CACS: A Novel Classification Algorithm Based on Concept Similarity

Authors:
Jing Peng;Dong-Qing Yang;Chang-Jie Tang;Jing Zhang;Jian-Jun Hu
Affiliations:
School of Electronics Engineering and Computer Science/ Peking University/ Beijing 100871, China;School of Electronics Engineering and Computer Science/ Peking University/ Beijing 100871, China;School of computer Science and Engineering, Sichuan University, Chengdu 610065, China;Chengdu Jiuheyuan Industry Company, Chengdu 610015, China;School of computer Science and Engineering, Sichuan University, Chengdu 610065, China
Venue:
ADMA '07 Proceedings of the 3rd international conference on Advanced Data Mining and Applications
Year:
2007

Citing 8
Cited 0

Rank aggregation methods for the Web

Proceedings of the 10th international conference on World Wide Web
Database-friendly random projections

PODS '01 Proceedings of the twentieth ACM SIGMOD-SIGACT-SIGART symposium on Principles of database systems
Hierarchical subspace sampling: a unified framework for high dimensional data reduction, selectivity estimation and nearest neighbor search

Proceedings of the 2002 ACM SIGMOD international conference on Management of data
Efficient Search for Approximate Nearest Neighbor in High Dimensional Spaces

SIAM Journal on Computing
Contrast Plots and P-Sphere Trees: Space vs. Time in Nearest Neighbour Searches

VLDB '00 Proceedings of the 26th International Conference on Very Large Data Bases
Database-friendly random projections: Johnson-Lindenstrauss with binary coins

Journal of Computer and System Sciences - Special issu on PODS 2001
Data Mining: Concepts and Techniques

Data Mining: Concepts and Techniques
Evolving accurate and compact classification rules with gene expression programming

IEEE Transactions on Evolutionary Computation

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper proposes a novel algorithm of classification based on the similarities among data attributes. This method assumes data attributes of dataset as basic vectors of m dimensions, and each tuple of dataset as a sum vector of all the attribute-vectors. Based on transcendental concept similarity information among attributes, this paper suggests a novel distance algorithm to compute the similarity distance of each pairs of attribute-vectors. In the method, the computing of correlation is turned to attribute-vectors and formulas of their projections on each other, and the correlation among any two tuples of dataset can be worked out by computing these vectors and formulas. Based on the correlation computing method, this paper proposes a novel classification algorithm. Extensive experiments prove the efficiency of the algorithm.