Mining heuristic evidence sentences for more interpretable document-level relation extraction
Current research on evidence sentences is aimed at developing document-level relational extraction models with improved interpretability. Evidence sentences extracted using existing methods are often incomplete, leading to poor relationship prediction accuracy. To address this problem, we developed...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-07-01
|
| Series: | Journal of King Saud University: Computer and Information Sciences |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1319157823001970 |
| _version_ | 1797749208258707456 |
|---|---|
| author | Taojie Zhu Jicang Lu Gang Zhou Xiaoyao Ding Panpan Guo Hao Wu |
| author_facet | Taojie Zhu Jicang Lu Gang Zhou Xiaoyao Ding Panpan Guo Hao Wu |
| author_sort | Taojie Zhu |
| collection | DOAJ |
| description | Current research on evidence sentences is aimed at developing document-level relational extraction models with improved interpretability. Evidence sentences extracted using existing methods are often incomplete, leading to poor relationship prediction accuracy. To address this problem, we developed a novel efficient heuristic rule and entity representation method. First, a heuristic rule is constructed according to the interactions between different mentions of the head and tail entities of the target entity pair, and evidence sentences are subsequently extracted. Second, pseudo documents, constructed according to the original document order, are used as input text to remove noisy statements. Finally, different representations of the same entity in different entity pairs are learned to represent it more accurately through the interactive mention of head and tail entities. Experiments on the document-level general domain dataset DocRED indicated that our heuristic rules improved sentence extraction by 6.01% compared to that achieved by the baseline model Paths-BiLSTM. In terms of relation prediction, the accuracy of the proposed method was comparable to those of existing models that use the entire document as input text; however, the input text used by the proposed method was shorter and more interpretable. |
| first_indexed | 2024-03-12T16:15:57Z |
| format | Article |
| id | doaj.art-e8db03f584bf48c490451f189a5b64b4 |
| institution | Directory Open Access Journal |
| issn | 1319-1578 |
| language | English |
| last_indexed | 2024-03-12T16:15:57Z |
| publishDate | 2023-07-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Journal of King Saud University: Computer and Information Sciences |
| spelling | doaj.art-e8db03f584bf48c490451f189a5b64b42023-08-09T04:32:08ZengElsevierJournal of King Saud University: Computer and Information Sciences1319-15782023-07-01357101643Mining heuristic evidence sentences for more interpretable document-level relation extractionTaojie Zhu0Jicang Lu1Gang Zhou2Xiaoyao Ding3Panpan Guo4Hao Wu5State Key Laboratory of Mathematical Engineering and Advanced Computing, Zhengzhou 450001, ChinaCorresponding author.; State Key Laboratory of Mathematical Engineering and Advanced Computing, Zhengzhou 450001, ChinaState Key Laboratory of Mathematical Engineering and Advanced Computing, Zhengzhou 450001, ChinaState Key Laboratory of Mathematical Engineering and Advanced Computing, Zhengzhou 450001, ChinaState Key Laboratory of Mathematical Engineering and Advanced Computing, Zhengzhou 450001, ChinaState Key Laboratory of Mathematical Engineering and Advanced Computing, Zhengzhou 450001, ChinaCurrent research on evidence sentences is aimed at developing document-level relational extraction models with improved interpretability. Evidence sentences extracted using existing methods are often incomplete, leading to poor relationship prediction accuracy. To address this problem, we developed a novel efficient heuristic rule and entity representation method. First, a heuristic rule is constructed according to the interactions between different mentions of the head and tail entities of the target entity pair, and evidence sentences are subsequently extracted. Second, pseudo documents, constructed according to the original document order, are used as input text to remove noisy statements. Finally, different representations of the same entity in different entity pairs are learned to represent it more accurately through the interactive mention of head and tail entities. Experiments on the document-level general domain dataset DocRED indicated that our heuristic rules improved sentence extraction by 6.01% compared to that achieved by the baseline model Paths-BiLSTM. In terms of relation prediction, the accuracy of the proposed method was comparable to those of existing models that use the entire document as input text; however, the input text used by the proposed method was shorter and more interpretable.http://www.sciencedirect.com/science/article/pii/S1319157823001970Document-level relation extractionHeuristic rulesEvidence sentencesEntity representation enhancement |
| spellingShingle | Taojie Zhu Jicang Lu Gang Zhou Xiaoyao Ding Panpan Guo Hao Wu Mining heuristic evidence sentences for more interpretable document-level relation extraction Journal of King Saud University: Computer and Information Sciences Document-level relation extraction Heuristic rules Evidence sentences Entity representation enhancement |
| title | Mining heuristic evidence sentences for more interpretable document-level relation extraction |
| title_full | Mining heuristic evidence sentences for more interpretable document-level relation extraction |
| title_fullStr | Mining heuristic evidence sentences for more interpretable document-level relation extraction |
| title_full_unstemmed | Mining heuristic evidence sentences for more interpretable document-level relation extraction |
| title_short | Mining heuristic evidence sentences for more interpretable document-level relation extraction |
| title_sort | mining heuristic evidence sentences for more interpretable document level relation extraction |
| topic | Document-level relation extraction Heuristic rules Evidence sentences Entity representation enhancement |
| url | http://www.sciencedirect.com/science/article/pii/S1319157823001970 |
| work_keys_str_mv | AT taojiezhu miningheuristicevidencesentencesformoreinterpretabledocumentlevelrelationextraction AT jicanglu miningheuristicevidencesentencesformoreinterpretabledocumentlevelrelationextraction AT gangzhou miningheuristicevidencesentencesformoreinterpretabledocumentlevelrelationextraction AT xiaoyaoding miningheuristicevidencesentencesformoreinterpretabledocumentlevelrelationextraction AT panpanguo miningheuristicevidencesentencesformoreinterpretabledocumentlevelrelationextraction AT haowu miningheuristicevidencesentencesformoreinterpretabledocumentlevelrelationextraction |