| Resumen |
This paper extends the Gamma associative classifier, making it able to deal with hybrid and
incomplete data. In addition, it also encompasses the gamma rough sets model for dealing with such data,
introducing the extended gamma rough sets. Some properties of such sets are demonstrated in this paper.
In turn, the novel extended gamma rough sets are used to improve the extended gamma associative classifier
by selecting the instances. The results indicate that the selection of instances significantly improves the
accuracy of the extended gamma associative classifier while reducing its computational cost. |
