Constraint Based Subspace Clustering for High Dimensional Uncertain Data

Author(s):  
Xianchao Zhang ◽  
Lu Gao ◽  
Hong Yu
2020 ◽  
Vol 130 ◽  
pp. 253-268
Author(s):  
Aluizio F.R. Araújo ◽  
Victor O. Antonino ◽  
Karina L. Ponce-Guevara

Author(s):  
Parul Agarwal ◽  
Shikha Mehta

Subspace clustering approaches cluster high dimensional data in different subspaces. It means grouping the data with different relevant subsets of dimensions. This technique has become very effective as a distance measure becomes ineffective in a high dimensional space. This chapter presents a novel evolutionary approach to a bottom up subspace clustering SUBSPACE_DE which is scalable to high dimensional data. SUBSPACE_DE uses a self-adaptive DBSCAN algorithm to perform clustering in data instances of each attribute and maximal subspaces. Self-adaptive DBSCAN clustering algorithms accept input from differential evolution algorithms. The proposed SUBSPACE_DE algorithm is tested on 14 datasets, both real and synthetic. It is compared with 11 existing subspace clustering algorithms. Evaluation metrics such as F1_Measure and accuracy are used. Performance analysis of the proposed algorithms is considerably better on a success rate ratio ranking in both accuracy and F1_Measure. SUBSPACE_DE also has potential scalability on high dimensional datasets.


2017 ◽  
Vol 2017 ◽  
pp. 1-9 ◽  
Author(s):  
Binbin Zhang ◽  
Weiwei Wang ◽  
Xiangchu Feng

Subspace clustering aims to group a set of data from a union of subspaces into the subspace from which it was drawn. It has become a popular method for recovering the low-dimensional structure underlying high-dimensional dataset. The state-of-the-art methods construct an affinity matrix based on the self-representation of the dataset and then use a spectral clustering method to obtain the final clustering result. These methods show that sparsity and grouping effect of the affinity matrix are important in recovering the low-dimensional structure. In this work, we propose a weighted sparse penalty and a weighted grouping effect penalty in modeling the self-representation of data points. The experimental results on Extended Yale B, USPS, and Berkeley 500 image segmentation datasets show that the proposed model is more effective than state-of-the-art methods in revealing the subspace structure underlying high-dimensional dataset.


2012 ◽  
Vol 45 (1) ◽  
pp. 434-446 ◽  
Author(s):  
Xiaojun Chen ◽  
Yunming Ye ◽  
Xiaofei Xu ◽  
Joshua Zhexue Huang

Sign in / Sign up

Export Citation Format

Share Document