Temporal Knowledge Graph Construction from Text
Project/Area Number |
21K17816
|
Research Category |
Grant-in-Aid for Early-Career Scientists
|
Allocation Type | Multi-year Fund |
Review Section |
Basic Section 61030:Intelligent informatics-related
|
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 – 2024-03-31
|
Project Status |
Granted (Fiscal Year 2022)
|
Budget Amount *help |
¥3,510,000 (Direct Cost: ¥2,700,000、Indirect Cost: ¥810,000)
Fiscal Year 2022: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2021: ¥2,340,000 (Direct Cost: ¥1,800,000、Indirect Cost: ¥540,000)
|
Keywords | Temporal Knowledge Graph / Knowledge Extraction / Financial Report / Financial Application / Knowledge Graph / Knowledge Representation / Relation Extraction / Event Extraction |
Outline of Research at the Start |
Knowledge Graph plays a key role in various artificial intelligence applications. Generally, a knowledge graph is built to express static knowledge. Nevertheless, knowledge usually changes over time. A knowledge graph without considering the time does not satisfy the change. Therefore, this project aims to study a temporal knowledge graph and to develop a novel framework for constructing temporal knowledge graphs from text. The outcomes of this research will expand the study on the knowledge graph area and facilitate the time-aware knowledge graph-based applications.
|
Outline of Annual Research Achievements |
This research aims to build a temporal knowledge graph from the text. In this fiscal year, we constructed FinKG, the temporal knowledge graph in the financial domain using information reported from SEC and market exchange. The temporal information including the report detail and stock price, is extracted using our financial ontology as a template. The temporal information is encoded at the edge of the knowledge graph. Overall, FinKG contained more than 30 million facts. Furthermore, we demonstrate the usefulness of FinKG with two applications: knowledge retrieval and stock price prediction. Knowledge retrieval reveals the complex connection among entities, while aggregated features from FinKG help neural models to better forecast stock prices.
|
Current Status of Research Progress |
Current Status of Research Progress
3: Progress in research has been slightly delayed.
Reason
The research project progressed as scheduled. We built the temporal knowledge graph as planned. Also, we could demonstrate the application of temporal knowledge graph in financial domain.
|
Strategy for Future Research Activity |
Currently, we use the rule-based method to extract the temporal clue from text. In the future, We plan to further investigate and improve the temporal clue extraction from text.
|
Report
(2 results)
Research Products
(3 results)