Information Technology Reference
In-Depth Information
(a) dataset13;
nc
=4;
σ
=0
.
3
(b) dataset15;
nc
=5;
σ
=0
.
3
(c) dataset22;
nc
=45;
σ
=0
.
15
(d) spiral;
nc
=2;
σ
=0
.
13
Fig. 6.35 Some clustering solutions given by Spectral-Shi. Each label shows: the
dataset name; the pre-established number of clusters and the
σ
value.
Fig. 6.36
A sample of the Dutch Handwritten Numerals dataset.
[173]. The size of the corpus (the number of different words present in all the
stemmed dataset) defines the number of features. In this sub-sample, only the
words that occur at least 40 times were considered, thus obtaining a corpus
of 565 words. The results of the experiments with LEGClust and Spectral
clustering are shown in Table 6.21.