Point-based Subspace Clustering Algorithms (PBSCA)
According to the philosophy of grid-based algorithms, clusters are defined as unions of dense units. In addition, a data point may “contribute” to more than one cluster in different subspaces through its projections. Moreover, the identification of clusters takes place after the establishment of the appropriate subspaces. In the present framework, a different philosophy is adopted. The clusters are defined in terms of data points and each data point contributes to a single cluster. Furthermore, the clusters as well as the subspaces in which they live are simultaneously determined in an iterative fashion. Two typical representatives of this category, the PROCLUS and ORCLUS, are described next. ...
Get Pattern Recognition, 4th Edition now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
