| Project/Area Number |
17200009
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (A)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Intelligent informatics
|
|---|
| Research Institution | Keio University |
|---|
Principal Investigator |
FURUKAWA Koichia Keio University, Graduate School of Media and Governance, Professor (10245615)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
FUJINAMI Tsutomu Japan Advanced Institute of Science and Techchnology, 知識科学研究科, Associate Professor (70303344)
OHGI Yuji Keio University, Graduate School of Media and Governance, Associate Professor (90317313)
SUWA Masaki Chukyo University, School of Information Science and Technology, Professor (50329661)
FUKUDA Ryoko Keio University, Faculty of Environment and Information studies, Assistant Professor (80383917)
KATO Takaki Keio University, Faculty of Environment and Information studies, Assistant Professor (30365481)
|
|---|
| Project Period (FY) |
2005 – 2007
|
| Project Status |
Completed (Fiscal Year 2007)
|
| Budget Amount *help |
¥41,080,000 (Direct Cost: ¥31,600,000、Indirect Cost: ¥9,480,000)
Fiscal Year 2007: ¥7,800,000 (Direct Cost: ¥6,000,000、Indirect Cost: ¥1,800,000)
Fiscal Year 2006: ¥7,800,000 (Direct Cost: ¥6,000,000、Indirect Cost: ¥1,800,000)
Fiscal Year 2005: ¥25,480,000 (Direct Cost: ¥19,600,000、Indirect Cost: ¥5,880,000)
|
|---|
| Keywords | skill science / verbalization of tacit knowledge / time series data mining / motion integrity constraints / meta cognition / physical models / abductive inference / skill discovery and diagnosis / 属性発見 / 身体スキル / スキルの力学モデル / 鞭力学 / 体のインピーダンスの調節 / 帰納論理プログラミング / 着眼点の発見 / 人工知能 / 認知科学 / 機械学習 / 感性情報学 / スキル科学 |
|---|
| Research Abstract |
We conducted this research project aiming to establish methodologies of verbalizing physical skills by means of knowledge discovery technologies. We selected cello bowing and batting as our research targets. More concretely, we conducted researches on (1) developing physical models of body skills and discovering key points, (2) extracting skill rules by means of time series data mining, (3) extracting key points by means of meta cognition and (4) building a skill discovery and diagnosis support system based on abduction.
In (1), we developed a pendulum model, a rotation model and a whip model to describe human skill motion and extracted key points from the developed models. In (2), we succeeded in extracting positive skill rules such as propagating motions from the body trunk to extremities. We also extracted motion integrity constraints as negative skill rules. In (3), we continuously collected self analysis in performing batting and observed the change of focus in consciousness and relation between the change of focus and rapid skill improvement. In (4), we designed a skill discovery and diagnosis support system based on abductive inference system PrologClA in the case of cello playing and we succeeded in showing the feasibility of the approach.
Furthermore, we held an International Symposium on Skill Science 2007 partly to publicize our research results of this project and we provided an opportunity for presenting research results as well as to discuss future direction of the skill science through the panel discussion.
|
