Knowledge graphs (KGs) have been widely accepted as powerful tools for modeling the complex relationships between concepts and developing knowledge-based services. In recent years, researchers in the field of power systems have explored KGs to develop intelligent dispatching systems for increasingly large power grids. With multiple power grid dispatching knowl-edge graphs (PDKGs) constructed by different agencies, the knowledge fusion of different PDKGs is useful for providing more accurate decision supports. To achieve this, entity alignment that aims at connecting different KGs by identifying equivalent enti-ties is a critical step. Existing entity alignment methods cannot integrate useful structural, attribute, and relational information while calculating entities' similarities and are prone to making many-to-one alignments, thus can hardly achieve the best perfor-mance. To address these issues, this paper proposes a collective entity alignment model that integrates three kinds of available information and makes collective counterpart assignments. This model proposes a novel knowledge graph attention network (KGAT) to learn the embeddings of entities and relations explic-itly and calculates entities' similarities by adaptively incorporat-ing the structural, attribute, and relational similarities. Then, we formulate the counterpart assignment task as an integer pro-gramming (IP) problem to obtain one-to-one alignments. We not only conduct experiments on a pair of PDKGs but also evaluate our model on three commonly used cross-lingual KGs. Experi- mental comparisons indicate that our model outperforms other methods and provides an effective tool for the knowledge fusion of PDKGs.

Linyao Yang;Chen Lv;Xiao Wang;Ji Qiao;Weiping Ding;Jun Zhang;Fei-Yue Wang

State Key Laboratory for Management and Control of Complex Systems,Institute of Automation,Chinese Academy of Sciences,Beijing 100190;School of Artificial Intelligence,University of Chinese Academy of Sciences,Beijing 100049,China;China Electric Power Research Institute,Beijing 100192,China;Qingdao Academy of Intelligent Industries,Qingdao 256200,China;School of Information Science and Technology,Nantong University,Nantong 226019,China;School of Electrical Engineering and Automation,Wuhan University,Wuhan 430072,China

自动化学报（英文版）

[1]Linyao Yang;Chen Lv;Xiao Wang;Ji Qiao;Weiping Ding;Jun Zhang;Fei-Yue Wang-.Collective Entity Alignment for Knowledge Fusion of Power Grid Dispatching Knowledge Graphs)[J].自动化学报（英文版）,2022(11):1990-2004

Dispatching,PDKGs,enti,explic,itly,incorporat

Collective,Entity,Alignment,Knowledge,Fusion,Power,Grid,Graphs,graphs,have,been,widely,accepted,powerful,tools,modeling,complex,relationships,between,concepts,developing,knowledge,services,In,recent,years,researchers,field,systems,explored,intelligent,dispatching,increasingly,large,grids,With,multiple,constructed,by,different,agencies,fusion,useful,providing,more,accurate,decision,supports,To,achieve,this,entity,that,aims,connecting,identifying,equivalent,critical,step,Existing,methods,cannot,structural,attribute,relational,information,while,calculating,entities,similarities,are,prone,making,many,alignments,thus,hardly,best,mance,address,these,issues,paper,proposes,collective,integrates,three,kinds,available,makes,counterpart,assignments,This,novel,attention,network,KGAT,learn,embeddings,calculates,adaptively,Then,formulate,task,integer,gramming,problem,obtain,We,conduct,experiments,pair,also,evaluate,our,commonly,used,cross,lingual,Experi,mental,comparisons,indicate,outperforms,other,provides,effective

0.534222