← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 6780.b · 生成模型:GPT-6

关系标识与别名应统一管理

理解等级:中级

kind of、type of等表面表达可映射到同一个受控关系,但须确认语义确实一致。实体也应按消歧结果关联稳定ID;规范化减少重复,不应把训练模型觉得相似的任何词都并成同一种关系。

这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • · 查看原书依据

      Most knowledge bases normalize the strings that define these relationships, so kind of and type of would be assigned a normalized string or ID to represent that particular relation. Some knowledge bases also normalize the nouns representing the objects in a knowledge base, using coreference resolution, as previously described. So the bigram Stanislav Petrov might be assigned a particular ID. Synonyms for Stanislav Petrov, like

    参考来源

    - 位置:《Natural Language Processing in Action, Second Edition》第11章《信息抽取与知识图谱》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Natural Language Processing in Action, Second Edition/_epub-src/OEBPS/Text/chapter-11.html) - 结论依据:摘录中关于关系标识与别名应统一管理的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。 - 原始内容:Most knowledge bases normalize the strings that define these relationships, so kind of and type of would be assigned a normalized string or ID to represent that particular relation. Some knowledge bases also normalize the nouns representing the objects in a knowledge base, using coreference resolution, as previously described. So the bigram Stanislav Petrov might be assigned a particular ID. Synonyms for Stanislav Petrov, like