| Project/Area Number |
16300065
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Sensitivity informatics/Soft computing
|
|---|
| Research Institution | University of Tsukuba |
|---|
Principal Investigator |
MIYAMOTO Sadaaki University of Tsukuba, Graduate School of Systems and Information Engineering, Professor, 大学院システム情報工学研究科, 教授 (60143179)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
SATO-ILIC Mika University of Tsukuba, Graduate School of Systems and Information Engineering, Associate Professor, 大学院システム情報工学研究科, 助教授 (60269214)
ENDO Yasunori University of Tsukuba, Graduate School of Systems and Information Engineering, Associate Professor, 大学院システム情報工学研究科, 助教授 (10267396)
ICHIHASHI Hidetomo University of Osaka Prefecture, Graduate School of Engineering, Professor, 大学院工学研究科, 教授 (30151476)
HONDA Katsuhiro University of Osaka Prefecture, Graduate School of Engineering, Research Associate, 大学院工学研究科, 助手 (80332964)
|
|---|
| Project Period (FY) |
2004 – 2006
|
| Project Status |
Completed (Fiscal Year 2006)
|
| Budget Amount *help |
¥14,100,000 (Direct Cost: ¥14,100,000)
Fiscal Year 2006: ¥4,100,000 (Direct Cost: ¥4,100,000)
Fiscal Year 2005: ¥4,100,000 (Direct Cost: ¥4,100,000)
Fiscal Year 2004: ¥5,900,000 (Direct Cost: ¥5,900,000)
|
|---|
| Keywords | fuzzy system / cluster analysis / learning algorithm / possibilistic clustering / multivariate analysis / data mining / information retrieval / cluster validity / ファジィc-平均法 / クリスプc-平均法 / カーネル関数 / 独立成分分析 / 許容範囲付きデータ / 動画検索 / カーネルトリック / 競合学習 / 不確実データ / 多次元尺度法 / 重み付き最小二乗法 / 主成分分析 / クリスプクラスタリング / ファジィクラスタリング / 自己組織化マップ |
|---|
| Research Abstract |
The aim of this project is to promote researches in methods of fuzzy clustering and related algorithms, whereby establishing a small scale 'center of excellence' which is known among researchers of this field of study in the world. To attain this goal, two activities have been done : one is to study various aspects of fuzzy and related clustering algorithms with applications ; the other is to make our studies known throughout the world. Concretely, we have done the next investigations for the former purpose :
(1) investigations of crisp and fuzzy clustering algorithms with nonlinear cluster boundaries,
(2) development of learning algorithms for clustering and combination with crisp/fuzzy clustering algorithms
(3) the kernel method : its fundamentals and applications to clustering algorithms
(4) combination between a clustering algorithms and methods in multivariate analysis
(5) applications to data mining
(6) applications to information and image retrieval
(7) algorithms in possibilistic clus
… More
tering
(8) clustering algorithms for data with tolerance
(9) visualization of clusters in high-dimensional spaces
(10) generalizations of validity functions for clusters and application to evaluation of different clustering algorithms
For the second goal to spread our results, we have organized many special sessions :
(1) we have organized special sessions of clustering in FUZZ-IEEE2004, 2005, and 2006 ; in FUZZ-IEEE2006, we have collected 18 papers including our own papers in 3 sessions that are comparable with ordinary sessions.
(2) We have organized special sessions in a number of other international conferences : KES, NOLTA, SCIS, MDAI and collected many papers not only those from our groups but from throughout the world,
(3) In Fuzzy System Symposium 2004, 2005, and 2006 in Japan, we have organized special sessions in clustering. A major part of studies in clustering in Japan have been presented in these sessions.
(4) We have prepared a manuscript for a monograph in clustering and made a contract with Springer. We will publish this book in next year.
Finally, the research report was made that collects major part of our studies in the three years. Less
|
