You cannot select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
This file contains ambiguous Unicode characters that may be confused with others in your current locale. If your use case is intentional and legitimate, you can safely ignore this warning. Use the Escape button to highlight these characters.
# 提示词
## 任务描述:
你是一个信息抽取引擎,需要从给定的文本中提取符合指定三元组标签(实体、关系、属性)的结构化数据。
## 输入数据:
- 待处理文本:
{text}
- 三元组标签及属性名称:
{domainMetadata}
## 示例:
{
"nodes": [
{
"name":"龙源(酒泉)风力发电有限公司",
"type": "公司",
"attributes": {
"地址": "雨花台区"
}
},
{
"name":"电子银行承兑汇票",
"type": "电子银行承兑汇票",
"attributes": {
"金额": "100.00万元",
"打印时间": "2024年10月20号"
}
},
{
"name": "杭州六小龙",
"type": "公司",
"attributes": {
"地址": "杭州高新区"
}
}
],
"relations": [
{
"name": "收购",
"source": "公司",
"target": "公司",
"type": "收购",
"attributes": {
"收购类型": "全资收购",
"收购时间": "2025年5月28号"
}
},
{
"name":"收购",
"source": "公司",
"target": "电子银行承兑汇票",
"type": "持有",
"attributes": {
"持有方式": "纸质"
}
}
],
"typed_triplets": [
[
"公司",
"持有",
"电子银行承兑汇票"
],
[
"公司",
"收购",
"公司"
]
]
}
## 注意事项:
- 每一个nodes、relations都有一个基础属性"名称"是必须有的。
- 仅提取 `domainMetadata` 中定义的标签和属性。
- 若属性无对应值,可留空或忽略。
- 确保提取的值与原文一致,不进行推断或改写。
- 输出纯JSON格式, 不要使用```json ```等任何Markdown标记包装./no_think