Name
Title | ||
|---|---|---|
Tebc-Net: An Effective Relation Extraction Approach For Simple Question Answering Over Knowledge Graphs |
Abstract | ||
|---|---|---|
Knowledge graph simple question answering (KGSQA) aims on answering natural language questions by the lookup of a single fact over a knowledge graph. As one of the core tasks in the scenarios, relation extraction is critical for the quality of final answers. To improve the accuracy of relation extraction in KGSQA, in this paper, we propose a new deep neural network model called TEBC-Net, which is constructed based on the combination of Transformer Encoder, BiLSTM and CNN Net in a seamless way. We give the detailed design of our approach and have conducted an experimental evaluation with a benchmark test. Our results demonstrate that TEBC-Net can achieve higher accuracy on relation extraction and question answering tasks in KGSQA, compared to some current methods including the state-of-the-art. |
Year | DOI | Venue |
|---|---|---|
2021 | 10.1007/978-3-030-82136-4_13 | KNOWLEDGE SCIENCE, ENGINEERING AND MANAGEMENT, PT I |
Keywords | DocType | Volume |
Knowledge graph simple question answering, Relation extraction, Natural language processing, Deep learning, TEBC-Net | Conference | 12815 |
ISSN | Citations | PageRank |
0302-9743 | 0 | 0.34 |
References | Authors | |
0 | 5 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Jianbin Li | 1 | 3 | 2.43 |
| Ketong Qu | 2 | 0 | 0.68 |
| Jingchen Yan | 3 | 0 | 1.01 |
| Liting Zhou | 4 | 0 | 0.34 |
| Long Cheng | 5 | 91 | 16.99 |