Ontology Training Dataset

该数据集包含了175个用户提示，旨在训练语言模型掌握与核心家庭关系相关的概念，涵盖了各种本体类别和属性。其中，有143个提示是关于核心家庭关系的手工制作内容，另外32个提示则是用于处理不在本体概念范围内的通用输入。规模上，该数据集共有175个用户提示，其任务是针对预定义的本体概念，对语言模型进行微调以实现知识捕获。

This dataset contains 175 user prompts, which are designed to train language models to master concepts related to core family relationships and cover various ontology categories and attributes. Among them, 143 prompts are handcrafted content focused on core family relationships, while the remaining 32 prompts are used to handle general inputs that fall outside the scope of the predefined ontology concepts. In terms of scale, this dataset comprises exactly 175 user prompts, and its core task is to fine-tune language models for knowledge capture targeting predefined ontology concepts.