Faces and text data.

Similarity matrix and pattern labels (download) from a data set consisting of 2700 black and white face images from 135 different individuals. For each individual there are 20 snapshots. The original data base as well as relative information can be downloaded in [1]. The original images have been transformed in black and white and downscaled in half of the original size. A Matlab script (download) with execution instructions.

Similarity matrix and pattern labels (download) from a data setconsisting of 3893 abstracts from documents: MEDLINE (1033 documents), CISI(1460 documents), and CRANFIELD (1400 documents) sets. Following instructions in [2] we build a data set consisting of a matrix consisting 3893 rows each one representing a document and 3303 columns. A Matlab script (download) with execution instructions.

[1]: Link 1

[2]: I.S. Dhillon, D. S. Modha, “Concept Decompositions for Large Sparse Text Data Using Clustering”, Machine Learning, Vol. 42, p. 143-175, 2001