In actual databases, there are a lot of hierarchy coding data, existing clustering algorithms don't consider the special treatment of these data structure, so lead nonideal performance and clustering result. This paper proposes a new clustering algorithm to deal with the hierarchy coding data structure (HCDS) that exists in many applications. The main contributions include: (1) proposes a new concept for HCDS and corresponding definitions. (2) Proposes and implements a new clustering algorithm---CHCC (Coding Hierarchy Computing Based Clustering Algorithm) based on HCDS. (3) Proposes a fast algorithm for hierarchy coding structure processing. (4) Applies the algorithm into the clustering analysis of transient population for public security, and through extensive experiments, proves the validity and efficiency of the algorithm.
[1]
Paul E. Green,et al.
K-modes Clustering
,
2001,
J. Classif..
[2]
ChengXiang Zhai,et al.
A mixture model for contextual text mining
,
2006,
KDD '06.
[3]
Philip S. Yu,et al.
Outlier detection for high dimensional data
,
2001,
SIGMOD '01.
[4]
Philip S. Yu,et al.
A Framework for Projected Clustering of High Dimensional Data Streams
,
2004,
VLDB.
[5]
Raymond J. Mooney,et al.
A probabilistic framework for semi-supervised clustering
,
2004,
KDD.
[6]
Philip S. Yu,et al.
A Framework for Clustering Evolving Data Streams
,
2003,
VLDB.
[7]
Jiawei Han,et al.
Data Mining: Concepts and Techniques
,
2000
.
[8]
Philip S. Yu,et al.
Clustering by pattern similarity in large data sets
,
2002,
SIGMOD '02.