青岛大学学报（自然科学版）2017，Vol.30Issue(3)：64-68,5.DOI:10.3969/j.issn.1006-1037.2017.08.14
稀疏低秩子空间聚类算法
Sparse Low Rank Subspace Clustering Algorithm
摘要
Abstract
The main goal of subspace clustering algorithm is to find its low dimension representation from high dimensional data.In the low-rank representation subspace algorithm,original data is used as a dictionary,however the noise from data is inevitable.To solve this problem,a sparse low rank subspace clustering algorithm with using sparse representation optimization dictionary is proposed.The improved algorithm can handle the appearance that the excessive deviation of final result perfectly.It is showed by the experimental results that the algorithm has a significant improvement compared with the sparse subspace algorithm and the low rank representation algorithm.关键词
稀疏低秩/字典学习/高维数据Key words
sparse low rank/dictionary learning/high dimensional data分类
信息技术与安全科学引用本文复制引用
解昊,赵志刚,吕慧显,王福驰,刘馨月..稀疏低秩子空间聚类算法[J].青岛大学学报（自然科学版）,2017,30(3):64-68,5.基金项目
山东省科学技术发展计划(批准号:2012YD01058)资助. （批准号:2012YD01058）
