EmoTwiCS

EmoTwiCS是一个专为荷兰语客户服务对话中情感轨迹建模设计的数据集，由根特大学的LT3语言与翻译技术团队创建。该数据集包含9489条从Twitter收集的对话，每条对话都标注了情感轨迹，包括客户情感的细粒度分类和维度评分。数据集的创建旨在解决现有情感识别对话数据集的不足，特别是在特定工业应用中遇到的数据噪声和领域特定性问题。EmoTwiCS数据集不仅适用于情感分析研究，还可应用于客户满意度监控、客户流失预防和客户服务交互的成功与否监控等多个领域。

EmoTwiCS is a dataset specifically designed for modeling emotional trajectories in Dutch customer service conversations, developed by the LT3 Language and Translation Technology Team at Ghent University. This dataset contains 9489 dialogues collected from Twitter, with each dialogue annotated with emotional trajectories, including fine-grained classification of customer emotions and dimensional sentiment scoring. The dataset was developed to address the limitations of existing emotion recognition dialogue datasets, particularly the issues of data noise and domain specificity encountered in specific industrial applications. The EmoTwiCS dataset is not only applicable to sentiment analysis research, but also can be applied to multiple fields including customer satisfaction monitoring, customer churn prevention, and monitoring of the success or failure of customer service interactions.