Title :
Approach of outlier mining based on high dimension sparse clustering
Author_Institution :
Dept. of Math., Huainan Normal Univ., Huainan, China
Abstract :
To solve the problem that CABOSFV is only effective to high dimension sparse clustering problems of two-value attributes, this paper presents an improved high dimension sparse clustering approach by defining “absolute attribute distance”. The experimental results show that this method is efficient in outlier mining, and easy to calculate and program, so it is a more practical method.
Keywords :
data mining; pattern clustering; statistical analysis; CABOSFV algorithm; absolute attribute distance; clustering algorithm based on sparse feature vector; high dimension sparse clustering; outlier mining; Clustering algorithms; Communications technology; Credit cards; Data mining; Indexes; Presses; CABOSFV; High dimension; Outlier mining; Sparse clustering;
Conference_Titel :
Artificial Intelligence, Management Science and Electronic Commerce (AIMSEC), 2011 2nd International Conference on
Conference_Location :
Deng Leng
Print_ISBN :
978-1-4577-0535-9
DOI :
10.1109/AIMSEC.2011.6010420
