Graph Convolution Over Multiple Latent Context-Aware Graph Structures for Event Detection
Event detection is a particularly challenging problem in information extraction. The current neural network models have proved that dependency tree can better capture the correlation between candidate trigger words and related context in the sentence. However, syntactic information conveyed by the o...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9200483/ |
| _version_ | 1818934542967242752 |
|---|---|
| author | Lei Li Li Jin Zequn Zhang Qing Liu Xian Sun Hongqi Wang |
| author_facet | Lei Li Li Jin Zequn Zhang Qing Liu Xian Sun Hongqi Wang |
| author_sort | Lei Li |
| collection | DOAJ |
| description | Event detection is a particularly challenging problem in information extraction. The current neural network models have proved that dependency tree can better capture the correlation between candidate trigger words and related context in the sentence. However, syntactic information conveyed by the original dependency tree is insufficient for detecting trigger since the dependency tree obtained from natural language processing toolkits ignores semantic context information. Existing approaches employ a static graph structure based on original dependency tree which is incompetent in terms of distinguishing interrelations among trigger words and contextual words. So how to effectively make use of relevant information while ignoring irrelevant information from the dependency trees remains a challenging research question. To address this problem, we investigate a graph convolutional network over multiple latent context-aware graph structures to perform event detection. We exploit a multi-head attention mechanism on BERT representation and original adjacency matrix to generate multiple latent context-aware graph structures (a “dynamic cutting” strategy), which can automatically learn how to select the useful dependency information. Furthermore, we investigate graph convolutional networks with residual connections to combine the local and non-local contextual information. Experimental results on ACE2005 dataset show that our model achieves competitive performances compared with the methods based on dependency tree for event detection. |
| first_indexed | 2024-12-20T05:05:56Z |
| format | Article |
| id | doaj.art-64ee17e5b0624186828f83da0c296642 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-20T05:05:56Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-64ee17e5b0624186828f83da0c2966422022-12-21T19:52:24ZengIEEEIEEE Access2169-35362020-01-01817143517144610.1109/ACCESS.2020.30248729200483Graph Convolution Over Multiple Latent Context-Aware Graph Structures for Event DetectionLei Li0https://orcid.org/0000-0001-5660-0409Li Jin1https://orcid.org/0000-0001-8833-4862Zequn Zhang2https://orcid.org/0000-0002-2138-845XQing Liu3https://orcid.org/0000-0002-0592-171XXian Sun4https://orcid.org/0000-0002-0038-9816Hongqi Wang5https://orcid.org/0000-0002-5172-571XChinese Academy of Sciences, Aerospace Information Research Institute, Beijing, ChinaChinese Academy of Sciences, Aerospace Information Research Institute, Beijing, ChinaChinese Academy of Sciences, Aerospace Information Research Institute, Beijing, ChinaChinese Academy of Sciences, Aerospace Information Research Institute, Beijing, ChinaChinese Academy of Sciences, Aerospace Information Research Institute, Beijing, ChinaChinese Academy of Sciences, Aerospace Information Research Institute, Beijing, ChinaEvent detection is a particularly challenging problem in information extraction. The current neural network models have proved that dependency tree can better capture the correlation between candidate trigger words and related context in the sentence. However, syntactic information conveyed by the original dependency tree is insufficient for detecting trigger since the dependency tree obtained from natural language processing toolkits ignores semantic context information. Existing approaches employ a static graph structure based on original dependency tree which is incompetent in terms of distinguishing interrelations among trigger words and contextual words. So how to effectively make use of relevant information while ignoring irrelevant information from the dependency trees remains a challenging research question. To address this problem, we investigate a graph convolutional network over multiple latent context-aware graph structures to perform event detection. We exploit a multi-head attention mechanism on BERT representation and original adjacency matrix to generate multiple latent context-aware graph structures (a “dynamic cutting” strategy), which can automatically learn how to select the useful dependency information. Furthermore, we investigate graph convolutional networks with residual connections to combine the local and non-local contextual information. Experimental results on ACE2005 dataset show that our model achieves competitive performances compared with the methods based on dependency tree for event detection.https://ieeexplore.ieee.org/document/9200483/Event detectiongraph convolutional networkmulti-head attention |
| spellingShingle | Lei Li Li Jin Zequn Zhang Qing Liu Xian Sun Hongqi Wang Graph Convolution Over Multiple Latent Context-Aware Graph Structures for Event Detection IEEE Access Event detection graph convolutional network multi-head attention |
| title | Graph Convolution Over Multiple Latent Context-Aware Graph Structures for Event Detection |
| title_full | Graph Convolution Over Multiple Latent Context-Aware Graph Structures for Event Detection |
| title_fullStr | Graph Convolution Over Multiple Latent Context-Aware Graph Structures for Event Detection |
| title_full_unstemmed | Graph Convolution Over Multiple Latent Context-Aware Graph Structures for Event Detection |
| title_short | Graph Convolution Over Multiple Latent Context-Aware Graph Structures for Event Detection |
| title_sort | graph convolution over multiple latent context aware graph structures for event detection |
| topic | Event detection graph convolutional network multi-head attention |
| url | https://ieeexplore.ieee.org/document/9200483/ |
| work_keys_str_mv | AT leili graphconvolutionovermultiplelatentcontextawaregraphstructuresforeventdetection AT lijin graphconvolutionovermultiplelatentcontextawaregraphstructuresforeventdetection AT zequnzhang graphconvolutionovermultiplelatentcontextawaregraphstructuresforeventdetection AT qingliu graphconvolutionovermultiplelatentcontextawaregraphstructuresforeventdetection AT xiansun graphconvolutionovermultiplelatentcontextawaregraphstructuresforeventdetection AT hongqiwang graphconvolutionovermultiplelatentcontextawaregraphstructuresforeventdetection |