Learning from Knowledge Graphs: Neural Fine-Grained Entity Typing with Copy-Generation Networks
Fine-grained entity typing (FET) aims to identify the semantic type of an entity in a plain text, which is a significant task for downstream natural language processing applications. However, most existing methods neglect rich known typing information about these entities in knowledge graphs. To add...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-07-01
|
| Series: | Entropy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1099-4300/24/7/964 |
| _version_ | 1797439596135448576 |
|---|---|
| author | Zongjian Yu Anxiang Zhang Huali Feng Huaming Du Shaopeng Wei Yu Zhao |
| author_facet | Zongjian Yu Anxiang Zhang Huali Feng Huaming Du Shaopeng Wei Yu Zhao |
| author_sort | Zongjian Yu |
| collection | DOAJ |
| description | Fine-grained entity typing (FET) aims to identify the semantic type of an entity in a plain text, which is a significant task for downstream natural language processing applications. However, most existing methods neglect rich known typing information about these entities in knowledge graphs. To address this issue, we take advantage of knowledge graphs to improve fine-grained entity typing through the use of a copy mechanism. Specifically, we propose a novel deep neural model called <span style="font-variant: small-caps;">CopyFet</span> for FET via a copy-generation mechanism. <span style="font-variant: small-caps;">CopyFet</span> can integrate two operations: (<b>i</b>) the regular way of making type inference from the whole type set in the generation model; (<b>ii</b>) the new copy mechanism which can identify the semantic type of a mention with reference to the type-copying vocabulary from a knowledge graph in the copy model. Despite its simplicity, this mechanism proves to be powerful since extensive experiments show that <span style="font-variant: small-caps;">CopyFet</span> outperforms state-of-the-art methods in FET on two benchmark datasets (FIGER (GOLD) and BBN). For example, <span style="font-variant: small-caps;">CopyFet</span> achieves the new state-of-the-art score of 76.4% and 83.6% on the accuracy metric in FIGER (GOLD) and BBN, respectively. |
| first_indexed | 2024-03-09T11:55:22Z |
| format | Article |
| id | doaj.art-384ec003280b4a38b524c1790507546a |
| institution | Directory Open Access Journal |
| issn | 1099-4300 |
| language | English |
| last_indexed | 2024-03-09T11:55:22Z |
| publishDate | 2022-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Entropy |
| spelling | doaj.art-384ec003280b4a38b524c1790507546a2023-11-30T23:09:34ZengMDPI AGEntropy1099-43002022-07-0124796410.3390/e24070964Learning from Knowledge Graphs: Neural Fine-Grained Entity Typing with Copy-Generation NetworksZongjian Yu0Anxiang Zhang1Huali Feng2Huaming Du3Shaopeng Wei4Yu Zhao5Financial Intelligence and Financial Engineering Key Laboratory of Sichuan Province, Southwestern University of Finance and Economics, Chengdu 611130, ChinaSchool of Computer Science, Carnegie Mellon University, Pittsburgh, PA 15213, USAFinancial Intelligence and Financial Engineering Key Laboratory of Sichuan Province, Southwestern University of Finance and Economics, Chengdu 611130, ChinaFinancial Intelligence and Financial Engineering Key Laboratory of Sichuan Province, Southwestern University of Finance and Economics, Chengdu 611130, ChinaFinancial Intelligence and Financial Engineering Key Laboratory of Sichuan Province, Southwestern University of Finance and Economics, Chengdu 611130, ChinaFinancial Intelligence and Financial Engineering Key Laboratory of Sichuan Province, Southwestern University of Finance and Economics, Chengdu 611130, ChinaFine-grained entity typing (FET) aims to identify the semantic type of an entity in a plain text, which is a significant task for downstream natural language processing applications. However, most existing methods neglect rich known typing information about these entities in knowledge graphs. To address this issue, we take advantage of knowledge graphs to improve fine-grained entity typing through the use of a copy mechanism. Specifically, we propose a novel deep neural model called <span style="font-variant: small-caps;">CopyFet</span> for FET via a copy-generation mechanism. <span style="font-variant: small-caps;">CopyFet</span> can integrate two operations: (<b>i</b>) the regular way of making type inference from the whole type set in the generation model; (<b>ii</b>) the new copy mechanism which can identify the semantic type of a mention with reference to the type-copying vocabulary from a knowledge graph in the copy model. Despite its simplicity, this mechanism proves to be powerful since extensive experiments show that <span style="font-variant: small-caps;">CopyFet</span> outperforms state-of-the-art methods in FET on two benchmark datasets (FIGER (GOLD) and BBN). For example, <span style="font-variant: small-caps;">CopyFet</span> achieves the new state-of-the-art score of 76.4% and 83.6% on the accuracy metric in FIGER (GOLD) and BBN, respectively.https://www.mdpi.com/1099-4300/24/7/964knowledge graphsfine-grained entity typingcopy-generation networkscross-entropy |
| spellingShingle | Zongjian Yu Anxiang Zhang Huali Feng Huaming Du Shaopeng Wei Yu Zhao Learning from Knowledge Graphs: Neural Fine-Grained Entity Typing with Copy-Generation Networks Entropy knowledge graphs fine-grained entity typing copy-generation networks cross-entropy |
| title | Learning from Knowledge Graphs: Neural Fine-Grained Entity Typing with Copy-Generation Networks |
| title_full | Learning from Knowledge Graphs: Neural Fine-Grained Entity Typing with Copy-Generation Networks |
| title_fullStr | Learning from Knowledge Graphs: Neural Fine-Grained Entity Typing with Copy-Generation Networks |
| title_full_unstemmed | Learning from Knowledge Graphs: Neural Fine-Grained Entity Typing with Copy-Generation Networks |
| title_short | Learning from Knowledge Graphs: Neural Fine-Grained Entity Typing with Copy-Generation Networks |
| title_sort | learning from knowledge graphs neural fine grained entity typing with copy generation networks |
| topic | knowledge graphs fine-grained entity typing copy-generation networks cross-entropy |
| url | https://www.mdpi.com/1099-4300/24/7/964 |
| work_keys_str_mv | AT zongjianyu learningfromknowledgegraphsneuralfinegrainedentitytypingwithcopygenerationnetworks AT anxiangzhang learningfromknowledgegraphsneuralfinegrainedentitytypingwithcopygenerationnetworks AT hualifeng learningfromknowledgegraphsneuralfinegrainedentitytypingwithcopygenerationnetworks AT huamingdu learningfromknowledgegraphsneuralfinegrainedentitytypingwithcopygenerationnetworks AT shaopengwei learningfromknowledgegraphsneuralfinegrainedentitytypingwithcopygenerationnetworks AT yuzhao learningfromknowledgegraphsneuralfinegrainedentitytypingwithcopygenerationnetworks |