Geometric concept data generator for testing classification algorithms
It is used to create data sets representing a geometric concept. The created data sets have two descriptive attributes corresponding to the two coordinates of a cartesian point, and a class attribute which is true if the point lie inside the geometric concept and false otherwise. A number of geometric concepts, with different boundary complexities, are available to be modeled.
It can be used to study classification algorithms performance, data complexity characterization, and meta-learning among other machine learning topics. Particularly, it was used in:
- Sanchez Tarrago, D., Herrera, F., Bello, R.: On the usefulness of Rough Sets Theory for Data Complexity: A case study on the domain of competence of C4.5. Central University Marta Abreu de Las Villas, Santa Clara, Cuba (2010). (text)
Developed with:
- Java 1.8
- NetBeans IDE 8.2
Dependencies:
- Weka 3.7
- meta-learning project (only for rough-set metric analisys)