| Presentation | 2011-11-21 An Imputation of Context Data using Random Forest Tsunenori ISHIOKA, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | When considering contextware services, we set the response variable to the services to provide, and explanatory variable to the context information statistics. Usually, the explanatory variables contain some missing data. It is obvious that missing at random (MAR), which the missing depends on only observations not non-observations, is superior to missing completely at random (MCAR), which the missing does not depend on the variables in an assumed model. Random Forest (RF) is subject to the assumption of MAR, so it derive the better results than those by other conventional methods. The RF imputation can be activated since the Version 4. While being aware that RF is an ensemble learning method for the classification and/or non-linear regressions, many statistician and engineers do not know the availability of the missing data imputation. In this paper, we present the RF imputation algorithm, indicating that it works pretty well by comparing to kernel method on support vector machines. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | ensemble learning / Random Forest / data imputation / missing data / MAR |
| Paper # | AI2011-21 |
| Date of Issue |
| Conference Information | |
| Committee | AI |
|---|---|
| Conference Date | 2011/11/14(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Artificial Intelligence and Knowledge-Based Processing (AI) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | An Imputation of Context Data using Random Forest |
| Sub Title (in English) | |
| Keyword(1) | ensemble learning |
| Keyword(2) | Random Forest |
| Keyword(3) | data imputation |
| Keyword(4) | missing data |
| Keyword(5) | MAR |
| 1st Author's Name | Tsunenori ISHIOKA |
| 1st Author's Affiliation | Research Division, The National Center for University Entrance Examinations() |
| Date | 2011-11-21 |
| Paper # | AI2011-21 |
| Volume (vol) | vol.111 |
| Number (no) | 310 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |