Open AccessCross-domain Knowledge Discovery based on Knowledge Graph and Patent Mining
Author(s) -
Fan Ye,
Taotao Fu,
Lin Gong,
Jun Gao
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1744/4/042155
Subject(s) - knowledge extraction , domain knowledge , computer science , domain (mathematical analysis) , graph , domain engineering , knowledge engineering , data mining , software mining , open knowledge base connectivity , knowledge graph , reuse , process (computing) , information retrieval , artificial intelligence , knowledge management , theoretical computer science , engineering , mathematics , programming language , personal knowledge management , organizational learning , mathematical analysis , software system , waste management , software construction , component based software engineering , software
This paper studies an approach on cross-domain knowledge discovery to assist the conceptual stage of the design process related to mechanical engineering. Variable methods and tools are proposed to obtain knowledge within a given domain until now. However, methods on cross-domain knowledge analysis is under-developed. In this paper, domain knowledge graph is built automatically by employing natural language process (NLP) and patent mining. They comprise patent documents obtaining and knowledge extraction. Then according to the international patent classification (IPC), the knowledge elements are divided to some different categories. The elements are stored in Databases and then the given domain knowledge graph is constructed after correlation analyses. The cross-domain knowledge surrounding the given domain knowledge is found by mining the correlation among cross-domain knowledge. The cross-domain knowledge can inspire designers about new design of a given domain. And the efficiency of knowledge reusing can be improved by domain knowledge graphs.