Development of feature selection methods for large-input, small-sample problems
| Project/Area Number |
25420438
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Control engineering/System engineering
|
|---|
| Research Institution | Kobe University |
|---|
Principal Investigator |
Abe Shigeo 神戸大学, 工学研究科, 名誉教授 (50294195)
|
|---|
| Project Period (FY) |
2013-04-01 – 2016-03-31
|
| Project Status |
Completed (Fiscal Year 2015)
|
| Budget Amount *help |
¥3,250,000 (Direct Cost: ¥2,500,000、Indirect Cost: ¥750,000)
Fiscal Year 2015: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2014: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2013: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
|
|---|
| Keywords | パターン認識 / 特徴選択 / ブロック追加 / ブロック削除 / サポートベクトルマシン |
|---|
| Outline of Final Research Achievements |
We developed feature selection methods for small-sample, large-input problems. To speed up the block addition and block deletion (BABD) methods developed previously, we developed the iterative BABD that repeatedly iterates BABD and incremental BABD that applies BABD to a block of features divided in advance. To speed up training of support vector machines that are used for feature selection, we developed the SMO-NM method that combines the sequential minimal optimization technique and the Newton Method. We evaluated the validity of the proposed methods using benchmark data sets.
|
Report
(4 results)
Research Products
(7 results)
