Electric power potential safety hazard risk grade classification method and system based on knowledge graph

The invention discloses an electric power potential safety hazard risk grade classification method and system based on a knowledge graph. An electric power potential safety hazard entity identification model is established to carry out entity identification on a long text; simplifying the long text...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: MU JINGLONG, YANG XUXIN, YE DEWU, JIN SHUANGYUAN, ZHANG XIN, LU TIANFENG, ZHANG DAYONG, WANG GANG, GAO SONG, CAO FEIYI, LI TIANNING, LOU ZHANHAO, SUN JIAN, DING YIXIN
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The invention discloses an electric power potential safety hazard risk grade classification method and system based on a knowledge graph. An electric power potential safety hazard entity identification model is established to carry out entity identification on a long text; simplifying the long text of the electric power hidden danger treatment measures through a simplified neural network model, and performing entity attribute extraction by adopting a rule-based information extraction algorithm; constructing an entity-based electric power potential safety hazard triple knowledge graph, and performing vectorization representation on an electric power potential safety hazard triple by using a TransH model; classifying the risk levels of the electric power hidden dangers by using a TextCNN model; the system comprises a power hidden danger entity identification model, a simplified neural network model, an entity attribute extraction algorithm module, a power hidden danger vectorization presentation layer module an