2021 Fiscal Year Research-status Report
Temporal Knowledge Graph Construction from Text
| Project/Area Number |
21K17816
|
|---|
| Research Institution | Japan Advanced Institute of Science and Technology |
|---|
Principal Investigator |
K. Natthawut 北陸先端科学技術大学院大学, 先端科学技術研究科, 助教 (40818100)
|
|---|
| Project Period (FY) |
2021-04-01 – 2023-03-31
|
| Keywords | Temporal Knowledge Graph / Knowledge Extraction |
|---|
| Outline of Annual Research Achievements |
This research aims to propose a framework to construct a temporal knowledge graph from the text. So far, we have proposed the knowledge extraction framework for extracting useful knowledge from a text by using both semantic and syntactic features. Our framework is up to 20% improvement compared with the state-of-the-art methods. Also, we manually built the temporal knowledge graph as the bootstrapping training data and prepared the dataset for extracting temporal clues in the financial domain. We selected the financial domain for this study because the financial data are extremely sensitive to temporal information.
|
| Current Status of Research Progress |
Current Status of Research Progress
2: Research has progressed on the whole more than it was originally planned.
Reason
The research project is progressed as scheduled. The preliminary framework for extracting knowledge is implemented. Also, we obtained the data and the initial results for further experiments.
|
| Strategy for Future Research Activity |
We plan to utilize the collected data in the financial domain to build the model for the temporal knowledge graph construction framework. Also, we plan to improve the temporal clue extraction from text based on our collected data.
|
| Causes of Carryover |
We could not attend the international conference due to the spread of COVID-19. Considering this situation, we will use the travel expenses to buy computing and storage devices in order to carry out further experiments.
|
Research Products
(1 results)
