Emotional Valence Coded in the Phonemic Content – Statistical Evidence Based on Corpus Analysis
Abstract
This study investigates the relationship between the phonemic content of texts in English and the emotional valence they inspire. The sublexical content is presented in terms of biphones composed by one vowel and one consonant. The statistical analysis of a vast corpus of emotionally evaluated sentences reveals a strong correlation between this sublexical presentation and the evaluations of valence provided by the readers. An initial test performed with other valence-rated prose texts makes believing that the feature observed within the corpus can be useful for the emotion classification of texts.
© 2020 Velina Slavova, published by Bulgarian Academy of Sciences, Institute of Information and Communication Technologies
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.