YUHAN03/yelp-rag-data

该数据集名为Yelp RAG Agent Data，是为Yelp业务智能代理演示提供的向量存储和源数据。它包括一个FAISS索引（包含60,823个块）、块元数据与嵌入映射、50,000条采样的Yelp评论以及Yelp业务元数据JSON文件。嵌入模型使用sentence-transformers/all-MiniLM-L6-v2。数据集主要用于文本检索任务，支持英语语言，适用于构建和测试基于检索增强生成（RAG）的代理系统，以分析和查询Yelp评论和业务信息。This dataset, named Yelp RAG Agent Data, provides vector store and source data for the Yelp Business Intelligence Agent demo. It includes a FAISS index (with 60,823 chunks), chunk metadata and embeddings mapping, 50k sampled Yelp reviews, and Yelp business metadata in JSON format. The embedding model used is sentence-transformers/all-MiniLM-L6-v2. The dataset is designed for text-retrieval tasks, supports the English language, and is intended for building and testing retrieval-augmented generation (RAG) agent systems to analyze and query Yelp reviews and business information.