move images for kag examples to _static/images (#214)
xionghuaidong
GitHub
|
Before Width: | Height: | Size: 7.5 KiB After Width: | Height: | Size: 7.5 KiB |
|
Before Width: | Height: | Size: 283 KiB After Width: | Height: | Size: 283 KiB |
|
Before Width: | Height: | Size: 68 KiB After Width: | Height: | Size: 68 KiB |
|
Before Width: | Height: | Size: 273 KiB After Width: | Height: | Size: 273 KiB |
|
Before Width: | Height: | Size: 84 KiB After Width: | Height: | Size: 84 KiB |
|
Before Width: | Height: | Size: 1.7 MiB After Width: | Height: | Size: 1.7 MiB |
|
Before Width: | Height: | Size: 1.7 MiB After Width: | Height: | Size: 1.7 MiB |
|
Before Width: | Height: | Size: 1.7 MiB After Width: | Height: | Size: 1.7 MiB |
|
Before Width: | Height: | Size: 3.1 MiB After Width: | Height: | Size: 3.1 MiB |
|
|
@ -316,7 +316,7 @@ The KAG framework provides checkpoint-based resumption functionality. If the tas
|
||||||
|
|
||||||
Currently, OpenSPG-KAG provides the [Knowledge Exploration](https://openspg.yuque.com/ndx6g9/cwh47i/mzq74eaynm4rqx4b) capability in product mode, along with the corresponding API documentation [HTTP API Reference](https://openspg.yuque.com/ndx6g9/cwh47i/qvbgge62p7argtd2).
|
Currently, OpenSPG-KAG provides the [Knowledge Exploration](https://openspg.yuque.com/ndx6g9/cwh47i/mzq74eaynm4rqx4b) capability in product mode, along with the corresponding API documentation [HTTP API Reference](https://openspg.yuque.com/ndx6g9/cwh47i/qvbgge62p7argtd2).
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
## 4. Reasoning-based question answering
|
## 4. Reasoning-based question answering
|
||||||
|
|
||||||
|
|
|
||||||
|
|
@ -316,7 +316,7 @@ KAG 框架基于 checkpoint 文件提供了断点续跑的功能。如果由于
|
||||||
|
|
||||||
当前，OpenSPG-KAG 在产品端已提供 [知识探查](https://openspg.yuque.com/ndx6g9/0.6/fw4ge5c18tyfl2yq) 能力，以及对应的 API 文档 [HTTP API Reference](https://openspg.yuque.com/ndx6g9/0.6/zde1yunbb8sncxtv)。
|
当前，OpenSPG-KAG 在产品端已提供 [知识探查](https://openspg.yuque.com/ndx6g9/0.6/fw4ge5c18tyfl2yq) 能力，以及对应的 API 文档 [HTTP API Reference](https://openspg.yuque.com/ndx6g9/0.6/zde1yunbb8sncxtv)。
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
## 4. 推理问答
|
## 4. 推理问答
|
||||||
|
|
||||||
|
|
|
||||||
|
|
@ -5,7 +5,7 @@
|
||||||
|
|
||||||
This example aims to demonstrate how to extract and construct entities and relations in a knowledge graph based on the SPG-Schema using LLMs.
|
This example aims to demonstrate how to extract and construct entities and relations in a knowledge graph based on the SPG-Schema using LLMs.
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
## 1. Precondition
|
## 1. Precondition
|
||||||
|
|
||||||
|
|
|
||||||
|
|
@ -5,7 +5,7 @@
|
||||||
|
|
||||||
本示例旨在展示如何基于 schema 的定义，利用大模型实现对图谱实体和关系的抽取和构建到图谱。
|
本示例旨在展示如何基于 schema 的定义，利用大模型实现对图谱实体和关系的抽取和构建到图谱。
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
## 1. 前置条件
|
## 1. 前置条件
|
||||||
|
|
||||||
|
|
|
||||||
|
|
@ -7,7 +7,7 @@
|
||||||
|
|
||||||
**Keywords**: semantic properties, dynamic multi-classification of entities, knowledge application in the context of hierarchical business knowledge and factual data.
|
**Keywords**: semantic properties, dynamic multi-classification of entities, knowledge application in the context of hierarchical business knowledge and factual data.
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
## 1. Precondition
|
## 1. Precondition
|
||||||
|
|
||||||
|
|
@ -69,7 +69,7 @@ knext reasoner execute --dsl "${ql}"
|
||||||
|
|
||||||
#### Scenario 1: Semantic attributes vs text attributes
|
#### Scenario 1: Semantic attributes vs text attributes
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
MobilePhone: "standard attribute" vs "text attribute".
|
MobilePhone: "standard attribute" vs "text attribute".
|
||||||
|
|
||||||
|
|
|
||||||
|
|
@ -5,7 +5,7 @@
|
||||||
|
|
||||||
**关键词**：语义属性，实体动态多分类，面向业务知识和事实数据分层下的知识应用
|
**关键词**：语义属性，实体动态多分类，面向业务知识和事实数据分层下的知识应用
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
## 1. 前置条件
|
## 1. 前置条件
|
||||||
|
|
||||||
|
|
@ -67,7 +67,7 @@ knext reasoner execute --dsl "${ql}"
|
||||||
|
|
||||||
#### 场景 1：语义属性对比文本属性
|
#### 场景 1：语义属性对比文本属性
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
电话号码：标准属性 vs 文本属性。
|
电话号码：标准属性 vs 文本属性。
|
||||||
|
|
||||||
|
|
|
||||||
|
|
@ -13,11 +13,11 @@ This example is based on the SPG framework to construct an industry chain enterp
|
||||||
|
|
||||||
Please refer to the document for knowledge modeling: [Schema of Enterprise Supply Chain Knowledge Graph](./schema/README.md), As shown in the example below:
|
Please refer to the document for knowledge modeling: [Schema of Enterprise Supply Chain Knowledge Graph](./schema/README.md), As shown in the example below:
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
Concept knowledge maintains industry chain-related data, including hierarchical relations, supply relations. Entity instances consist of only legal representatives and transfer information. Company instances are linked to product instances based on the attributes of the products they produce, enabling deep information mining between company instances, such as supplier relationships, industry peers, and shared legal representatives. By leveraging deep contextual information, more credit assessment factors can be provided.
|
Concept knowledge maintains industry chain-related data, including hierarchical relations, supply relations. Entity instances consist of only legal representatives and transfer information. Company instances are linked to product instances based on the attributes of the products they produce, enabling deep information mining between company instances, such as supplier relationships, industry peers, and shared legal representatives. By leveraging deep contextual information, more credit assessment factors can be provided.
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
Within the industrial chain, categories of product and company events are established. These categories are a combination of indices and trends. For example, an increase in price consists of the index "价格" (price) and the trend "上涨" (rising). Causal knowledge sets the events of a company's profit decrease and cost increase due to a rise in product prices. When a specific event occurs, such as a significant increase in rubber prices, it is categorized under the event of a price increase. As per the causal knowledge, a price increase in a product leads to two event types: a decrease in company profits and an increase in company costs. Consequently, new events are generated:"三角\*\*轮胎公司成本上涨事件" and "三角\*\*轮胎公司利润下跌".
|
Within the industrial chain, categories of product and company events are established. These categories are a combination of indices and trends. For example, an increase in price consists of the index "价格" (price) and the trend "上涨" (rising). Causal knowledge sets the events of a company's profit decrease and cost increase due to a rise in product prices. When a specific event occurs, such as a significant increase in rubber prices, it is categorized under the event of a price increase. As per the causal knowledge, a price increase in a product leads to two event types: a decrease in company profits and an increase in company costs. Consequently, new events are generated:"三角\*\*轮胎公司成本上涨事件" and "三角\*\*轮胎公司利润下跌".
|
||||||
|
|
||||||
|
|
|
||||||
|
|
@ -13,11 +13,11 @@
|
||||||
|
|
||||||
建模参考 [基于 SPG 建模的产业链企业图谱](./schema/README_cn.md)，如下图示意。
|
建模参考 [基于 SPG 建模的产业链企业图谱](./schema/README_cn.md)，如下图示意。
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
概念知识维护着产业链相关数据，包括上下位层级、供应关系；实体实例仅有法人代表、转账信息，公司实例通过生产的产品属性和概念中的产品节点挂载，实现了公司实例之间的深度信息挖掘，例如供应商、同行业、同法人代表等关系。基于深度上下文信息，可提供更多的信用评估因子。
|
概念知识维护着产业链相关数据，包括上下位层级、供应关系；实体实例仅有法人代表、转账信息，公司实例通过生产的产品属性和概念中的产品节点挂载，实现了公司实例之间的深度信息挖掘，例如供应商、同行业、同法人代表等关系。基于深度上下文信息，可提供更多的信用评估因子。
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
产业链中建立了产品和公司事件类别，该类别属于指标和趋势的一种组合，例如价格上涨，是由指标：价格，趋势：上涨两部分构成。
|
产业链中建立了产品和公司事件类别，该类别属于指标和趋势的一种组合，例如价格上涨，是由指标：价格，趋势：上涨两部分构成。
|
||||||
|
|
||||||
|
|
|
||||||