Paper Search Console

Home Search Page About Contact

Journal Title

Title of Journal: Knowl Inf Syst

Search In Journal Title:

Abbravation: Knowledge and Information Systems

Search In Journal Abbravation:

Publisher

Springer London

Search In Publisher:

DOI

10.1007/bf01692011

Search In DOI:

ISSN

0219-3116

Search In ISSN:
Search In Title Of Papers:

Scaling cut criterionbased discriminant analysis

Authors: Xiangrong Zhang Yudi He Licheng Jiao Ruochen Liu Jie Feng Sisi Zhou
Publish Date: 2014/04/06
Volume: 43, Issue: 3, Pages: 633-655
PDF Link

Abstract

Dimension reduction has always been a major problem in many applications of machine learning and pattern recognition In this paper the scaling cut criterionbased supervised dimension reduction methods for data analysis are proposed The scaling cut criterion can eliminate the limit of the hypothesis that data distribution of each class is homoscedastic Gaussian To obtain a more reasonable mapping matrix and reduce the computational complexity local scaling cut criterionbased dimension reduction is raised which utilized the localization strategy of the input data The localized knearest neighbor graph is introduced which relaxes the withinclass variance and enlarges the betweenclass margin Moreover by kernelizing the scaling cut criterion and local scaling cut criterion both methods are extended to efficiently model the nonlinear variability of the data Furthermore the optimal dimension scaling cut criterion is proposed which can automatically select the optimal dimension for the dimension reduction methods The approaches have been tested on several datasets and the results have shown a better and efficient performance compared with other linear and nonlinear dimension reduction techniquesThis work was supported by the National Basic Research Program of China 973 Program Grant 2013CB329402 the National Natural Science Foundation of China Nos 61272282 61203303 and 61272279 the Program for New Century Excellent Talents in University NCET130948 and Fundamental Research Funds for the Central Universities Grant K50511020011

Keywords:

References

.
Search In Abstract Of Papers:
Other Papers In This Journal:
  1. Modified algorithms for synthesizing high-frequency rules from different data sources
  2. Domain-agnostic discovery of similarities and concepts at scale
  3. A transversal hypergraph approach for the frequent itemset hiding problem
  4. Bootstrapping trust of Web services based on trust patterns and Hidden Markov Models
  5. A hybrid scheme for energy-efficient object tracking in sensor networks
  6. An ontology-based procedure for generating object model from text description
  7. Effective and efficient classification on a search-engine model
  8. A multimodal virtual reality interface for 3D interaction with VTK
  9. Ontology alignment design patterns
  10. A framework for anomaly detection in maritime trajectory behavior
  11. Knowledge management in image-based analysis of blood vessel structures
  12. Remodeling the network for microgroup detection on microblog
  13. Fuzzy clustering-based discretization for gene expression classification
  14. On the difficulty of automatically detecting irony: beyond a simple case of negation
  15. A spatial relation-based framework to perform visual information extraction
  16. Action rule discovery from incomplete data
  17. Architecture for knowledge discovery and knowledge management
  18. Managing Multiuser Database Buffers Using Data Mining Techniques
  19. On link privacy in randomizing social networks
  20. Redesigning business processes: a methodology based on simulation and process mining techniques
Search Result: