| Presentation | 2019-07-25 Bayes Optimal Classification on Decision Tree Model and Its Approximative Algorithm Using Ensemble Learning Nao Dobashi, Shota Saito, Toshiyasu Matsushima, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In this paper we consider classification problem about discrete category $y$ regarding discrete variables $bm{x}$. Decision tree model is one of the model expressing $P(y|bm{x})$. Previously, regarding the classification problem whose $P(y|bm{x})$ is unknown, algorithms which estimate one or more decision tree models from given data and classify using them have been studied. However, these algorithms are not necessarily theoretically optimal. On the other hand, Suko et al. proposed Bayes optimal classification by considering every decision tree model (model class). However, we have to calculate the sum of every decision tree model which is included in model class in order to do this. Therefore, they proposed an algorithm achiving Bayes optimal classification under restriction. In addition, Arai et al. proposed an approximative algorithm under moderate restriction. In our study, we propose an approximative algorithm of the Bayes optimal classification. In this algorithm, we construct multiple model classes based on the idea of ensemble learning and average the outputs of these model classes with their posterior distributions. Also, experiments using synthetic data are performed to make sure its effectiveness and performance. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Decision Tree Model / Ensemble Learning / Bayes Optimal Classification |
| Paper # | IT2019-17 |
| Date of Issue | 2019-07-18 (IT) |
| Conference Information | |
| Committee | IT |
|---|---|
| Conference Date | 2019/7/25(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | NATULUCK-Iidabashi-Higashiguchi Ekimaeten |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | freshman session, general |
| Chair | Jun Muramatsu(NTT) |
| Vice Chair | Tadashi Wadayama(Nagoya Inst. of Tech.) |
| Secretary | Tadashi Wadayama(Saga Univ.) |
| Assistant | Hideki Yagi(UEC) |
| Paper Information | |
| Registration To | Technical Committee on Information Theory |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Bayes Optimal Classification on Decision Tree Model and Its Approximative Algorithm Using Ensemble Learning |
| Sub Title (in English) | |
| Keyword(1) | Decision Tree Model |
| Keyword(2) | Ensemble Learning |
| Keyword(3) | Bayes Optimal Classification |
| 1st Author's Name | Nao Dobashi |
| 1st Author's Affiliation | Waseda University(Waseda Univ.) |
| 2nd Author's Name | Shota Saito |
| 2nd Author's Affiliation | Waseda University(Waseda Univ.) |
| 3rd Author's Name | Toshiyasu Matsushima |
| 3rd Author's Affiliation | Waseda University(Waseda Univ.) |
| Date | 2019-07-25 |
| Paper # | IT2019-17 |
| Volume (vol) | vol.119 |
| Number (no) | IT-149 |
| Page | pp.pp.11-16(IT), |
| #Pages | 6 |
| Date of Issue | 2019-07-18 (IT) |