| 基于支持向量机的空间数据拟合 |
| |
| 引用本文： | 冯杨民,梁晶.基于支持向量机的空间数据拟合[J].海洋测绘,2009,29(3):31-33. |
| |
| 作者姓名： | 冯杨民 梁晶 |
| |
| 作者单位： | 长安大学,地质工程与测绘学院,陕西,西安,710054 |
| |
| 摘 要： | 空间数据拟合是以搜集空间信息、建立合理的数据解算模型，用以表达空间数据之间的关系。支持向量机应用于空间数据拟合时，建立在结构风险最小的原则上，使得其模型有较强的泛化性能。借鉴泰勒公式的思想，在拟合空间起伏较大数据时，提出了对支持向量残差再拟合，以提高数据拟合的精度。最后，通过实例验证了残差再拟合算法的有效性。
|
| 关 键 词： | 空间数据库 支持向量机 优化 残差再拟合 |
Space Data Fitting Based on the Support Vector Machine-SVM |
| |
| FENG Yang-min,LIANG Jing.Space Data Fitting Based on the Support Vector Machine-SVM[J].Hydrographic Surveying and Charting,2009,29(3):31-33. |
| |
| Authors: | FENG Yang-min LIANG Jing |
| |
| Institution: | ( School of Geology Engineering and Geomatics, Chang＇ an University, Xi＇ an, Shaanxi,710054) |
| |
| Abstract: | Space data fitting is used to illustrate the relationship of them by the collection of spatial information and the construction of rational data processing model. The usage of support vector machine in space data fitting is based on the minimal risk. Taking the advantage of Taylor formula, the second fitting of the SVM residual error is proposed in order to improve the fitting accuracy when fitting the undulating space data. The validity of the thought of the paper is demonstrated by an exampe. |
| |
| Keywords: | spatial database support vector machine（SVM） optimization second fitting of the residual error |
| 本文献已被 维普 万方数据 等数据库收录！ |
| 点击此处可从《海洋测绘》浏览原始摘要信息 |
| 点击此处可从《海洋测绘》下载免费的PDF全文 |
