2014 Fiscal Year Final Research Report
Generative double articulation analyzer based on nonparametric Bayesian approach
| Project/Area Number |
24700233
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Sensitivity informatics/Soft computing
|
|---|
| Research Institution | Ritsumeikan University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2012-04-01 – 2015-03-31
|
| Keywords | 機械学習 / 時系列解析 / ノンパラメトリックベイズ |
|---|
| Outline of Final Research Achievements |
In this project, we successfully developed a nonparametric Bayesian double articulation analyzer. The analyzer integrates two inference processes which were previously treated as different learning processes. One is a segmentation process and the other is a chunking process. To develop the learning method, we proposed an integrated generative model and derived efficient blocked Gibbs sampling procedure. In addition to that, we developed various methods related to driver support system by using a double articulation analyzer.
|
| Free Research Field |
創発システム論
|
