Sofie Labat
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Sofie Labat has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- Thomas Demeester
Ghent University, Ghent, Belgium
4 shared publications - Véronique Hoste
LT3, Ghent University, Ghent, Belgium
4 shared publications - Benedikt Szmrecsanyi
KU Leuven
1 shared publication Els Lefever
1 shared publication- Haidee Kotze
Utrecht University
1 shared publication - Quanqi Du
Language and Translation Technology Team Ghent University Groot-Brittanniëlaan 45 9000 Ghent Belgium
1 shared publication
Latest Publications
- Open Access
A Customer Journey in the Land of Oz: Leveraging the Wizard of Oz Technique to Model Emotions in Customer Service Interactions
Emotion-aware customer service needs in-domain conversational data, rich annotations, and predictive capabilities, but existing resources for emotion recognition are often out-of-domain, narrowly labeled, and focused on post-hoc detection. To address this, we conducted a controlled Wizard of Oz (WOZ) experiment to elicit interactions with targeted affective trajectories. The resulting corpus, EmoWOZ-CS, contains 2,148 bilingual (Dutch-English) written dialogues from 179 participants across commercial aviation, e-commerce, online travel agencies, and telecommunication scenarios. Our contributions are threefold: (1) Evaluate WOZ-based operator-steered valence trajectories as a design for emotion research; (2) Quantify human annotation performance and variation, including divergences between self-reports and third-party judgments; (3) Benchmark detection and forward-looking emotion inference in real-time support. Findings show neutral dominates participant messages; desire and gratitude are the most frequent non-neutral emotions. Agreement is moderate for multilabel emotions and valence, lower for arousal and dominance; self-reports diverge notably from third-party labels, aligning most for neutral, gratitude, and anger. Objective strategies often elicit neutrality or gratitude, while suboptimal strategies increase anger, annoyance, disappointment, desire, and confusion. Some affective strategies (cheerfulness, gratitude) foster positive reciprocity, whereas others (apology, empathy) can also leave desire, anger, or annoyance. Temporal analysis confirms successful conversation-level steering toward prescribed trajectories, most distinctly for negative targets; positive and neutral targets yield similar final valence distributions. Benchmarks highlight the difficulty of forward-looking emotion inference from prior turns, underscoring the complexity of proactive emotion-aware support.
Open AccessA Customer Journey in the Land of Oz: Leveraging the Wizard of Oz Technique to Model Emotions in Customer Service Interactions
Emotion-aware customer service needs in-domain conversational data, rich annotations, and predictive capabilities, but existing resources for emotion recognition are often out-of-domain, narrowly labeled, and focused on post-hoc detection. To address this, we conducted a controlled Wizard of Oz (WOZ) experiment to elicit interactions with targeted affective trajectories. The resulting corpus, EmoWOZ-CS, contains 2,148 bilingual (Dutch-English) written dialogues from 179 participants across commercial aviation, e-commerce, online travel agencies, and telecommunication scenarios. Our contributions are threefold: (1) Evaluate WOZ-based operator-steered valence trajectories as a design for emotion research; (2) Quantify human annotation performance and variation, including divergences between self-reports and third-party judgments; (3) Benchmark detection and forward-looking emotion inference in real-time support. Findings show neutral dominates participant messages; desire and gratitude are the most frequent non-neutral emotions. Agreement is moderate for multilabel emotions and valence, lower for arousal and dominance; self-reports diverge notably from third-party labels, aligning most for neutral, gratitude, and anger. Objective strategies often elicit neutrality or gratitude, while suboptimal strategies increase anger, annoyance, disappointment, desire, and confusion. Some affective strategies (cheerfulness, gratitude) foster positive reciprocity, whereas others (apology, empathy) can also leave desire, anger, or annoyance. Temporal analysis confirms successful conversation-level steering toward prescribed trajectories, most distinctly for negative targets; positive and neutral targets yield similar final valence distributions. Benchmarks highlight the difficulty of forward-looking emotion inference from prior turns, underscoring the complexity of proactive emotion-aware support.
- 21 May 2025
- Published by Springer Nature in Language Resources and Evaluation
- p. 1-36
UniC: a dataset for emotion analysis of videos with multimodal and unimodal labels
- 21 May 2025
- Published by Springer Nature in Language Resources and Evaluation
- p. 1-36
UniC: a dataset for emotion analysis of videos with multimodal and unimodal labels
- Open Access
- 8 December 2023
- Published by Springer Nature in Language Resources and Evaluation
- Vol. 58 (2) , 505-546
EmoTwiCS: a corpus for modelling emotion trajectories in Dutch customer service dialogues on Twitter
Due to the rise of user-generated content, social media is increasingly adopted as a channel to deliver customer service. Given the public character of online platforms, the automatic detection of emotions forms an important application in monitoring customer satisfaction and preventing negative word-of-mouth. This paper introduces EmoTwiCS, a corpus of 9489 Dutch customer service dialogues on Twitter that are annotated for emotion trajectories. In our business-oriented corpus, we view emotions as dynamic attributes of the customer that can change at each utterance of the conversation. The term ‘emotion trajectory’ refers therefore not only to the fine-grained emotions experienced by customers (annotated with 28 labels and valence-arousal-dominance scores), but also to the event happening prior to the conversation and the responses made by the human operator (both annotated with 8 categories). Inter-annotator agreement (IAA) scores on the resulting dataset are substantial and comparable with related research, underscoring its high quality. Given the interplay between the different layers of annotated information, we perform several in-depth analyses to investigate (i) static emotions in isolated tweets, (ii) dynamic emotions and their shifts in trajectory, and (iii) the role of causes and response strategies in emotion trajectories. We conclude by listing the advantages and limitations of our dataset, after which we give some suggestions on the different types of predictive modelling tasks and open research questions to which EmoTwiCS can be applied. The dataset is made publicly available athttps://lt3.ugent.be/resources/emotwics.
- 8 December 2023
- Published by Springer Nature in Language Resources and Evaluation
- Vol. 58 (2) , 505-546
Open AccessEmoTwiCS: a corpus for modelling emotion trajectories in Dutch customer service dialogues on Twitter
Due to the rise of user-generated content, social media is increasingly adopted as a channel to deliver customer service. Given the public character of online platforms, the automatic detection of emotions forms an important application in monitoring customer satisfaction and preventing negative word-of-mouth. This paper introduces EmoTwiCS, a corpus of 9489 Dutch customer service dialogues on Twitter that are annotated for emotion trajectories. In our business-oriented corpus, we view emotions as dynamic attributes of the customer that can change at each utterance of the conversation. The term ‘emotion trajectory’ refers therefore not only to the fine-grained emotions experienced by customers (annotated with 28 labels and valence-arousal-dominance scores), but also to the event happening prior to the conversation and the responses made by the human operator (both annotated with 8 categories). Inter-annotator agreement (IAA) scores on the resulting dataset are substantial and comparable with related research, underscoring its high quality. Given the interplay between the different layers of annotated information, we perform several in-depth analyses to investigate (i) static emotions in isolated tweets, (ii) dynamic emotions and their shifts in trajectory, and (iii) the role of causes and response strategies in emotion trajectories. We conclude by listing the advantages and limitations of our dataset, after which we give some suggestions on the different types of predictive modelling tasks and open research questions to which EmoTwiCS can be applied. The dataset is made publicly available athttps://lt3.ugent.be/resources/emotwics.
- 23 October 2023
- Published by John Benjamins Publishing Company
- p. 250-276
Processing and prescriptivism as constraints on language variation and change
We investigate the choice between the relative markers which and that in 8,283 restrictive relative clauses on subject position, with inanimate antecedents, in a written corpus consisting of British and Australian Hansard materials over five sampling years (1901, 1935, 1965, 1995, 2015). Our aim is to determine how processing-related factors and prescriptivism-related factors influence processes of language variation and change across two varieties of English. We analyse how the language-external variables of period and variety (British, Australian) interact with two groups of language-internal variables, namely predictors related to language processing and linguistic predictors associated with prescriptivism. The analysis shows that the relativiser that has been on the rise over the past century. The increase is particularly pronounced in the British Hansard, as which was comparatively frequent in early twentieth-century British material. As to the relative importance of predictors, we find that language-external predictors are the most important in conditioning the variation in relative markers, followed by processing-related constraints. Prescriptivism-related variables tend to generally be less important in this type of variation.
- 23 October 2023
- Published by John Benjamins Publishing Company
- p. 250-276
Processing and prescriptivism as constraints on language variation and change
We investigate the choice between the relative markers which and that in 8,283 restrictive relative clauses on subject position, with inanimate antecedents, in a written corpus consisting of British and Australian Hansard materials over five sampling years (1901, 1935, 1965, 1995, 2015). Our aim is to determine how processing-related factors and prescriptivism-related factors influence processes of language variation and change across two varieties of English. We analyse how the language-external variables of period and variety (British, Australian) interact with two groups of language-internal variables, namely predictors related to language processing and linguistic predictors associated with prescriptivism. The analysis shows that the relativiser that has been on the rise over the past century. The increase is particularly pronounced in the British Hansard, as which was comparatively frequent in early twentieth-century British material. As to the relative importance of predictors, we find that language-external predictors are the most important in conditioning the variation in relative markers, followed by processing-related constraints. Prescriptivism-related variables tend to generally be less important in this type of variation.
- Open Access
EmoTwiCS: A Corpus for Modelling Emotion Trajectories in Dutch Customer Service Dialogues on Twitter
Due to the rise of user-generated content, social media is increasingly adopted as a channel to deliver customer service. Given the public character of these online platforms, the automatic detection of emotions forms an important application in monitoring customer satisfaction and preventing negative word-of-mouth. This paper introduces EmoTwiCS, a corpus of 9,489 Dutch customer service dialogues on Twitter that are annotated for emotion trajectories. In our business-oriented corpus, we view emotions as dynamic attributes of the customer that can change at each utterance of the conversation. The term `emotion trajectory' refers therefore not only to the fine-grained emotions experienced by customers (annotated with 28 labels and valence-arousal-dominance scores), but also to the event happening prior to the conversation and the responses made by the human operator (both annotated with 8 categories). Inter-annotator agreement (IAA) scores on the resulting dataset are substantial and comparable with related research, underscoring its high quality. Given the interplay between the different layers of annotated information, we perform several in-depth analyses to investigate (i) static emotions in isolated tweets, (ii) dynamic emotions and their shifts in trajectory, and (iii) the role of causes and response strategies in emotion trajectories. We conclude by listing the advantages and limitations of our dataset, after which we give some suggestions on the different types of predictive modelling tasks and open research questions to which EmoTwiCS can be applied. The dataset is available upon request and will be made publicly available upon acceptance of the paper.
Open AccessEmoTwiCS: A Corpus for Modelling Emotion Trajectories in Dutch Customer Service Dialogues on Twitter
Due to the rise of user-generated content, social media is increasingly adopted as a channel to deliver customer service. Given the public character of these online platforms, the automatic detection of emotions forms an important application in monitoring customer satisfaction and preventing negative word-of-mouth. This paper introduces EmoTwiCS, a corpus of 9,489 Dutch customer service dialogues on Twitter that are annotated for emotion trajectories. In our business-oriented corpus, we view emotions as dynamic attributes of the customer that can change at each utterance of the conversation. The term `emotion trajectory' refers therefore not only to the fine-grained emotions experienced by customers (annotated with 28 labels and valence-arousal-dominance scores), but also to the event happening prior to the conversation and the responses made by the human operator (both annotated with 8 categories). Inter-annotator agreement (IAA) scores on the resulting dataset are substantial and comparable with related research, underscoring its high quality. Given the interplay between the different layers of annotated information, we perform several in-depth analyses to investigate (i) static emotions in isolated tweets, (ii) dynamic emotions and their shifts in trajectory, and (iii) the role of causes and response strategies in emotion trajectories. We conclude by listing the advantages and limitations of our dataset, after which we give some suggestions on the different types of predictive modelling tasks and open research questions to which EmoTwiCS can be applied. The dataset is available upon request and will be made publicly available upon acceptance of the paper.