AUGUST

AUGUST是一个自动生成的对话推荐数据集，由香港中文大学（深圳）创建。该数据集通过数据到文本的生成过程，从结构化图表中生成大规模高质量的推荐对话，利用真实世界的用户-项目信息。数据集整合了传统推荐数据集中的丰富个性化用户配置文件、知识图谱中的丰富外部知识以及人类到人类对话推荐数据集中的对话能力。该数据集旨在解决对话推荐系统中训练数据不足的问题，特别是在低资源环境下，通过自动合成数据来提高推荐系统的有效性。

AUGUST is an automatically generated conversational recommendation dataset developed by The Chinese University of Hong Kong, Shenzhen. Leveraging real-world user-item information, this dataset generates large-scale, high-quality recommendation dialogues from structured schemas via a data-to-text generation process. It integrates rich personalized user profiles from traditional recommendation datasets, abundant external knowledge from knowledge graphs, and conversational capabilities from human-to-human conversational recommendation datasets. This dataset aims to address the shortage of training data for conversational recommendation systems, particularly in low-resource scenarios, by automatically synthesizing data to enhance the effectiveness of recommendation systems.