公文知识图谱 (Official Document Knowledge Graph) - 公文知识图谱 (Mobile Application Knowledge Graph)

Time: 2/6/2023 Authors: Wenbo Zhou & Weizhuo Li & Bianyu Yang

Mail: b20090724@njupt.edu.cn & liweizhuo@amss.ac.cn & b20090721@njupt.edu.cn

Description: ODKG is an official document knowledge graph. You can use its original resources and visit the website to enjoy its services.

1. Introduction：

In this project, we present an official document knowledge graph, namely ODKG, which aims to collect the documents for effective management.

We design a lightweight ontology of official documents. It can bring a well-defined schema of collected documents so that they could share more linkage with each other. We present the algorithms of element extraction, document archiving knowledge alignment during the process of construction, and further evaluate the corresponding algorithms with our constructed datasets. Experimental results show that several algorithms can be competent to above tasks to some extent. Finally, we list three use cases of ODKG that are helpful for managers and can improve the efficiency of their document management.

2. Usage:

MAKG consists of five important resources, including Ontology (Knowledge Schema), Statistics of official documents, Structured document, Test data for document archiving and element extraction.

Ontology:

We design one lightweight ontology of documents. It can bring a well-defined schema of collected documents so that these documents could share more linkage with each other. It contains 39 basic classes, 10 relations and 30 properties.

For the this file, it needs to install protege to open it.

Statistics of official documents:

This dataset is statistics of crawlled documents from from the Ministry of Industry and information technology, nuclear safety administration, the Ministry of Science and Technology, and the Government official website.

Encyclopedias:

This dataset contains the structureed triples of documents contained extracted value of elements.

Test data:

It is corpus to test the models for the tasks of Document Archiving and Element Extraction.

3. Use-Cases:

We list the main use-cases of ODKG for effective management.

Semantic Retrieval: ODKG can achieve semantic retrieval. For example, if users query one document, ODKG can display its comprehensive structured information to them in the form of knowledge cards. Benefited from this case, managers can quickly understand the key contents of documents.
Association Discovery. ODKG can recommend some similar documents for managers, which is helpful for them to further analyze the reliability and substation of other news and public sentiments.
Auxiliary Decision-Making. The comprehensive information of ODKG and its services can assist users to make better decisions. Nevertheless, it depends on the scale of ODKG and the quality of extracted elements.

Felt	Verdi
Data sist oppdatert	30. mai 2023
Metadata sist oppdatert	30. mai 2023
Opprettet	30. mai 2023
Format	ZIP
Lisens	Other (Non-Commercial)
Datastore active	False
Has views	False
Id	735e9458-0bfc-435b-a75b-9344dbd70fbd
Package id	1cf87cb1-d1ed-430f-8070-1c882d452ca8
Position	0
State	active
Url type	upload

公文知识图谱 (Mobile Application Knowledge Graph)