PhillyMac/Marie_Curie_Corpus

该数据集是由Deku Corpus Builder自动生成的，用于基于RAG的AI应用。数据集主题为玛丽·居里（Marie Curie），类型为人物传记，包含1,214个项目，所有文本都预先计算了嵌入向量（使用all-MiniLM-L6-v2模型）。每条记录包含文本内容、来源URL、来源标题、来源域名、相关性评分（0-1）、质量评分（0-1）、检测到的主题JSON数组、字符长度、相关主题名称、主题类型（"personality"或"topic"）、提取日期以及预计算的384维嵌入向量。数据集设计用于与现有嵌入语料库集成，嵌入使用sentence-transformers/all-MiniLM-L6-v2模型，与FAISS索引兼容。

This corpus was automatically generated by the Deku Corpus Builder for use in RAG-based AI applications. The subject is Marie Curie, with a subject type of personality, containing 1,214 items, all with pre-computed embeddings (all-MiniLM-L6-v2 model). Each record contains: text content, source URL, source title, source domain, relevance score (0-1), quality score (0-1), JSON array of detected topics, character count, subject name, subject type ("personality" or "topic"), extraction date, and pre-computed 384-dimensional embedding vector. The dataset is designed to be integrated with existing embedded corpuses, using the sentence-transformers/all-MiniLM-L6-v2 model embeddings compatible with FAISS indexing.