Amazon-M2
收藏资源简介:
Amazon-M2是由亚马逊公司推出的首个多语言多地区购物会话数据集,包含来自六个不同地区的数百万用户会话,主要语言包括英语、德语、日语、法语、意大利语和西班牙语。该数据集通过收集真实用户会话数据构建,每个样本包含用户按时间顺序交互的产品列表,并提供包含产品标题、品牌、颜色等属性的产品属性表。Amazon-M2旨在通过丰富的语义属性和大规模数据,增强个性化推荐和用户偏好理解,适用于多种现有任务和新任务,如产品推荐和标题生成。
Amazon-M2 is the first multilingual and multi-regional shopping conversation dataset launched by Amazon. It houses millions of user conversation sessions across six distinct regions, with major languages including English, German, Japanese, French, Italian and Spanish. This dataset is constructed by collecting real-world user conversation data. Each sample contains a chronological list of products that users have interacted with, alongside a product attribute table detailing attributes such as product title, brand, color and other relevant product features. Amazon-M2 aims to enhance personalized recommendation and user preference understanding through its rich semantic attributes and large-scale dataset, and is applicable to a variety of existing and emerging tasks such as product recommendation and title generation.




