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：
为克服聚类算法对初始聚类中心选取敏感这一缺点,结合文本样本集中各个词所处位置不同而具有不同重要程度（权值）可客观反映文本数据本来特征的特点,提出一个考虑样本点分布密度优选初始聚类中心的最大熵核FCM算法（WKMEFCM）。实验结果表明,该算法与C均值方法、模糊C均值方法 FCM、最大熵核FCM相比,其聚类结果更加稳定、准确,聚类效果更好。
英文摘要：
To overcome the weakness of the sensitivity of clustering algorithms in selecting initial clustering centers,combining that words in text sample set locating in different positions have different importance（weights）and may reflect native characteristics of text data objectively,a maximum entropy kernel fuzzy text clustering algorithm（WKMEFCM）was proposed by optimally choosing initial centers based on sample distribution density.Experimental results show that the algorithm is more stable,accurate and effective compared with the C-means clustering algorithm,fuzzy C-means clustering algorithm FCM and maximal entropy and kernel FCM algorithm.
同期刊论文项目
同项目期刊论文
Interval-valued analysis for discriminative gene selection and tissue sample classification using mi
A fuzzy supervised learning method with dynamical parameter estimation for nonlinear discriminant an
Estimating Local False Discovery Rates to Identify the Differentially Expressed Genes in Microarrays
期刊信息
位置：