To solve the low efficiency of approximate queries caused by the large sizes of the knowledge graphs in the real world,an embedding-based approximate query method is proposed.First,the nodes in the query graph are cla...To solve the low efficiency of approximate queries caused by the large sizes of the knowledge graphs in the real world,an embedding-based approximate query method is proposed.First,the nodes in the query graph are classified according to the degrees of approximation required for different types of nodes.This classification transforms the query problem into three constraints,from which approximate information is extracted.Second,candidates are generated by calculating the similarity between embeddings.Finally,a deep neural network model is designed,incorporating a loss function based on the high-dimensional ellipsoidal diffusion distance.This model identifies the distance between nodes using their embeddings and constructs a score function.k nodes are returned as the query results.The results show that the proposed method can return both exact results and approximate matching results.On datasets DBLP(DataBase systems and Logic Programming)and FUA-S(Flight USA Airports-Sparse),this method exhibits superior performance in terms of precision and recall,returning results in 0.10 and 0.03 s,respectively.This indicates greater efficiency compared to PathSim and other comparative methods.展开更多
Aiming at the relation linking task for question answering over knowledge base,especially the multi relation linking task for complex questions,a relation linking approach based on the multi-attention recurrent neural...Aiming at the relation linking task for question answering over knowledge base,especially the multi relation linking task for complex questions,a relation linking approach based on the multi-attention recurrent neural network(RNN)model is proposed,which works for both simple and complex questions.First,the vector representations of questions are learned by the bidirectional long short-term memory(Bi-LSTM)model at the word and character levels,and named entities in questions are labeled by the conditional random field(CRF)model.Candidate entities are generated based on a dictionary,the disambiguation of candidate entities is realized based on predefined rules,and named entities mentioned in questions are linked to entities in knowledge base.Next,questions are classified into simple or complex questions by the machine learning method.Starting from the identified entities,for simple questions,one-hop relations are collected in the knowledge base as candidate relations;for complex questions,two-hop relations are collected as candidates.Finally,the multi-attention Bi-LSTM model is used to encode questions and candidate relations,compare their similarity,and return the candidate relation with the highest similarity as the result of relation linking.It is worth noting that the Bi-LSTM model with one attentions is adopted for simple questions,and the Bi-LSTM model with two attentions is adopted for complex questions.The experimental results show that,based on the effective entity linking method,the Bi-LSTM model with the attention mechanism improves the relation linking effectiveness of both simple and complex questions,which outperforms the existing relation linking methods based on graph algorithm or linguistics understanding.展开更多
To revise stratified web ontology language(OWL)ontologies,the kernel revision operator is extended by defining novel conflict stratification and the incision function based on integer linear programming(ILP).The ILP-b...To revise stratified web ontology language(OWL)ontologies,the kernel revision operator is extended by defining novel conflict stratification and the incision function based on integer linear programming(ILP).The ILP-based model considers an optimization problem of minimizing a linear objective function which is suitable for selecting the minimal number of axioms to remove when revising ontologies.Based on the incision function,a revision algorithm is proposed to apply ILP to all minimal incoherence-preserving subsets(MIPS).Although this algorithm can often find a minimal number of axioms to remove,it is very time-consuming to compute MIPS.Thus,an adapted revision algorithm to deal with unsatisfiable concepts individually is also given.Experimental results reveal that the proposed ILP-based revision algorithm is much more efficient than the commonly used algorithm based on the hitting set tree.In addition,the adapted algorithm can achieve higher efficiency,while it may delete more axioms.展开更多
To solve the ambiguity and uncertainty in the labeling process of power equipment corrosion datasets,a novel hierarchical annotation method(HAM)is proposed.Firstly,large boxes are used to label a large area covering t...To solve the ambiguity and uncertainty in the labeling process of power equipment corrosion datasets,a novel hierarchical annotation method(HAM)is proposed.Firstly,large boxes are used to label a large area covering the range of corrosion,provided that the area is visually continuous and adjacent to corrosion that cannot be clearly divided.Secondly,in each labeling box established in the first step,regions with distinct corrosion and relative independence are labeled to form a second layer of nested boxes.Finally,a series of comparative experiments are conducted with other common annotation methods to validate the effectiveness of HAM.The experimental results show that,with the help of HAM,the recall of YOLOv5 increases from 50.79%to 59.41%;the recall of Faster R-CNN+VGG16 increases from 66.50%to 78.94%;the recall of Faster R-CNN+Res101 increases from 78.32%to 84.61%.Therefore,HAM can effectively improve the detection ability of mainstream models in detecting metal corrosion.展开更多
基金The State Grid Technology Project(No.5108202340042A-1-1-ZN).
文摘To solve the low efficiency of approximate queries caused by the large sizes of the knowledge graphs in the real world,an embedding-based approximate query method is proposed.First,the nodes in the query graph are classified according to the degrees of approximation required for different types of nodes.This classification transforms the query problem into three constraints,from which approximate information is extracted.Second,candidates are generated by calculating the similarity between embeddings.Finally,a deep neural network model is designed,incorporating a loss function based on the high-dimensional ellipsoidal diffusion distance.This model identifies the distance between nodes using their embeddings and constructs a score function.k nodes are returned as the query results.The results show that the proposed method can return both exact results and approximate matching results.On datasets DBLP(DataBase systems and Logic Programming)and FUA-S(Flight USA Airports-Sparse),this method exhibits superior performance in terms of precision and recall,returning results in 0.10 and 0.03 s,respectively.This indicates greater efficiency compared to PathSim and other comparative methods.
基金The National Natural Science Foundation of China(No.61502095).
文摘Aiming at the relation linking task for question answering over knowledge base,especially the multi relation linking task for complex questions,a relation linking approach based on the multi-attention recurrent neural network(RNN)model is proposed,which works for both simple and complex questions.First,the vector representations of questions are learned by the bidirectional long short-term memory(Bi-LSTM)model at the word and character levels,and named entities in questions are labeled by the conditional random field(CRF)model.Candidate entities are generated based on a dictionary,the disambiguation of candidate entities is realized based on predefined rules,and named entities mentioned in questions are linked to entities in knowledge base.Next,questions are classified into simple or complex questions by the machine learning method.Starting from the identified entities,for simple questions,one-hop relations are collected in the knowledge base as candidate relations;for complex questions,two-hop relations are collected as candidates.Finally,the multi-attention Bi-LSTM model is used to encode questions and candidate relations,compare their similarity,and return the candidate relation with the highest similarity as the result of relation linking.It is worth noting that the Bi-LSTM model with one attentions is adopted for simple questions,and the Bi-LSTM model with two attentions is adopted for complex questions.The experimental results show that,based on the effective entity linking method,the Bi-LSTM model with the attention mechanism improves the relation linking effectiveness of both simple and complex questions,which outperforms the existing relation linking methods based on graph algorithm or linguistics understanding.
基金The National Natural Science Foundation of China(No.61602259,U1736204)Research Foundation for Advanced Talents of Nanjing University of Posts and Telecommunications(No.NY216022)the National Key Research and Development Program of China(No.2018YFC0830200).
文摘To revise stratified web ontology language(OWL)ontologies,the kernel revision operator is extended by defining novel conflict stratification and the incision function based on integer linear programming(ILP).The ILP-based model considers an optimization problem of minimizing a linear objective function which is suitable for selecting the minimal number of axioms to remove when revising ontologies.Based on the incision function,a revision algorithm is proposed to apply ILP to all minimal incoherence-preserving subsets(MIPS).Although this algorithm can often find a minimal number of axioms to remove,it is very time-consuming to compute MIPS.Thus,an adapted revision algorithm to deal with unsatisfiable concepts individually is also given.Experimental results reveal that the proposed ILP-based revision algorithm is much more efficient than the commonly used algorithm based on the hitting set tree.In addition,the adapted algorithm can achieve higher efficiency,while it may delete more axioms.
基金The National Key R&D Program of China(No.2018YFC0830200)the Open Research Fund from State Key Laboratory of Smart Grid Protection and Control(No.NARI-T-2-2019189)+1 种基金Rapid Support Project(No.61406190120)the Fundamental Research Funds for the Central Universities(No.2242021k10011).
文摘To solve the ambiguity and uncertainty in the labeling process of power equipment corrosion datasets,a novel hierarchical annotation method(HAM)is proposed.Firstly,large boxes are used to label a large area covering the range of corrosion,provided that the area is visually continuous and adjacent to corrosion that cannot be clearly divided.Secondly,in each labeling box established in the first step,regions with distinct corrosion and relative independence are labeled to form a second layer of nested boxes.Finally,a series of comparative experiments are conducted with other common annotation methods to validate the effectiveness of HAM.The experimental results show that,with the help of HAM,the recall of YOLOv5 increases from 50.79%to 59.41%;the recall of Faster R-CNN+VGG16 increases from 66.50%to 78.94%;the recall of Faster R-CNN+Res101 increases from 78.32%to 84.61%.Therefore,HAM can effectively improve the detection ability of mainstream models in detecting metal corrosion.
