• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>21st international conference on software engineering and data engineering 2012 >GAP: Genetic Algorithm Based Projected Clustering Method
【24h】

GAP: Genetic Algorithm Based Projected Clustering Method

机译：GAP：基于遗传算法的投影聚类方法

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Clustering is the division of dataset into groups called clusters, where objects in the same group are similar to each other and objects in different groups are dissimilar. Clustering techniques which find clusters in full dimensional space by considering all attributes of dataset tend to fail for high dimensional data due to various problems associated with it. Subspace and projected clustering methods find clusters that exist in subspaces of dataset have emerged as a possible solution to the challenges associated with clustering high dimensional data. Projected clustering methods output partition of points. In subspace clustering, points may be assigned to multiple subspace clusters. But in many application domains partition of points is required. In such cases projected clustering is preferable to subspace clustering. Genetic Algorithms have proven to be promising for solving complex optimization problems. In this paper, we propose a Genetic Algorithm Projected clustering method (GAP clustering method) to find subspace clusters that are present in the dataset. In GAP clustering method, Genetic Algorithm has been used to find optimal cluster centers of subspace clusters by optimizing a subspace cluster validation index. The proposed method has been applied to find subspace clusters present in synthetic and real datasets.
机译：聚类是将数据集划分为称为聚类的组，其中同一组中的对象彼此相似，而不同组中的对象则不同。通过考虑数据集的所有属性在全维空间中找到聚类的聚类技术由于与之相关的各种问题而往往无法用于高维数据。子空间和投影聚类方法发现存在于数据集子空间中的聚类已成为解决与聚类高维数据相关的挑战的可能解决方案。投影聚类方法输出点的分区。在子空间聚类中，可以将点分配给多个子空间聚类。但是在许多应用程序域中，需要对点进行分区。在这种情况下，投影聚类优于子空间聚类。遗传算法已被证明对于解决复杂的优化问题很有前途。在本文中，我们提出了一种遗传算法投影聚类方法（GAP聚类方法）来查找数据集中存在的子空间聚类。在GAP聚类方法中，遗传算法已用于通过优化子空间聚类验证指标来找到子空间聚类的最佳聚类中心。所提出的方法已被应用于寻找合成和真实数据集中存在的子空间簇。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号