快速搜索与发现密度峰值聚类(clustering by fast search and find of density peaks,DPC)算法对聚类中心点进行了全新的定义,能够得到更优的聚类结果。但该算法需要手动选取聚类中心,容易出现多选、漏选聚类中心的问题。提出一种自动...快速搜索与发现密度峰值聚类(clustering by fast search and find of density peaks,DPC)算法对聚类中心点进行了全新的定义,能够得到更优的聚类结果。但该算法需要手动选取聚类中心,容易出现多选、漏选聚类中心的问题。提出一种自动选取聚类中心的密度峰值聚类算法。将参数积γ引入新算法以扩大聚类中心的选取范围,利用KL散度的差异性度量准则对聚类中心点和非聚类中心点进行清晰划分,以Dkl排序图中的拐点作为分界点实现了对聚类中心的自动选取。在人工以及UCI数据集上的实验表明,新算法能够在自动选取聚类中心的同时,获得更好的聚类效果。展开更多
移动时间层次聚类(Travel-Time based Hierarchical Clustering,TTHC)是一种新的势能聚类算法,尽管具有较好的聚类效果,但是该算法需要人工设定聚类数目,而且在分配样本的时候仅根据相似度,忽略了距离和势能的影响.针对以上问题,提出一...移动时间层次聚类(Travel-Time based Hierarchical Clustering,TTHC)是一种新的势能聚类算法,尽管具有较好的聚类效果,但是该算法需要人工设定聚类数目,而且在分配样本的时候仅根据相似度,忽略了距离和势能的影响.针对以上问题,提出一种自动确定聚类中心的移动时间势能聚类算法.首先计算每个数据点的势能和相似度,然后根据相似度确定数据点的父节点,得到数据点与父节点的距离;然后,根据数据点与父节点的相似度、距离和数据点的势能得到综合考量值,根据综合考量值自动确定聚类中心;最后,将剩余数据点分配到比其势能小且与其相似度最大的数据点所属类簇,得到聚类结果.将新算法与TTHC算法进行比较,在人工数据集和真实数据集上的实验结果表明,新算法不仅能够自动确定聚类数目,而且采用了更优的分配机制,可以产生更好的聚类结果.展开更多
Aimed at the problem that the traditional suppressed fuzzy C-means clustering algorithms ignore the real needs of different objects, applying the same suppressed parameter for modifying membership degrees of all the o...Aimed at the problem that the traditional suppressed fuzzy C-means clustering algorithms ignore the real needs of different objects, applying the same suppressed parameter for modifying membership degrees of all the objects, a novel partition region-based suppressed fuzzy C-means clustering algorithm with better capacity of adaptability and robustness is proposed in this paper. The model based on the real needs of different objects is built, making it clear to decide whether to proceed with further determination; in addition, the external user-defined suppressed parameter is automatically selected according to the intrinsic structural characteristic of each dataset, making the proposed method become robust to the fluctuations in the incoming dataset and initial conditions. Experimental results show that the proposed method is more robust than its counterparts and overcomes the weakness of the original suppressed clustering algorithm in most cases.展开更多
This paper introduces niching particle swarm optimiza- tion （nichePSO） into clustering analysis and puts forward a cluster- ing algorithm which uses nichePSO to optimize density functions. Firstly, this paper improv...This paper introduces niching particle swarm optimiza- tion （nichePSO） into clustering analysis and puts forward a cluster- ing algorithm which uses nichePSO to optimize density functions. Firstly, this paper improves main swarm training models and in- creases their ability of space searching. Secondly, the radius of sub-swarms is defined adaptively according to the actual clus- tering problem, which can be useful for the niches＇ forming and searching. At last, a novel method that distributes samples to the corresponding cluster is proposed. Numerical results illustrate that this algorithm based on the density function and nichePSO could cluster unbalanced density datasets into the correct clusters auto- matically and accurately.展开更多
文摘快速搜索与发现密度峰值聚类(clustering by fast search and find of density peaks,DPC)算法对聚类中心点进行了全新的定义,能够得到更优的聚类结果。但该算法需要手动选取聚类中心,容易出现多选、漏选聚类中心的问题。提出一种自动选取聚类中心的密度峰值聚类算法。将参数积γ引入新算法以扩大聚类中心的选取范围,利用KL散度的差异性度量准则对聚类中心点和非聚类中心点进行清晰划分,以Dkl排序图中的拐点作为分界点实现了对聚类中心的自动选取。在人工以及UCI数据集上的实验表明,新算法能够在自动选取聚类中心的同时,获得更好的聚类效果。
文摘移动时间层次聚类(Travel-Time based Hierarchical Clustering,TTHC)是一种新的势能聚类算法,尽管具有较好的聚类效果,但是该算法需要人工设定聚类数目,而且在分配样本的时候仅根据相似度,忽略了距离和势能的影响.针对以上问题,提出一种自动确定聚类中心的移动时间势能聚类算法.首先计算每个数据点的势能和相似度,然后根据相似度确定数据点的父节点,得到数据点与父节点的距离;然后,根据数据点与父节点的相似度、距离和数据点的势能得到综合考量值,根据综合考量值自动确定聚类中心;最后,将剩余数据点分配到比其势能小且与其相似度最大的数据点所属类簇,得到聚类结果.将新算法与TTHC算法进行比较,在人工数据集和真实数据集上的实验结果表明,新算法不仅能够自动确定聚类数目,而且采用了更优的分配机制,可以产生更好的聚类结果.
基金supported by the National Natural Science Foundation of China(61401363)the Science and Technology on Avionics Integration Laboratory and Aeronautical Science Foundation(20155153034)+1 种基金the Fundamental Research Funds for the Central Universities(3102016AXXX0053102015BJJGZ009)
文摘Aimed at the problem that the traditional suppressed fuzzy C-means clustering algorithms ignore the real needs of different objects, applying the same suppressed parameter for modifying membership degrees of all the objects, a novel partition region-based suppressed fuzzy C-means clustering algorithm with better capacity of adaptability and robustness is proposed in this paper. The model based on the real needs of different objects is built, making it clear to decide whether to proceed with further determination; in addition, the external user-defined suppressed parameter is automatically selected according to the intrinsic structural characteristic of each dataset, making the proposed method become robust to the fluctuations in the incoming dataset and initial conditions. Experimental results show that the proposed method is more robust than its counterparts and overcomes the weakness of the original suppressed clustering algorithm in most cases.
基金supported by the National Natural Science Foundation of China (708710157103100271171030)
文摘This paper introduces niching particle swarm optimiza- tion （nichePSO） into clustering analysis and puts forward a cluster- ing algorithm which uses nichePSO to optimize density functions. Firstly, this paper improves main swarm training models and in- creases their ability of space searching. Secondly, the radius of sub-swarms is defined adaptively according to the actual clus- tering problem, which can be useful for the niches＇ forming and searching. At last, a novel method that distributes samples to the corresponding cluster is proposed. Numerical results illustrate that this algorithm based on the density function and nichePSO could cluster unbalanced density datasets into the correct clusters auto- matically and accurately.
