基于自步学习的半监督中智聚类算法

Semi-Supervised Neutrosophic Clustering Algorithm Based on Self-paced Learning

现有的中智聚类算法,大多利用样本固有的标签信息改善聚类效果,并未关注不同样本对聚类结果的不同程度的影响.针对这一问题,本文提出一种基于自步学习的半监督中智聚类算法(SSNCSPL).首先,引入半监督学习理论,可以通过约束隶属度矩阵与标签矩阵的误差平方和最小,实现标签信息引导聚类结果的目的;其次,引入自步学习刻画不同样本对聚类结果的重要程度的差异性;最后,在各类数据实验表明,与现有算法比较,所提算法既有更好的聚类效果也能有效地减少噪声对模型性能的影响。

In the semi-supervised neutrosophic clustering algorithm,the guidance of labeled samples is usually used to improve the clustering effect of data,but the importance of different samples to the clustering result is not fully considered.In order to solve this problem,this paper proposes a semi-supervised NeocI clustering algorithm based on self-step learning(SSNCSPL).Firstly,semi-supervised learning was introduced into the model,which could minimize the sum of squares of the error between the membership matrix and the label matrix,and realize the purpose of the label information to guide the clustering result.Secondly,a self-learning mechanism was introduced into the model to describe the different importance degrees of different samples to the clustering results.Finally,the data experiments show that the proposed algorithm can achieve better clustering effect compared with the existing excellent algorithms.In addition,the experimental results also show that the proposed algorithm can effectively reduce the influence of noise on the clustering performance of the model.