Research on robust independent component analysis applicable to only a few observed signals
| Project/Area Number |
26730013
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Statistical science
|
|---|
| Research Institution | The University of Tokyo |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2014-04-01 – 2017-03-31
|
| Project Status |
Completed (Fiscal Year 2016)
|
| Budget Amount *help |
¥3,770,000 (Direct Cost: ¥2,900,000、Indirect Cost: ¥870,000)
Fiscal Year 2016: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2015: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
Fiscal Year 2014: ¥1,950,000 (Direct Cost: ¥1,500,000、Indirect Cost: ¥450,000)
|
|---|
| Keywords | 独立成分分析 / 機械学習 / 信号処理 / 統計科学 / ウェブデータ解析 / 自然言語処理 / 教育工学 / ウェブデータ分析 / 頑健性 / ウェブマイニング / カリキュラム分析 |
|---|
| Outline of Final Research Achievements |
Independent Component Analysis (ICA) is a method estimating the sources from only the mixed signals. ICA is widely used in machine learning and signal processing and is known to be useful in many applications such as image separation, sound separation, and feature extraction from images. In this research, we improved ICA and enabled it to order the sources uniquely. Moreover, we enabled ICA to separate the Gaussian noises so that we can estimate the number of the non-Gaussian sources. In addition, we applied some machine learning methods including ICA to the applications of web data analysis, natural language processing, and educational engineering.
|
Report
(4 results)
Research Products
(17 results)
