CALL FOR PROPOSALS
Computational Stylistics Workshop on Emotion and Sentiment Analysis in Literature
Paris, June 16-17 2022
Generally speaking, sentiment analysis is used in natural language processing to encompass all research studies dealing with opinions, sentiments or emotion of an individual or a community, as can be seen from the title of this book : Sentiment Analysis: Mining Opinions, Sentiments, and Emotions (Liu, 2015). As far as computational literary studies are concerned, Kim and Klinger (2021) argue that whereas sentiment analysis is mainly focused on text polarity (positive or negative), emotion analysis rather seeks to identify an affective state. Since literature has the unique ability to arouse strong feelings in readers, it is therefore not surprising that computational tools have paved the way for new methods for studying characterisation, narratology and style. In this respect, it is possible to distinguish a wide array of computational approaches to sentiment analysis in literary texts focusing on the one hand on the emotional trajectory of the story itself (Reagan et al., 2016 ; Schmidt, 2019), its main protagonists (Nalisnick and Baird, 2013a, 2013b ; Yavuz, 2020 ) or the relationships they share (Jhavar and Mirza, 2018), and on the other hand, on the evolution of the vocabulary of emotions in literary texts (Rastier, 1995 ; Mohammad, 2012 ; Leemans et al., 2017), in some cases in relation to specific places (Heuser et al, 2016). See Kim and Klingler (2018) for a detailed survey.
In practice, depending on the time span of the analysed corpus, the automatic detection of sentiment in texts has mostly relied on existing or customised lexicons. However, in recent years, an increasing number of computational tools specifically geared towards literary texts have been developed, such as SentiArt (Jacobs, 2019) which uses artificial intelligence techniques and SentText (Schmid et al., 2021). Consequently, unlike other texts, literary texts seem to possess some linguistic specificities that can only be taken into account by adapting existing sentiment analysis methods. For instance, with regards to the great complexity of literary language, Gius et al. (2020) propose to focus primarily on interesting text segments without attempting to assess the polarity of words denoting sentiments. Nevertheless, several domains intrinsically linked with emotions in literary texts have scarcely been explored with computational tools: the author’s stylistic choices when conveying emotions or even sentiment detection in relation to both literary criticism and readers’ response. More specifically, as it has been of great interest to linguists and narratologists in recent years, readers’ response to literary texts could potentially enable to better understand the diversity of the reading experience which mixes self-appropriating the emotions described in the text, identifying to a certain degree with the characters/narrator and feeling suspense, curiosity or surprise (Baroni, 2007).
Following previous initiatives such as the 2019 Montpellier conference “Questioning the Text in the Era of ‘Mechanical Intelligence’: Digital Stylistics between Disciplinary and Interdisciplinary studies” and the 2021 lecture series on Sentiment Analysis in Literary Studies, this workshop intends to enable researchers to share their experience on sentiment analysis in literary texts and also to foster new approaches in the field of digital literary studies, mainly in relation the following topics:
• Generic lexicon and sentiment detection in literature;
• Visualising emotions in literature;
• Emotion and style (metaphors, repetitions, clichés…);
• The history of emotions in literature;
• Emotions in popular literature;
• Literary critics and readers’ opinions on literary texts and authors;
• Affect and characters’ gestures or more generally characters’ description;
• The relationship between emotions/sentiments and literary genres;
• The identification the causes of emotions/sentiments;
• The reader’s emotions.
Submission Procedure
Proposals (2 pages, i.e. between 600 and 1,000 words) in English or in French shall be sent before March 1, 2022 both to Dominique Legallois (dominique.legallois@sorbonne-nouvelle.fr) and Suzanne Mpouli (suzanne.mpouli@u-paris.fr).
Authors will be notified of a decision on their proposal in mid-April.
Organizers
Dominique Legallois (Université Sorbonne Nouvelle, Lattice)
Suzanne Mpouli (Université de Paris, Centre des Humanités Numériques – Direction Générale Déléguée des Bibliothèques et Musées)
Practical Information
The workshop will be held in a hybrid format and participants can choose to present in English or in French. The event is free of charge and is endorsed by the SIG Digital Literary Stylistics.
Scientific Committee
Jean-Baptiste Camps, École Nationale des Chartes
Francesca Frontini, Istituto di Linguistica Computazionale “A. Zampolli”
Ioana Galleron, Université Sorbonne Nouvelle
Berenike Herrmann, University of Bielefeld
Olivier Kraif, Université de Grenoble Alpes
Thierry Poibeau, Université Sorbonne Nouvelle
Glenn Roe, Sorbonne Université
References
Baroni, R. (2007). La Tension Narrative . Paris: Seuil.
Gius, E., Murawska, A., Schmidt, O., Sökefeld, C. & Vauth, M. (2020). “Sentiment Sensitivity. Using Sentiment Analysis in Literary Studies to Analyze Genre and the Depiction of Illness”, Book of Abstracts Digital Humanities 2020 .
Heuser, R., Moretti, F. & Steiner E. (2016). “The Emotions of London”. Stanford Literary Lab Pamphlets, 13. https://litlab.stanford.edu/LiteraryLabPamphlet13.pdf
Jacobs, A. M. (2019). “Sentiment Analysis for Words and Fiction Characters from the Perspective of Computational (Neuro-)Poetics”, Frontiers in Robotics and AI, 6 (53).
Jhavar, H. a& Mirza, P. (2018). “EMOFIEL: Mapping Emotions of Relationships in a Story”, Companion Proceedings of the Web Conference 2018 , pp. 243–246.
Kim E. & Klingler, R. (2021). “A Survey on Sentiment and Emotion Analysis for Computational Literary Studies”, Zeitschrift für digitale Geisteswissenschafte n, doi: 10.17175/2019_008_v2
Leemans, I., van der Zwaan, J. M., Maks, I., Kuijpers, E. & Steenbergh, K. (2017). “Mining Embodied Emotions: A Comparative Analysis of Sentiment and Emotion in Dutch Texts, 1600-1800”, Digital Humanities Quarterly , 11(4).
Mohammad, S. M. (2012). “From Once Upon a Time to Happily Ever After: Tracking Emotions in Mail and Books”, Decision Support Systems 53 (4), pp. 730–741.
Nalisnick, E. T. & Baird, H. S. (2013a). “Character-to-character Sentiment Analysis in Shakespeare’s Plays”, Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers) , pp. 479-483.
Nalisnick, E. T. & Baird, H. S. (2013b). “Extracting Sentiment Networks from Shakespeare’s Plays”, 2013 12th International Conference on Document Analysis and Recognition , pp. 758-762.
Novakova, I. & Siepmann, D. (Eds.). (2020). Phraseology and Style in Subgenres of the Novel , London: Palgrave Macmillan.
Rastier, F., ed. (1995). L’Analyse Thématique des Données Textuelles : L’Exemple des Sentiments . Paris: Didier.
Reagan, A. J., Mitchell, L., Kiley, D., Danforth, C. M. & Dodds, P. S. (2016). “The Emotional Arcs of Stories are Dominated by Six Basic Shapes”, EPJ Data Science , 5(1), pp. 1-12.
Schmidt, T. (2019). “Distant Reading Sentiments and Emotions in Historic German Plays”, Abstract Booklet DH Budapest 2019 , pp. 57-60.
Schmidt, T., Dangel, J. & Wolff, C. (2021). “SentText: A Tool for Lexicon-based Sentiment Analysis in Digital Humanities”, Information between Data and Knowledge. Information Science and its Neighbors from Data Science to Digital Humanities. Proceedings of the 16th International Symposium of Information Science , pp. 156-172.
Yavuz, M. C. (2010). “Analyses of Character Emotions in Dramatic Works by using EmoLex Unigrams”, Proceedings of the Seventh Italian Conference on Computational Linguistics, CLiC-it’20 .