基于密度自适应邻域相似图的半监督谱聚类

谱聚类是基于谱图划分理论的一种聚类算法，传统的谱聚类算法属于无监督学习算法，只能利用单一数据来进行聚类。针对这种情况，提出一种基于密度自适应邻域相似图的半监督谱聚类（DAN-SSC）算法。DAN-SSC算法在传统谱聚类算法的基础上结合了半监督学习的思想，很好地解决了传统谱聚类算法无法充分利用所有数据，不得不对一些有标签数据进行舍弃的问题；将少量的成对约束先验信息扩散至整个空间，使其能更好地对聚类过程进行指导。实验结果表明，DAN-SSC算法具有可行性和有效性。

Semi-supervised spectral clustering based on density adaptive neighbor similarity graph

The spectral clustering is a clustering algorithm based on the theory of spectral partitioning. The traditional spectral clustering algorithm belongs to unsupervised learning algorithms and can only utilize a single type of data to cluster. Based on the situation, this paper proposed a semi-supervised spectral clustering algorithm based on density adaptive neighbor similarity graph(DAN-SSC). DAN-SSC algorithm combined the idea of semi-supervised learning on the basis of the traditional spectral clustering algorithm, it solved the problem that the traditional spectral clustering algorithm couldn''t fully utilize all the data and had to abandon some labeled data. It also spread a small amount of pairwise constrained prior information to the entire space and made it guide the process of clustering better. The results of experiments show that the proposed algorithm is feasible and effective.