Developing Nonlinear Feature Selection Algorithm for Ultra High-Dimensional Data
| Project/Area Number |
16K16114
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Intelligent informatics
|
|---|
| Research Institution | Institute of Physical and Chemical Research (2017)
Kyoto University (2016) |
|---|
Principal Investigator |
Yamada Makoto 国立研究開発法人理化学研究所, 革新知能統合研究センター, ユニットリーダー (00581323)
|
|---|
| Project Period (FY) |
2016-04-01 – 2018-03-31
|
| Project Status |
Completed (Fiscal Year 2017)
|
| Budget Amount *help |
¥1,950,000 (Direct Cost: ¥1,500,000、Indirect Cost: ¥450,000)
Fiscal Year 2017: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2016: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
|
|---|
| Keywords | 特徴選択 / 非線形 / 機械学習 |
|---|
| Outline of Final Research Achievements |
We have developed a nonlinear feature selection algorithm for ultra-high dimensional data (more than 1 million features with tens of thousand data samples). To the best of our knowledge, this is the first algorithm that scales to such data. Moreover, for non-machine learning researchers, we developed a python package "pyHSICLasso" and distributed the code through Github. Now, we can install the software using "pip install pyHSICLasso". Finally, our research paper entitled "Ultra High-Dimensional Nonlinear Feature
Selection for Big Biological Data" was accepted to a top-tier data mining journal IEEE Transactions on Knowledge and Data Engineering (TKDE).
|
Report
(3 results)
Research Products
(15 results)
