Abstract
One predictive system is superior to another one if it delivers a higher percentage of correct predictions. Two of the paradigms that have proven to be effective in design of viable predictive systems are `Rough Sets' and `Neural Network'. In this paper we investigate the integration of these two methodologies for the purpose of building a hybrid predictive system superior to a pure neural network. Our findings are validated using fifty training-test sets pairs generated out of our original data set using the statistical approach of random sampling.
| Original language | English |
|---|---|
| Pages | 499-504 |
| Number of pages | 6 |
| State | Published - 1999 |
| Event | Proceedings of the 1999 Artificial Neural Networks in Engineering Conference (ANNIE '99) - St. Louis, MO, USA Duration: Nov 7 1999 → Nov 10 1999 |
Conference
| Conference | Proceedings of the 1999 Artificial Neural Networks in Engineering Conference (ANNIE '99) |
|---|---|
| City | St. Louis, MO, USA |
| Period | 11/7/99 → 11/10/99 |