Research on Chinese Semantic Relation Extraction in Marine Engine Rooms Based on Multi-Feature Fusion

This research addresses the challenges of weak dependency relation learning caused by excessive contextual distances between segmented words in extracting entity relations within marine engine room knowledge graphs. Specifically, we have developed an advanced model that integrates an enhanced Chines...

Full description

Saved in:
Bibliographic Details
Main Authors: Xicai Liu, Zhengquan Wang, Fubo Wang
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10804127/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850122036930674688
author Xicai Liu
Zhengquan Wang
Fubo Wang
author_facet Xicai Liu
Zhengquan Wang
Fubo Wang
author_sort Xicai Liu
collection DOAJ
description This research addresses the challenges of weak dependency relation learning caused by excessive contextual distances between segmented words in extracting entity relations within marine engine room knowledge graphs. Specifically, we have developed an advanced model that integrates an enhanced Chinese syntactic structure with multi-feature fusion for extracting Chinese semantic relationships in engine rooms. Firstly, we construct a relational structure graph using a syntactic dependency tree based on the dependency relations among segmented words. This graph is then transformed into a character adjacency matrix. To incorporate syntactic graph structural features, we process this matrix and BERT-encoded embeddings using a graph convolutional neural network (GCN). Moreover, we utilize an attention mechanism to combine the syntactic graph structural features with the context features extracted by BERT, resulting in a multi-feature fused representation. Finally, we utilize this representation to extract entity relations by training a relation selector through reinforcement learning to optimize the relation embeddings and enhance the accuracy of relation judgment. Experimental results on a Chinese semantic dataset for marine engine rooms show that our model achieves an F1 score of 87.64%, outperforming several baseline models. These findings highlight that the fusion of semantic and syntactic structural features enhances the model’s informational content and interpretative capabilities.
format Article
id doaj-art-117ec000f97d49c7a957b3ebb393857a
institution OA Journals
issn 2169-3536
language English
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-117ec000f97d49c7a957b3ebb393857a2025-08-20T02:34:56ZengIEEEIEEE Access2169-35362024-01-011219201319202710.1109/ACCESS.2024.351861410804127Research on Chinese Semantic Relation Extraction in Marine Engine Rooms Based on Multi-Feature FusionXicai Liu0https://orcid.org/0009-0007-0144-3785Zhengquan Wang1Fubo Wang2School of Ocean, Yantai University, Yantai, ChinaYantai CIMC Raffles Ocean Technology Group, Yantai, ChinaYantai CIMC Raffles Ocean Technology Group, Yantai, ChinaThis research addresses the challenges of weak dependency relation learning caused by excessive contextual distances between segmented words in extracting entity relations within marine engine room knowledge graphs. Specifically, we have developed an advanced model that integrates an enhanced Chinese syntactic structure with multi-feature fusion for extracting Chinese semantic relationships in engine rooms. Firstly, we construct a relational structure graph using a syntactic dependency tree based on the dependency relations among segmented words. This graph is then transformed into a character adjacency matrix. To incorporate syntactic graph structural features, we process this matrix and BERT-encoded embeddings using a graph convolutional neural network (GCN). Moreover, we utilize an attention mechanism to combine the syntactic graph structural features with the context features extracted by BERT, resulting in a multi-feature fused representation. Finally, we utilize this representation to extract entity relations by training a relation selector through reinforcement learning to optimize the relation embeddings and enhance the accuracy of relation judgment. Experimental results on a Chinese semantic dataset for marine engine rooms show that our model achieves an F1 score of 87.64%, outperforming several baseline models. These findings highlight that the fusion of semantic and syntactic structural features enhances the model’s informational content and interpretative capabilities.https://ieeexplore.ieee.org/document/10804127/Marine engine roomrelation extractionBERT modelgraph convolutional neural networkmulti-feature fusion
spellingShingle Xicai Liu
Zhengquan Wang
Fubo Wang
Research on Chinese Semantic Relation Extraction in Marine Engine Rooms Based on Multi-Feature Fusion
IEEE Access
Marine engine room
relation extraction
BERT model
graph convolutional neural network
multi-feature fusion
title Research on Chinese Semantic Relation Extraction in Marine Engine Rooms Based on Multi-Feature Fusion
title_full Research on Chinese Semantic Relation Extraction in Marine Engine Rooms Based on Multi-Feature Fusion
title_fullStr Research on Chinese Semantic Relation Extraction in Marine Engine Rooms Based on Multi-Feature Fusion
title_full_unstemmed Research on Chinese Semantic Relation Extraction in Marine Engine Rooms Based on Multi-Feature Fusion
title_short Research on Chinese Semantic Relation Extraction in Marine Engine Rooms Based on Multi-Feature Fusion
title_sort research on chinese semantic relation extraction in marine engine rooms based on multi feature fusion
topic Marine engine room
relation extraction
BERT model
graph convolutional neural network
multi-feature fusion
url https://ieeexplore.ieee.org/document/10804127/
work_keys_str_mv AT xicailiu researchonchinesesemanticrelationextractioninmarineengineroomsbasedonmultifeaturefusion
AT zhengquanwang researchonchinesesemanticrelationextractioninmarineengineroomsbasedonmultifeaturefusion
AT fubowang researchonchinesesemanticrelationextractioninmarineengineroomsbasedonmultifeaturefusion