We present a new subspace clustering method called SuMC (Subspace Memory Clustering), which allows to efficiently divide a dataset D c RN into k 2 N pairwise disjoint clusters of possibly different dimensions. Since our approach is based on the memory compression, we do not need to explicitly specify dimensions of groups: in fact we only need to specify the mean number of scalars which is used to describe a data-point. In the case of one cluster our method reduces to a classical Karhunen-Loeve (PCA) transform. We test our method on some typical data from UCI repository and on data coming from real-life experiments.
pl
dc.subject.en
subspace clustering
pl
dc.subject.en
PCA
pl
dc.subject.en
projection clustering
pl
dc.description.volume
24
pl
dc.identifier.doi
10.4467/20838476SI.15.013.3035
pl
dc.identifier.eissn
2083-8476
pl
dc.title.journal
Schedae Informaticae
pl
dc.language.container
eng
pl
dc.affiliation
Wydział Matematyki i Informatyki : Instytut Informatyki i Matematyki Komputerowej
pl
dc.affiliation
Wydział Matematyki i Informatyki : Instytut Matematyki
pl
dc.subtype
Article
pl
dc.rights.original
OTHER; otwarte czasopismo; ostateczna wersja wydawcy; w momencie opublikowania; 0