convai-challenge/conv_ai_3

该数据集是为2020年SCAI EMNLP研讨会中的Conv AI 3挑战而创建的，旨在研究对话系统中如何处理用户提出的模糊问题。数据集包含用户初始请求、话题描述、澄清需求、问题ID、澄清问题及其答案等字段。数据集的主要目标是帮助系统识别模糊问题，并提出好的澄清问题，而不是直接回答问题。

This dataset was created for the Conv AI 3 Challenge at the 2020 SCAI EMNLP Workshop, aiming to investigate how dialogue systems handle ambiguous questions posed by users. The dataset comprises fields including initial user requests, topic descriptions, clarification requirements, question IDs, clarification questions and their corresponding answers. The core goal of this dataset is to help systems identify ambiguous questions and generate high-quality clarification questions, rather than directly answering those questions.

原始信息汇总

数据集概述

数据集描述

数据集摘要

Conv AI 3挑战赛是2020年EMNLP研讨会的一部分，旨在研究对话系统在用户请求可能含糊不清时的应对策略。主要目标是识别出含糊的请求，并提出澄清问题，而不是直接尝试回答。

支持的任务和排行榜

[更多信息待补充]

语言

• 英语

数据集结构

数据实例

以下是数据集中的几个示例：

json { "topic_id": 8, "facet_id": "F0968", "initial_request": "I want to know about appraisals.", "topic_desc": "Find information about the appraisals in nearby companies.", "clarification_need": 2, "question_id": "F0001", "question": "are you looking for a type of appraiser", "answer": "im looking for nearby companies that do home appraisals", "facet_desc": "Get the TYPE of Appraisals", "conversation_context": [], "context_id": 968 }

json { "topic_id": 8, "facet_id": "F0969", "initial_request": "I want to know about appraisals.", "topic_desc": "Find information about the type of appraisals.", "clarification_need": 2, "question_id": "F0005", "question": "are you looking for a type of appraiser", "facet_desc": "Get the TYPE of Appraisals", "answer": "yes jewelry", "conversation_context": [], "context_id": 969 }

json { "topic_id": 293, "facet_id": "F0729", "initial_request": "Tell me about the educational advantages of social networking sites.", "topic_desc": "Find information about the educational benefits of the social media sites", "clarification_need": 2, "question_id": "F0009", "question": "which social networking sites would you like information on", "answer": "i don have a specific one in mind just overall educational benefits to social media sites", "facet_desc": "Detailed information about the Networking Sites.", "conversation_context": [ { "question": "what level of schooling are you interested in gaining the advantages to social networking sites", "answer": "all levels" }, { "question": "what type of educational advantages are you seeking from social networking", "answer": "i just want to know if there are any" } ], "context_id": 976573 }

数据字段

• topic_id: 主题ID。
• initial_request: 发起对话的查询（文本）。
• topic_desc: 主题的完整描述。
• clarification_need: 澄清需求的标签，范围从1到4。
• facet_id: 方面ID。
• facet_desc: 方面的完整描述。
• question_id: 问题ID。
• question: 系统可以向用户提出的澄清问题。
• answer: 对澄清问题的回答。

数据分割

[更多信息待补充]

数据集创建

策划理由

[更多信息待补充]

源数据

初始数据收集和规范化

[更多信息待补充]

源语言生产者

[更多信息待补充]

注释

注释过程

[更多信息待补充]

注释者

[更多信息待补充]

个人和敏感信息

[更多信息待补充]

使用数据的注意事项

数据集的社会影响

[更多信息待补充]

偏见的讨论

[更多信息待补充]

其他已知限制

[更多信息待补充]

附加信息

数据集策展人

[更多信息待补充]

许可信息

[更多信息待补充]

引用信息

plaintext @misc{aliannejadi2020convai3, title={ConvAI3: Generating Clarifying Questions for Open-Domain Dialogue Systems (ClariQ)}, author={Mohammad Aliannejadi and Julia Kiseleva and Aleksandr Chuklin and Jeff Dalton and Mikhail Burtsev}, year={2020}, eprint={2009.11352}, archivePrefix={arXiv}, primaryClass={cs.CL} }

贡献

感谢@rkc007添加此数据集。