By Topic

The Outliers Mining Algorithm Based on Constrained Concept Lattice

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$33 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

5 Author(s)
Jiang Yiyong ; TaiYuan Univ. of Sci. & Technol., Taiyuan ; Zhang Jifu ; Cai Jianghui ; Zhang Sulan
more authors

Traditional outlier mining methods regard outliers from overall point, and hardly find bias data in low dimensional subspace. Constrained concept lattice, with characteristics of higher constructing efficiency , practicability and pertinency, is a new concept lattice structure. Outlier mining algorithm RCLOM based on constrained concept lattice is presented for bias data in low dimensional subspace. The intension of constrained concept lattice nodes is regarded as subspace and sparsity coefficient is computed for every intension reductions of the nodes. If sparsity coefficient of k dimensional intension reduction is less than the sparsity coefficient threshold value, k-1 dimensional intensions are judged whether it is dense subspace. If all subspaces are dense, objects contained in the intension are seen as bias data or outliers in k dimensional subspace. In the end, the algorithm is feasible and effective for mining outliers in low dimensional subspace by the example.

Published in:

Data, Privacy, and E-Commerce, 2007. ISDPE 2007. The First International Symposium on

Date of Conference:

1-3 Nov. 2007