Rule extraction from support vector machines: A review
Neurocomputing
| Hi-index | 0.00 |
In this work, a procedure for rule extraction fromsupport vector machines is proposed. Our method, firstdetermines prototype vectors by means of k-means. Then,these vectors are combined with the support vectors usinggeometric methods to define ellipsoids in the input space,which are later translated to if-then rules. In this way, it ispossible to give an interpretation to the knowledgeacquired by the SVM. On the other hand, the extractedrules render possible the integration of SVMs withsymbolic AI systems.