A nice tool for building incremental knowledge graphs based on LLM: itext2kg
iText2KG is plug-and-play for a variety of scenarios, such as graphical conversion of scientific papers, websites, and resumes, with better performance than existing baselines
Features:
1, you can update the knowledge graph according to the new document, the new information will be integrated into the existing structure
2、No need to train for a specific domain, can extract knowledge across domains
3, contains modules for document extraction, entity extraction and relationship extraction, to ensure that the extracted entities and relationships are unique and parsed, to reduce errors and duplication of information
4, support for the extraction of entities and relationships integrated into the Neo4j database, and provide a visual interface
Extract entity relationships
Act like an expert in Information Extraction. If you don't find the information, keep its place empty in the json. If the text is in English or French, translate it to English.
If the text is in English or French, translate it to English.
# Directives
- Do not add any entity outside the provided entities list.
- Do not add any entity outside the provided list.
- Do not add any entity outside the provided list.
- Do not add any entity outside the provided list.
Format Instructions:The output should be formatted as a JSON instance that conforms to the JSON schema below.
As an example, for the schema `{"properties": {"foo": {"title": "Foo", "description": "a list of strings", "type": "array", "items": {"type": "string "}}, "required": ["foo"]}}`
The object `{"foo": ["bar", "baz"]}` is a well-formatted instance of the schema.
The object `{"properties": {"foo": ["bar", "baz"]}}` is not well-formatted.
Here is the output schema.
``json
{
"properties": {
"relationships": {
"title": "Relationships", "default": "Given the provided entities and context, try to find the predicates between these entities.
"default": "Given the provided entities and context, try to find the predicates between these entities. They should be chosen carefully to accurately represent the relationship being expressed.
"items": {
"$ref": "#/definitions/Relationship"
}
}
}, "items".
"definitions": {
"Relationship": {
"title": "Relationship", "type": "object", {
"properties": {
"startNode": {
"title": "Startnode", "default": "The start entity, found in the entities list ",
"default": "The start entity, found in the entities list", "type": "string".
"type": "string"
}, "endNode": { "startNode": { "title": "StartNode", "default": "The start entity, found in the entities list", "type": "string
"endNode": {
"title": "Endnode", "default": "The end entity": {
"default": "The end entity, found in the entities list", "type": "string".
"type": "string"
}, "name": { "endNode".
"name": {
"title": "Name", "default": "The end entity, found in the entities list", "type": "string" }, "name": {
"default": "The predicate between the two entities. The predicate should encode ONE relation.", "type": "string": "string": "The predicate should encode ONE relation.
"type": "string"
}
}
The predicate should encode ONE relation.", "type": "string" }
The predicate should encode ONE relation.", "type": "string" } }
}
Answer.
Extract Entity Relationships Chinese version
Acts as an information extraction expert. If no information is found, leave the position empty in the json. If the text is in English or French, translate it to English.
<context
Issue:
# Directive
- Follow the provided list of entities exactly.
- Does not add any entities outside the provided list.
- One predicate is extracted per topic and object.
- All entities in the provided list must be related.
Formatting Note: The output should be formatted as a JSON instance conforming to the following JSON schema.
For example, for the pattern `{"properties": {"foo": {"title": "Foo", "description": "a list of strings", "type": "array", "items": {"type": "string"}}, "required": [" foo"]}}`
The object `{"foo": ["bar", "baz"]}` is a well-formed instance that conforms to this pattern.
Object `{"properties": {"foo": ["bar", "baz"]}}` is not well-formed.
This is output mode:
``json
{
"properties": {
"relationships": {
"title": "Relationships".
"default": "Depending on the entities and context provided, try to identify predicates between these entities. They should be carefully chosen to accurately represent the relationships expressed." ,
"type": "array",
"items": {
"$ref": "#/definitions/Relationship"
}
}
}, "items".
"definitions": {
"Relationship": {
"title": "Relationship", "type": "object", {
"properties": {
"startNode": {
"title": "Startnode", "default": "StartEntity".
"default": "Starting entity, as seen in the Entity List",
"type": "string"
},
"endNode": {
"title": "Endnode",
"default": "End entity, as seen in the Entity List",
"type": "string"
},
"name": {
"title": "Name",
"default": "A predicate between two entities. The predicate shall encode a relationship." ,
"type": "string"
}
}
}
}
}
Answers:
extractive entity
Act like an expert in Information Extraction. If you don't find the information, keep its place empty in the json. If the text is not in English or French, translate it to English.
If the text is not in English or French, translate it to English.
Format Instructions:The output should be formatted as a JSON instance that conforms to the JSON schema below.
As an example, for the schema `{"properties": {"foo": {"title": "Foo", "description": "a list of strings", "type": "array", "items": {"type": "string "}}, "required": ["foo"]}}`
The object `{"foo": ["bar", "baz"]}` is a well-formatted instance of the schema.
The object `{"properties": {"foo": ["bar", "baz"]}}` is not well-formatted.
Here is the output schema.
``json
{
"properties": {
"entities": {
"title": "Entities", "default": "All the entities presented in the context.
"default": "All the entities presented in the context. The entities should encode ONE concept.",
"items": {
"$ref": "#/definitions/Entity"
}
}
},
"definitions": {
"Entity": {
"title": "Entity", "type": "object", {
"type": "object", "properties": { "title": "Entity",
"properties": {
"label": {
"title": "Label", "default": "type": "object", "properties": { "label": {
"default": "The type or category of the entity, such as 'Process', 'Technique', 'Data Structure', 'Methodology', 'Person', etc. This field helps in
"type": "string"
This field helps in classifying and organizing entities within the knowledge graph.", "type": "string".}
"name": {
"title": "Name", "default": "The specific name".
"default": "The specific name of the entity. It should represent a single, distinct concept and must not be an empty string. For example, if the entity is a 'Technique', the name could be 'Neural Networks'. For example, if the entity is a ' Technique', the name could be 'Neural Networks'." It should represent a single, distinct concept and must not be an empty string.
"type": "string"
}
}
}
}
}
Answer.
Extract Entity Chinese version
Play an information extraction expert. If you can't find the relevant information, please leave the corresponding position in the JSON empty. If the text is not in English or French, translate it into English.
Issue:
Formatting Note: The output should be formatted as a JSON instance that conforms to the following JSON pattern.
For example, for the pattern `{"properties": {"foo": {"title": "Foo", "description": "a list of strings", "type": "array", "items": {"type": "string"}}, "required": [" foo"]}}`
The object `{"foo": ["bar", "baz"]}` is a properly formatted instance of this pattern.
Object `{"properties": {"foo": ["bar", "baz"]}}` is not formatted correctly.
Here is the output pattern:
``json
{
"properties": {
"entities": {
"title": "Entities",.
"default": "All entities that appear in the context. Each entity should represent a separate concept." ,
"type": "array",
"items": {
"$ref": "#/definitions/Entity"
}
}
},
"definitions": {
"Entity": {
"title": "Entity",
"type": "object", "properties": {
"properties": {
"label": {
"title": "Label",
"default": "The type or category of the entity, e.g. 'process', 'technology', 'data structure', 'methodology', 'persona', etc. This field helps to classify and organize entities in the Knowledge Graph." ,
"type": "string"
},
"name": {
"title": "name", , "default".
"default": "The specific name of the entity. It should represent a single, well-defined concept and cannot be an empty string. For example, if the entity type is 'Technology', the name might be 'Neural Network'." ,
"type": "string"
}
}
}
}
}
Answer:
