欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
聚类算法对初始值的依赖性较大,通常容易陷入局部最优,很难得到稳定的聚类结果。为改善该问题,本文提出了一种改进的加权模糊核聚类算法,将迭代自组织的数据分析算法（ISODATA）的思想引入到加权模糊核聚类算法（WFKCA）中,利用聚类中心分裂/合并的中间结果来调整初始中心,降低了WFKCA算法收敛于局部最优的可能。改进算法采用特征空间中的计算度量,将单值标准差阈值扩展为标准差阈值向量,并增加了对聚类中心的调整幅度。实验结果显示,该算法在不同结构和维数的数据集上都取得了更稳定的聚类精度。
英文摘要：
The dependence of clustering algorithm on initial values is generally liable to stick to a local optimum,therefore makes it difficult to obtain a stable clustering result.To overcome this shortcoming,an improved algorithm for weighted fuzzy kernel clustering analysis is proposed.The idea of iterative self-organizing data analysis technique algorithm（ISODATA） is introduced into the weighted fuzzy kernel clustering algorithm（WFKCA）,and initial center vectors are adjusted by the intermediate results from splitting and/or merging of clustering centers to reduce the possibility of local optimum.The algorithm uses a matchable measurement from the feature space,extends the single-value standard deviation threshold to standard deviation threshold vector,and increases the adjustment range of clustering centers.Simulation results show that the algorithm can achieve more stable clustering accuracy on the benchmark data sets.
同期刊论文项目
同项目期刊论文
An optimal symmetrical null space criterion of Fisher discriminant for feature extraction and recogn
Dominance-based Rough Set Approach and Knowledge Reductions in Incomplete Ordered Information System
An extreme case of the generalized optimal discriminant transformation and its application to face r
位置：