| 基于遗传算法的多分类器融合模型在信用评估中的应用 |
| |
| 引用本文： | 叶强,张洁.基于遗传算法的多分类器融合模型在信用评估中的应用[J].哈尔滨工业大学学报,2006,38(9):1504-1505,1536. |
| |
| 作者姓名： | 叶强 张洁 |
| |
| 作者单位： | 哈尔滨工业大学,管理学院,哈尔滨,150001 |
| |
| 基金项目： | 国家自然科学基金
,
黑龙江省自然科学基金 |
| |
| 摘 要： | 为探讨基于遗传算法的多分类器融合模型,并基于多分类器融合技术,建立新的客户信用分类模型,该模型通过使用分类融合器,将多个单分类器得到的客户信用评估结果进行合并,从而综合不同分类器的局部优势,提高分类性能.采用线性分类融合器,并通过遗传算法对分类融合器进行优化.实验表明,该方法在客户信用评估中的效果明显优于传统的运用单个分类器的方法.
|
| 关 键 词： | 多分类器融合模型 遗传算法 信用评估 |
| 文章编号： | 0367-6234（2006）09-1504-02 |
| 收稿时间： | 2005-12-15 |
| 修稿时间： | 2005-12-15 |
Application of multiple classifiers syncretizing model in credit evaluation |
| |
| YE Qiang,ZHANG Jie.Application of multiple classifiers syncretizing model in credit evaluation[J].Journal of Harbin Institute of Technology,2006,38(9):1504-1505,1536. |
| |
| Authors: | YE Qiang ZHANG Jie |
| |
| Affiliation: | School of Management, Harbin Institute of Technology, Harbin 150001, China |
| |
| Abstract: | By combining multiple classifiers,a new model for customer credit evaluation was established in this study.The evaluation ability was improved by combining multiple results of different classifiers into a single new result.Genetic algorithm was adopted to optimize the linear fusion model.Experimentation results showed that the evaluation ability of this new method was better than the traditional evaluation methods. |
| |
| Keywords: | model of multiple classifiers syncretizing genetic algorithm credit evaluation |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
