| 一种改进的少数类样本过抽样算法 |
| |
| 引用本文： | 许丹丹,蔡立军,王勇. 一种改进的少数类样本过抽样算法[J]. 计算机工程, 2012, 38(4): 67-69 |
| |
| 作者姓名： | 许丹丹 蔡立军 王勇 |
| |
| 作者单位： | 1. 西北工业大学理学院,西安,710129
2. 西北工业大学计算机学院,西安,710072 |
| |
| 基金项目： | 国家自然科学基金资助项目（60873196） |
| |
| 摘 要： | 针对偏斜数据集的分类问题，提出一种改进的少数类样本过抽样算法(B-ISMOTE)。在边界少数类实例及其最近邻实例构成的 n维球体空间内进行随机插值，以此产生虚拟少数类实例，减小数据的不均衡程度。在实际数据集上进行实验，结果证明，与SMOTE算法和B-SMOTE算法相比，B-ISMOTE算法具有较优的分类性能。
|
| 关 键 词： | 偏斜数据集 分类 过抽样 虚拟实例 n维球体空间 |
| 收稿时间： | 2011-07-18 |
Improved Over-sampling Algorithm of Minority Class Sample |
| |
| XU Dan-dan , CAI Li-jun , WANG Yong. Improved Over-sampling Algorithm of Minority Class Sample[J]. Computer Engineering, 2012, 38(4): 67-69 |
| |
| Authors: | XU Dan-dan CAI Li-jun WANG Yong |
| |
| Affiliation: | 1. School of Science, Northwestern Polytechnical University, X i＇ an 710129, China; 2. School of Computer, Northwestern Polytechnical University, Xi＇an 710072, China) |
| |
| Abstract: | Aiming at the classification of the skewed dataset, this paper proposes an improved over-sampling algorithm of minority class sample, named B-ISMOTE. It improves the data unbalanced distribution of degree through randomized interpolation to produce virtual minority class instances in the sphere space, which constitute of the borderline minority class instances and its nearest neighbor. Experimental results on the real datasets show that compared with SMOTE algorithm and B-SMOTE algorithm, B-ISMOTE algorithm has better classification performance. |
| |
| Keywords: | skewed dataset classification over-sampling virtual instance n dimension sphere space |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《计算机工程》浏览原始摘要信息 |
|
点击此处可从《计算机工程》下载免费的PDF全文 |
|
