| Presentation | 2008-11-07 Averaging Naive Bayes Trees Mori KUROKAWA, Hiroyuki YOKOYAMA, Akito SAKURAI, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Naive Bayes (NB) is a simplified Bayesian classifier under the conditional independence assumption and Augmented Naive Bayes (ANB) models are extensions of NB by relaxing the assumption. Averaged One-Dependence Estimator (AODE) is an excellent ANB model with high classification accuracy. However, AODE limits the models' flexibility. In this paper, we propose averaging method of Naive Bayes Trees (NBTs) with flexible structures and show experimental results on classification accuracy. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Augmented Naive Bayes / Averaged One-Dependence Estimator (AODE) / Naive Bayes Tree (NBT) / model averaging |
| Paper # | NC2008-62 |
| Date of Issue |
| Conference Information | |
| Committee | NC |
|---|---|
| Conference Date | 2008/10/31(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 | Neurocomputing (NC) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Averaging Naive Bayes Trees |
| Sub Title (in English) | |
| Keyword(1) | Augmented Naive Bayes |
| Keyword(2) | Averaged One-Dependence Estimator (AODE) |
| Keyword(3) | Naive Bayes Tree (NBT) |
| Keyword(4) | model averaging |
| 1st Author's Name | Mori KUROKAWA |
| 1st Author's Affiliation | Web data computing group, KDDI R & D Laboratories Inc.() |
| 2nd Author's Name | Hiroyuki YOKOYAMA |
| 2nd Author's Affiliation | Web data computing group, KDDI R & D Laboratories Inc. |
| 3rd Author's Name | Akito SAKURAI |
| 3rd Author's Affiliation | Department of Science and Technology, Keio University |
| Date | 2008-11-07 |
| Paper # | NC2008-62 |
| Volume (vol) | vol.108 |
| Number (no) | 281 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |