2022 Fiscal Year Final Research Report
Fundamnetal techniques for knowledge discovery based on local importance
| Project/Area Number |
19K20350
|
|---|
| Research Category |
Grant-in-Aid for Early-Career Scientists
|
| Allocation Type | Multi-year Fund |
|---|
| Review Section |
Basic Section 61030:Intelligent informatics-related
|
|---|
| Research Institution | Hosei University (2022)
Toyohashi University of Technology (2020-2021)
National Institute of Informatics (2019) |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2019-04-01 – 2023-03-31
|
| Keywords | 列挙アルゴリズム / データマイニング / 近傍領域 |
|---|
| Outline of Final Research Achievements |
Throughout the entire period, we were able to get a basic idea for a framework for developing enumeration or reconfiguration algorithms. However, the other goal, that is, revealing "what is the importance measure of problems" did not achieve. The specific reason for this may come from the difficulty of meeting people in the other field due to the extreme decrease in human interaction caused by the Corona disaster. Recently, when we develop an algorithm, users ask us to clarify "why such a solution is outputted?" This kind of explainability is highly demanded such as k-best enumeration, which have non-explicit indicators behind them as well as machine learning. We will continue to address this topic after the completion of this project.
|
| Free Research Field |
離散アルゴリズム
|
| Academic Significance and Societal Importance of the Research Achievements |
列挙問題，つまり，一定条件を満たした解を全て列挙する問題はさまざまな分野において見られる基礎的な問題である．本研究成果の学術的意義は，本課題の一つの課題である種々の問題に対する列挙アルゴリズムの構築を実際に行った点である．一方で，このアルゴリズムを利用して"解きたい問題"においては何に着目したら良いかを導き出す，というもう一つの課題については成果を上げられなかった．この点については，今後同様の問題意識を持って取り組み成果を上げていきたい．
|
