Experimentation and Results Discussion
All experimentation is focused on the previously shown approaches to fuzzy information granulation. Most experiments were done with benchmark datasets which will be described in the following paragraph. Two benchmark dataset types were used: classificatio
- PDF / 368,771 Bytes
- 13 Pages / 439.37 x 666.142 pts Page_size
- 40 Downloads / 250 Views
Experimentation and Results Discussion
All experimentation is focused on the previously shown approaches to fuzzy information granulation. Most experiments were done with benchmark datasets which will be described in the following paragraph. Two benchmark dataset types were used: classification, and identification. Classification type benchmarks datasets are described in Table 4.1, showing name, number of features, number of classes, and sample size. Where these datasets are iris, wine, glass identification, seeds, image segmentation, Haberman’s survival, and mammographic mass [1]. The iris dataset, has 4 input features (petal length, petal width, sepal length, and sepal width), and 3 outputs (iris setosa, iris virginica, and iris versicolor) with 50 samples of each flower type, with a total of 150 elements in the dataset. The wine dataset, with 13 input features of different constituents (Alcohol, malic acid, ash, alkalinity of ash, magnesium, total phenols, flavanoids, nonflavanoid phenols, proanthocyanidins, color intensity, hue, OD280/OD315 of diluted wines, and proline) identifying 3 distinct italian locations where the wine came from. With 59, 71, and 48 elements respectively in each class, for a total of 178 elements in the whole dataset. The glass identification dataset, has 9 input variables (refractive index, sodium, magnesium, aluminum, silicon, potassium, calcium, barium, and iron), and 7 classes (building windows float processed, building windows non float processed, vehicle windows float processed, containers, tableware, and headlamps). With 70, 76, 17, 13, 9, and 29 elements respectively in each class, for a total of 214 elements in the whole dataset. The seeds dataset, with 7 input features (area, perimeter, compactness, length of kernel, width of kernel, asymmetry coefficient, and length of kernel groove) and 3 output classes (Kama, Rosa, and Canadian) with 70 samples of each class, for a total of 210 elements in the dataset. The image segmentation dataset, with 19 input features (column of the center pixel region, row of the center pixel region, number of pixels in a region, result of line extraction algorithm, lines of high contrast, mean contrast of horizontally adjacent pixels in the region, standard deviation contrast of horizontally adjacent pixels in the region, mean contrast of vertically adjacent pixels in the region, standard deviation contrast of vertically adjacent pixels in the region, © The Author(s) 2017 M.A. Sanchez et al., Type-2 Fuzzy Granular Models, SpringerBriefs in Computational Intelligence, DOI 10.1007/978-3-319-41288-7_4
37
38
4
Experimentation and Results Discussion
Table 4.1 Description of classification benchmark datasets used for experimentation Dataset
No. of features
No. of classes
Sample size
Iris Wine Glass Seed Image segmentation Haberman’s survival Mammographic mass
4 13 9 7 19 3 5
3 3 6 3 7 2 2
150 178 214 210 2310 306 830
Table 4.2 Description of identification benchmark datasets used for experimentation Dataset
No. of inputs
No. of outputs
Sample size
Data Loading...
