Name
Abstract | ||
|---|---|---|
The Kohonen self-organizing map can be an effective tool for knowledge discovery in large databases. One drawback however is the sensitivity of the Kohonen network's performance to the proper selection of its learning parameters. The manual setting of parameters such as the neighborhood size, the rate of decay of the neighborhood function, and the shape and form of the activation function is difficult and requires much experimentation for each data set. In this paper, we explore the use of simulated annealing to automate the process of parameter selection for a Kohonen network with minimal operator interaction. |
Year | Venue | Field |
|---|---|---|
2001 | INFORMATION REUSE AND INTEGRATION | Computer science,Knowledge extraction,Artificial intelligence,Machine learning,Software mining |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
1 | 3 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Ashraf M. Abdelbar | 1 | 243 | 25.43 |
| Awad H. Khalil | 2 | 3 | 4.21 |
| Emad A. M. Andrews | 3 | 18 | 2.74 |