Abstract | ||
---|---|---|
In this paper we show how to reduce the computational cost of Clustering
by
Compression, proposed by Cilibrasi & Vitànyi, from O(n4) to O(n2). To that end, we adopte the Weighted Paired Group Method using Averages (WPGMA) method to the same similarity measure, based
on compression, used in Clustering
by
Compression. Consequently, our proposed approach has easily classified thousands of data, where Cilibrasi & Vitànyi proposed algorithm
shows its limits just for a hundred objects. We give also results of experiments.
|
Year | DOI | Venue |
---|---|---|
2010 | 10.1007/978-3-642-15387-7_49 | Knowledge-Based Intelligent Information & Engineering Systems |
Keywords | Field | DocType |
classified thousand,kolmogorov information,weighted paired group method,similarity measure,computational cost,hundred object,nyi proposed algorithm,information theory,classification | Information theory,Fuzzy clustering,Similarity measure,Correlation clustering,Kolmogorov complexity,Pattern recognition,Computer science,Normalized compression distance,Artificial intelligence,Cluster analysis | Conference |
Volume | ISSN | ISBN |
6276 | 0302-9743 | 3-642-15386-0 |
Citations | PageRank | References |
1 | 0.36 | 8 |
Authors | ||
5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Said Fouchal | 1 | 19 | 2.21 |
Murat Ahat | 2 | 15 | 3.52 |
Ivan Lavallée | 3 | 24 | 6.76 |
Marc Bui | 4 | 23 | 9.28 |
Sofiane Ben Amor | 5 | 1 | 0.36 |