| 改进型特征权重自调节K-均值聚类算法 |
| |
| 引用本文： | 支晓斌,许朝晖.改进型特征权重自调节K-均值聚类算法[J].西安邮电学院学报,2014,19(6):26-31. |
| |
| 作者姓名： | 支晓斌 许朝晖 |
| |
| 作者单位： | 1. 西安邮电大学理学院,陕西西安,710121
2. 西安邮电大学通信与信息工程学院,陕西西安,710121 |
| |
| 基金项目： | 陕西省自然科学基金资助项目 |
| |
| 摘 要： | 针对特征权重自调节K-均值聚类(FWSA-KM)算法对噪声敏感的问题,提出一种改进型特征权重自调节K-均值聚类(IFWSA-KM)算法。用一种非欧氏距离代替FWSA-KM算法中的欧氏距离,以增加聚类算法的抗噪声性能。通过用人工数据和真实数据的对比性实验,可验证IFWSA-KM算法的有效性。
|
| 关 键 词： | 聚类算法 特征权重 鲁棒性 非欧氏距离 |
K-means clustering algorithm with an improved feature weight self-adjustment mechanism |
| |
| ZHI Xiaobin,XU Zhaohui.K-means clustering algorithm with an improved feature weight self-adjustment mechanism[J].Journal of Xi'an Institute of Posts and Telecommunications,2014,19(6):26-31. |
| |
| Authors: | ZHI Xiaobin XU Zhaohui |
| |
| Affiliation: | ZHI Xiaobin , XU Zhaohui ( 1. School of Science, Xi＇an University of Posts and Telecommunications, Xi＇an 710121, China; 2. School of Conmaunicafion and Information Engineering, Xi＇an University of Posts and Tel , Xi＇an 710121, China) |
| |
| Abstract: | K-means with a feature weight self-adjustment mechanism（FWSA-KM）clustering algorithm is sensitive to noise.Therefore K-means with an improved feature weight selfadjustment mechanism（IFWSA-KM）clustering algorithm is proposed in this paper.IFWSA-KM clustering algorithm can have some anti-noise performance by using a non-Euclidean distance.The effectiveness of IFWSA-KM algorithm is demonstrated by comparative experiments on synthetic and real data. |
| |
| Keywords: | clustering algorithm feature weighting robust non-Euclidean distance |
| 本文献已被 维普 等数据库收录！ |
|
