An Annotated Corpus of Relational Strategies in Customer Service

本数据集名为‘客户服务中的关系策略标注语料库’，由Next IT Corporation创建，旨在通过分析客户服务对话中的关系策略，提升智能虚拟代理（IVAs）的交互质量。数据集包含8000条来自旅行和电信领域的IVAs与用户的对话，每条对话均经过详细标注，以识别如问候、背景故事、理由、感激、抱怨或情感等语言类别。此数据集不仅有助于IVAs更好地理解和响应用户意图，还为研究多意图检测提供了基础。通过此数据集，研究者可以探索如何自动化识别和分类关系策略，从而使IVAs在处理客户服务时更加高效和人性化。

This dataset is titled "Relational Strategy Annotated Corpus in Customer Service" and was developed by Next IT Corporation, with the goal of enhancing the interaction quality of intelligent virtual agents (IVAs) by analyzing relational strategies in customer service dialogues. The corpus comprises 8,000 dialogues between IVAs and users from the travel and telecommunications industries, with each dialogue thoroughly annotated to identify linguistic categories including greetings, background narratives, justifications, expressions of gratitude, complaints, and emotional states. This corpus not only helps IVAs better understand and respond to user intentions but also provides a foundational resource for research on multi-intent detection. By leveraging this corpus, researchers can explore methods to automatically identify and classify relational strategies, thereby enabling IVAs to deliver more efficient and human-centric customer service interactions.