Memetic suggestions: A comparative sentiment analysis of multimodal and textual posts
| Authors | |
|---|---|
| Year of publication | 2024 |
| Type | Appeared in Conference without Proceedings |
| MU Faculty or unit | |
| Citation | |
| Description | This linguistic study focuses on the use of memes on social media and their utilization in the posters’ representation and the reception they receive. The study looks at social media posts regarding certain polarizing topics – both textual and memetic – and, using sentiment analysis, determines the answer to the question: Does usage of memetic elements improve the sentiments of the audience towards the more positive? The research takes into account a number of posts gathered from social media platforms, such as Facebook and X, and compares pairs by the same poster on the same topic in order to retain a similar baseline of presumed popularity of the contribution. The study then looks into how these pairs (of which one is textual and one memetic) are perceived by the public – if the memetic element evokes more positive responses than the textual one. A sample size of the reactions to these posts are subjected to sentiment analysis using both artificial intelligence tools and subsequent manual qualitative evaluation of the post by the author regarding its contextual cues and accuracy of the AI’s determined sentiment. The overall sentiment values, extracted from a few such pairs of posts, then portray whether the tactic of using memetic elements to engage the crowd and skew the opinion of polarizing topics is successful in these instances. By a mixed qualitative and quantitative examination of the data, the project gives insight into representation manipulation on social media and the power of memetic elements as communication units in digital environments. The study thus observes how the paths of textual and multimodal communication diverge in their effects on the audience and draws conclusions from the results. |
| Related projects: |