Human behavior modeling with hierarchical Bayesian models
| Project/Area Number |
20700181
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
Perception information processing/Intelligent robotics
|
|---|
| Research Institution | The University of Tokyo |
|---|
Principal Investigator |
SHIMOSAKA Masamichi The University of Tokyo, 大学院・情報理工学系研究科, 助教 (40431796)
|
|---|
| Project Period (FY) |
2008 – 2009
|
| Project Status |
Completed (Fiscal Year 2009)
|
| Budget Amount *help |
¥4,160,000 (Direct Cost: ¥3,200,000、Indirect Cost: ¥960,000)
Fiscal Year 2009: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
Fiscal Year 2008: ¥2,340,000 (Direct Cost: ¥1,800,000、Indirect Cost: ¥540,000)
|
|---|
| Keywords | 統計的情報処理 / 機械学習 / 行動モデリング |
|---|
| Research Abstract |
In this research, a statistical approach of human behavior modeling is proposed and verified. In this study, issues on automatic extraction of latent factors from human behavior data are focused. In contrast to the conventional human behavior modeling where behavior models are estimated independently from each individual, the proposed model, which is formulated with Dirichlet process priors, automatically couples similar human behavior and provides robust estimation from the coupled behavior data. Empirical evaluation using human behavior sensor data shows that our model improves better estimation accuracy.
|
Report
(3 results)
Research Products
(11 results)
