KEEL: A software tool to assess evolutionary algorithms for Data Mining problems (regression, classification, clustering, pattern mining and so on)

KEEL-dataset - data set description

Corel Image Features (layoutHistogram) data set

Description
Files and additional references

Description

This section describes main characteristics of the layoutHistogram data set and its attributes:

General information

Corel Image Features (layoutHistogram) data set
Type	Unsupervised	Origin	Real world
Features	33	(Real / Integer / Nominal)	(32 / 1 / 0)
Instances	66616	Missing values?	No

Attribute description

Attribute	Domain	Attribute	Domain	Attribute	Domain
ID	[1, 68040]	H2S1I3	[0.000000, 0.250024]	H3S2I2	[0.000000, 0.250026]
H1S1I1	[0.000000, 0.250026]	H2S1I4	[0.000000, 0.250021]	H3S2I3	[0.000000, 0.250026]
H1S1I2	[0.000000, 0.250026]	H2S2I1	[0.000000, 0.250000]	H3S2I4	[0.000000, 0.250026]
H1S1I3	[0.000000, 0.250026]	H2S2I2	[0.000000, 0.250000]	H4S1I1	[0.000000, 0.250012]
H1S1I4	[0.000000, 0.250026]	H2S2I3	[0.000000, 0.249687]	H4S1I2	[0.000000, 0.250012]
H1S2I1	[0.000000, 0.250026]	H2S2I4	[0.000000, 0.249896]	H4S1I3	[0.000000, 0.250014]
H1S2I2	[0.000000, 0.250026]	H3S1I1	[0.000000, 0.250026]	H4S1I4	[0.000000, 0.249802]
H1S2I3	[0.000000, 0.250026]	H3S1I2	[0.000000, 0.250026]	H4S2I1	[0.000000, 0.250015]
H1S2I4	[0.000000, 0.250026]	H3S1I3	[0.000000, 0.250026]	H4S2I2	[0.000000, 0.250004]
H2S1I1	[0.000000, 0.250025]	H3S1I4	[0.000000, 0.250026]	H4S2I3	[0.000000, 0.250011]
H2S1I2	[0.000000, 0.250012]	H3S2I1	[0.000000, 0.250026]	H4S2I4	[0.000000, 0.250000]

Additional information

This dataset is a part of the Corel Image Features data set of the UCI Repository. It contains image features extracted from a Corel image collection (http://corel.digitalriver.com/). The ID attribute is common in each of the 4 parts of the data set (colorHistogram, colorMoments, colorTexture and layoutHistogram).

In this part, the features represent the color histogram layout of each one where HS color space is divided into 8 subspaces (8 colors : 4 ranges of H and 2 ranges of S). Moreover, each image is splitted into 4 sub-images (one horizontal split and one vertical split). The value in each dimension is the density of each color in the entire sub-image.

Warning: Note that this dataset has only data from 66616 images, due to the data related to 1424 images (ID: 62482-63905 ) were missing in the original source (UCI Repository).

Files and additional references

In this section you can download some files related to the layoutHistogram data set:

The complete data set already formatted in KEEL format can be downloaded from here .
The header file associated to this data set can be downloaded from here .
This is not a native data set from the KEEL project. It has been obtained from the UCI Machine Learning repository. The original page where the data set can be found is: http://archive.ics.uci.edu/ml/datasets/Corel+Image+Features.