The final, formatted version of the article will be published soon.
ORIGINAL RESEARCH article
Front. Psychol.
Sec. Emotion Science
Volume 15 - 2024 |
doi: 10.3389/fpsyg.2024.1401084
This article is part of the Research Topic Emotions and Artificial Intelligence View all 12 articles
Emotion Topology: Extracting Fundamental Components of Emotions from text using Word Embeddings
Provisionally accepted- 1 Polish Academy of Sciences, Warsaw, Poland
- 2 University of Warsaw, Warsaw, Masovian, Poland
This exploratory study examined the potential of word embeddings, an automated numerical representation of written text, as a novel method for emotion decomposition analysis. Drawing from a substantial dataset scraped from a Social Media site, we constructed emotion vectors to extract the dimensions of emotions, as annotated by the readers of the texts, directly from human language. Our findings demonstrated that word embeddings could yield emotional components akin to those found in previous literature, offering an alternative perspective not bounded by theoretical presuppositions, as well as showing that the dimensional structure of emotions is reflected in the semantic structure of their text-based expressions. Our study highlights word embeddings as a promising tool for uncovering the nuances of human emotions and comments on the potential of this approach for other psychological domains, providing a basis for future studies. The exploratory nature of this research paves the way for further development and refinement of this method, promising to enrich our understanding of emotional constructs and psychological phenomena in a more ecologically valid and data-driven manner.
Keywords: word embeddings 1, emotion decomposition 2, natural language processing 3, valence4, Arousal5
Received: 14 Mar 2024; Accepted: 03 Sep 2024.
Copyright: © 2024 Plisiecki and Sobieszek. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence:
Hubert W. Plisiecki, Polish Academy of Sciences, Warsaw, Poland
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.