This section describes main characteristics of the penbased data set and its attributes:
General information
Pen-Based Recognition of Handwritten Digits Multi-class Imbalanced data set |
Type | Imbalanced | Origin | Real world |
Features | 16 | (Real / Integer / Nominal) | (16 / 0 / 0) |
Instances | 1100 |
IR | 1.95 |
% Positive instances | 33.9 | % Negative instances | 66.1 | Missing values? | No |
Attribute description
Attribute | Domain | Attribute | Domain |
at1 | [0.0, 100.0] | at9 | [0.0, 100.0] |
at2 | [0.0, 100.0] | at10 | [0.0, 100.0] |
at3 | [0.0, 100.0] | at11 | [0.0, 100.0] |
at4 | [0.0, 100.0] | at12 | [0.0, 100.0] |
at5 | [0.0, 100.0] | at13 | [0.0, 100.0] |
at6 | [0.0, 100.0] | at14 | [0.0, 100.0] |
at7 | [0.0, 100.0] | at15 | [0.0, 100.0] |
at8 | [0.0, 100.0] | at16 | [0.0, 100.0] |
class | {0, 1, 2, 3, 4, 5, 6, 7, 8, 9} |
Additional information
An imbalanced version of the Pen-Based Recognition of Handwritten Digits data set, where there are some classes with a small number of examples while other classes have a large number of examples.
In this section you can download some files related to the penbased data set:
- The complete data set already formatted in KEEL format can be downloaded from
here.
- A copy of the data set already partitioned by means of a 5-folds cross validation procedure can be downloaded from here.
- The header file associated to this data set can be downloaded from here.
|