.png){kind=icon}
.png){kind=icon}
Return
An automatic construction and intelligent retrieval method for knowledge graph for machining process specifications
.png){kind=icon}
PRE
AI
DOI：10.1080/0951192X.2025.2544546.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In order to semantically organize and intelligently reuse the machining process knowledge contained in a large number of machining process specifications (MPS) and improve the efficiency of process design, this paper proposes an automatic construction and intelligent retrieval method for knowledge graph for machining process specifications (KG-MPS). First, a machining process information model is established to achieve structured multi-level representation of MPS, meeting the requirements for knowledge graph construction. Subsequently, a joint extraction model for process entity-relationship based on multi-modal feature fusion is developed, which uses natural language processing technology to extract process data from MPS to realize automatic construction of KG-MPS. Furthermore, an intelligent retrieval technology of MPS based on subgraph matching is proposed, which can semantically match the search requirements expressed in natural language with the historical machining process knowledge base and retrieve the reusable machining process instances precisely and intelligently. Finally, in the field of cold heading die machining, the superiority of the proposed KG-MPS construction method is verified by serial experiments, and the developed application system verifies that the proposed intelligent retrieval method can effectively support the reuse of MPS.
Keywords:
Knowledge graph
natural language processing
joint extraction
intelligent retrieval
machining knowledge reuse
machining process specifications
Journal
I
IF:
4
Papers:
2.3K
Citations:
3.4K