| Article ID: | iaor20011595 |
| Country: | Cuba |
| Volume: | 21 |
| Issue: | 3 |
| Start Page Number: | 247 |
| End Page Number: | 259 |
| Publication Date: | Sep 2000 |
| Journal: | Revista de Investigacin Operacional |
| Authors: | Elizondo William Castillo, Zelaya Javier Trejos |
| Keywords: | neural networks |
We present a study of the main methods for two-mode classification, hierarchical and partitioning. A Lance & Williams recurrence formula is proposed for hierarchical two-mode classification using the Eckes & Orlik criterion. Also, the simulated annealing technique is implemented for global optimization in two-mode partitioning, obtaining better results than usual methods like alternating exchanges and dynamic clusters or