University of Twente Student Theses
Exploring the relationship between linguistic alignment and sentiment - in online discussions
Wentzel, S.D. (2023) Exploring the relationship between linguistic alignment and sentiment - in online discussions.
|
PDF
46MB |
| Abstract: | In conversations, people often converge on their linguistic behavior. This linguistic alignment has been researched in combination with many different factors, but a gap was found in studying the relationship between linguistic alignment and sentiment. This thesis aimed to find such a relationship. Specifically, the Internet Argument Corpus was investigated. Linguistic alignment on the lexical level was investigated with a new measure: the time-based overlap. The alignment scores over time of discussions were clustered, resulting in groups with similar patterns of lexical alignment behaviors. Sentiment was investigated with VADER and the scores over time were also clustered to find patterns. These results were combined and correlation and association coefficients were computed to find how sentiment and alignment relate, along with contingency tables. Additionally, associations with discussion length and topic were investigated as the previous results hinted that these might be involved. Results showed no clear association between sentiment and alignment, but some between alignment and topic, and even more between sentiment and topic. Results also indicate that the length of discussions might be related. This implies that if sentiment and linguistic alignment interplay, the correlation is complex and likely depends on other factors, such as this discussion length and topic. |
| Item Type: | Essay (Master) |
| Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
| Subject: | 17 linguistics and theory of literature, 54 computer science |
| Programme: | Interaction Technology MSc (60030) |
| Link to this item: | https://purl.utwente.nl/essays/97481 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page