Data Set

The following describes the data sets that will be used throughout this investigation.

Glass identification

Vina performs a comparative study based on: rules, Beagle, the algorithm of the nearest neighbor and the discriminant analysis. Beagle is a product available through VRS Consulting, Inc. It is determined whether the vessel was of a "float" type glass or not.

The study of the classification of glass types was carried out through criminological research.

Data SetCharacteristic:	Multivariate	Number ofInstances:	214	Area:	Computer
AttributeCharacteristic:	Real	Number of Attributes	10
Associated Task:	Classification	Missing Values?	N/A

Attributes info:

1. Identifier: 1 to 214

2. RI: Refractive index

3. Na: Sodium

4. Mg: Magnesium

5. Al: Aluminum

6. Yes: Silicon

7. K: Potassium

8. Ca: Calcium

9. Ba: Barium

10. Fe: Iron

11. Type of glass:

    1. building_windows_float_processed

    2. building_windows_non_float_processed

    3. vehicle_windows_float_processed

    4. vehicle_windows_non_float_processed (none in this database)

    5. containers

    6. tableware

    7. headlamps

Ecoli

The data set created by Kenta Nakai by the Institute of Cellular and Molecular Biology, presents a knowledge base that contains the locality of proteins in various sites.

Data SetCharacteristic:	Multivariate	Number ofInstances:	336	Area:	Life
AttributeCharacteristic:	Real	Number of Attributes	8
Associated Task:	Classification	Missing Values?	N/A

Attributes info:

1. Sequence name: Access number of the SWISS-PROT database

2. mcg: the McGeoch method for recognition of the signal sequence.

3. gvh: von Heijne method of signal sequence recognition.

4. labio: signal peptidase II Result consensus sequence von Heijne. binary attribute.

5. chg: The presence of charge in N-terminal of the predicted lipoproteins. binary attribute.

6. aac: Score of the discriminant analysis of the amino acid content of the outer membrane and periplasmic proteins.

7. alm1: ALOM membrane score encompassing the region prediction program.

8. alm2: ALOM program score after excluding the putative scissile signal regions of the sequence.

Iris Data

This is perhaps the most well-known database found in the pattern recognition literature. Fisher's article is a classic in the field and is often referred to as this. The data set contains 3 classes of 50 cases each, where each class refers to a type of iris plant. One class is linearly separable from the other 2; The latter are not linearly separable from each other.

Data SetCharacteristic:	Multivariate	Number ofInstances:	150	Area:	Life
AttributeCharacteristic:	Real	Number of Attributes	4
Associated Task:	Classification	Missing Values?	N/A

Attributes info:

1. Sepal length in cm

2. Separate width in cm

3. Petal length in cm

4. Petal width in cm

5. Class

   1. Irisi setosa

   2. Irisi versicolor

   3. Iris virginica