实体识别优化
优化内容中的实体标注,提升AI引擎识别和引用准确性
实体识别概述
**实体识别(NER)**识别文本中的关键实体[1],是AI理解内容的基础[2]。
实体类型
基础实体[3]
- 人名(PERSON)、地名(LOCATION)、组织(ORGANIZATION)、时间(DATE)
领域实体[4]
- 产品名、品牌名、技术术语、行业概念
优化方法
明确标注[5]
- 首次出现完整名称、使用Schema标记、上下文清晰
一致性[6]
- 统一命名、避免歧义、规范表达
结构化[7]
- Schema.org标记、JSON-LD格式、微数据
技术实施
Schema标记示例[8]
{
"@context": "https://schema.org",
"@type": "Person",
"name": "张三",
"jobTitle": "CEO",
"worksFor": {
"@type": "Organization",
"name": "公司名称"
}
}
相关资源
参考文献
-
Stanford NLP. (2024). "Named Entity Recognition". NLP Guide. https://nlp.stanford.edu/software/CRF-NER.html
-
spaCy. (2024). "Entity Recognition". Documentation. https://spacy.io/usage/linguistic-features#named-entities
-
OntoNotes. (2024). "Entity Types". Annotation Standard. https://catalog.ldc.upenn.edu/LDC2013T19
-
ACL. (2024). "Domain-Specific NER". Research. https://www.aclweb.org/
-
Google. (2024). "Entity Markup". Structured Data. https://developers.google.com/search/docs/appearance/structured-data/intro-structured-data
-
W3C. (2024). "Naming Conventions". Best Practices. https://www.w3.org/
-
Schema.org. (2024). "Entity Schemas". Vocabulary. https://schema.org/
-
Google. (2024). "JSON-LD". Implementation Guide. https://developers.google.com/search/docs/appearance/structured-data/intro-structured-data#structured-data-format
更新日期:2025-11
词条状态:✅ 已完成