University of Twente Student Theses
To what extent do uncertainty, sentiment, and authors influence the amount of anthropomorphic behavior on social media?
Harbers, Thom (2023) To what extent do uncertainty, sentiment, and authors influence the amount of anthropomorphic behavior on social media?
|
PDF
683kB |
| Abstract: | Social media serves as a platform for people to share their thoughts but also provides a place to share knowledge. For certain accounts, people tend to attribute human-like qualities to non-human agents. We are interested in seeing what factors and strategies influence the amount of anthropomorphic behavior elicited, with the goal of creating social entities. We performed an analysis of three different Twitter accounts, which we labeled with different message composition strategies. By analyzing the replies to a multitude of tweets for each account, we examined whether uncertainty and sentiment in a non-human agent’s behavior heightened the amount of anthropomorphic behavior elicited by consumers. A large dataset was investigated using natural language processing techniques. Anthropomorphism was measured by examining the use of pronouns within replies, as these contained a lot of first-person (“I"), second-person (“you"), and third-person pronouns (“he/she"), as well as explicit mentioning of authors’ names. Uncertainty and sentiment of a non-human agent’s behavior resulted in insignificant results with no correlation between them and anthropomorphism. Our findings do suggest that there is a significant difference between the authors from the accounts analyzed (human written vs. computer generated), however, there may have been additional or different factors to the ones analyzed that resulted in this. |
| Item Type: | Essay (Bachelor) |
| Clients: | Unknown organization, Netherlands |
| Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
| Subject: | 54 computer science |
| Programme: | Computer Science BSc (56964) |
| Link to this item: | https://purl.utwente.nl/essays/96120 |
| 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