Extracting semantic link network of words from text for semantics-based applications

doi:10.1016/j.eswa.2024.125768

	Extracting semantic link network of words from text for semantics-based applications
	Li, Jiazheng 1,2; Zhou, Jian 1,2; Zhuge, Hai 3,4
	2025-03-05
发表期刊	EXPERT SYSTEMS WITH APPLICATIONS
ISSN	0957-4174
卷号	263 页码:20
摘要	Transforming text into a Semantic Link Network of Words (in short W-SLN) is an approach to extracting the basic semantics from text for supporting natural language processing applications based on basic semantics within text. This paper proposes an approach to extracting W-SLN with a set of reasoning rules on semantic links for deriving implicit semantic links. Different from previous relation extraction approaches, the approach to extracting WSLN is based on the following techniques: identification of nominal snippets for extracting semantic links, construction of semantic link patterns for extracting various types of relations from nominal snippets, and a general algorithm for automatically discovering reasoning rules with probabilities updated with inputting new texts. The approach achieves comparable performance with other rule-based and semi-supervised relation extraction approaches in terms of extraction recall and precision on the latest benchmark BenchIE. An unsupervised text summarization approach is developed by using W-SLN and a set of heuristic rules on W-SLN. Experiments on summarizing scientific papers, news and legal case reports show that the summarization approach outperforms other summarization baselines and can improve the richness, coherence, diversity and conciseness of the summary. Further, a framework based on W-SLN for question-answering on texts is introduced and verified through comparison with other approaches. The applications in text summarization and question answering applications demonstrate the generalizability of the W-SLN in supporting applications through the basic semantic representation of text.
关键词	Natural Language Processing Relation Extraction Semantic Link Network Text Summarization
DOI	10.1016/j.eswa.2024.125768
收录类别	SCI
语种	英语
资助项目	National Natural Science Foundation of China[61876048]
WOS研究方向	Computer Science ; Engineering ; Operations Research & Management Science
WOS类目	Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic ; Operations Research & Management Science
WOS记录号	WOS:001362895900001
出版者	PERGAMON-ELSEVIER SCIENCE LTD
引用统计
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/41159
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Zhuge, Hai
作者单位	1.Chinese Acad Sci, Key Lab Intelligent Informat Proc, Inst Comp Technol, Beijing, Peoples R China 2.Univ Chinese Acad Sci, Beijing, Peoples R China 3.Great Bay Univ, Sch Comp & Informat Technol, Dongguan 523000, Guangdong, Peoples R China 4.Great Bay Inst Adv Study, Dongguan 523000, Guangdong, Peoples R China
推荐引用方式 GB/T 7714	Li, Jiazheng,Zhou, Jian,Zhuge, Hai. Extracting semantic link network of words from text for semantics-based applications[J]. EXPERT SYSTEMS WITH APPLICATIONS,2025,263:20.
APA	Li, Jiazheng,Zhou, Jian,&Zhuge, Hai.(2025).Extracting semantic link network of words from text for semantics-based applications.EXPERT SYSTEMS WITH APPLICATIONS,263,20.
MLA	Li, Jiazheng,et al."Extracting semantic link network of words from text for semantics-based applications".EXPERT SYSTEMS WITH APPLICATIONS 263(2025):20.