| Presentation | 2015-12-19 A Learning Method of Distributed Support Vector Machine Based on Transfer Control of Data Yukawa Kiichiro, Mikawa Kenta, Goto Masayuki, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Due to the development of information technology, the analysis of big data stored in various databases has become more important. From this kind of circumstance, the importance of Distributed Data Mining (DDM), which is the technique to implement data mining while each database doesn't transmit raw data to each other, has been advocated. As one of the methods, Forrero et al. proposed the method of learning optimal support vector machine (SVM) using the Alternating Direction Method of Multipliers (ADMM) in DDM. This method can learn the optimal hyperplane with low iterations and communication costs for any network structure without sharing their data. However, when the statistical characteristics of data stored in each node are quite different, this method requires large iterations until convergence. In this study, we propose a new learning method, which reduces the number of iterations considering network structure, provided that the all nodes are connected with each other. To verify the effectiveness of proposed method, a simulation experiment is conducted. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Distributed Data Mining / Support Vector Machine / Graph Structure |
| Paper # | AI2015-51 |
| Date of Issue | 2015-12-11 (AI) |
| Conference Information | |
| Committee | AI |
|---|---|
| Conference Date | 2015/12/18(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | Toshiharu Sugawara(Waseda Univ.) |
| Vice Chair | Tsunenori Mine(Kyushu Univ.) / Daisuke Katagami(Tokyo Polytechnic Univ.) |
| Secretary | Tsunenori Mine(Kyoto Univ.) / Daisuke Katagami(Shizuoka Univ.) |
| Assistant | Yuichi Sei(Univ. of Electro-Comm.) |
| Paper Information | |
| Registration To | Technical Committee on Artificial Intelligence and Knowledge-Based Processing |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | A Learning Method of Distributed Support Vector Machine Based on Transfer Control of Data |
| Sub Title (in English) | |
| Keyword(1) | Distributed Data Mining |
| Keyword(2) | Support Vector Machine |
| Keyword(3) | Graph Structure |
| 1st Author's Name | Yukawa Kiichiro |
| 1st Author's Affiliation | Waseda University(Waseda Univ.) |
| 2nd Author's Name | Mikawa Kenta |
| 2nd Author's Affiliation | Waseda University(Waseda Univ.) |
| 3rd Author's Name | Goto Masayuki |
| 3rd Author's Affiliation | Waseda University(Waseda Univ.) |
| Date | 2015-12-19 |
| Paper # | AI2015-51 |
| Volume (vol) | vol.115 |
| Number (no) | AI-381 |
| Page | pp.pp.149-154(AI), |
| #Pages | 6 |
| Date of Issue | 2015-12-11 (AI) |