referents are presented in contemporary Russian school textbooks. In this paper, we analyze the distribution of sentiment words and phrases in a Corpus of Russian school textbooks on History (Grades 10-11) and Social Sciences (Grades 5-11). The results of the study demonstrate that the discourse within (1) History textbooks used in the 10th and 11th grades of Russian schools and (2) Social Studies textbooks written by Nikitin for Russian schools (Grades 5-11) contains predominantly negative sentiment: the writers select negatively valenced words and prefer presenting negative referents. By contrast, the discourse within the set of Social Studies textbooks written by Bogolubov revealed a predominantly positive bias. The authors discuss the implications of these trends in relation to the potential impact of the tone of educational discourse on learning.