Name
Title | ||
|---|---|---|
New Hybrid Methodology Based on Particle Swarm Optimization with Genetic Algorithms to Improve the Search of Parsimonious Models in High-Dimensional Databases |
Abstract | ||
|---|---|---|
Our previous PSO-PARSIMONY methodology (a heuristic to search for accurate and low-complexity models with particle swarm optimization) shows a good balance between accuracy and complexity with small databases, but gets stuck in local minima in high-dimensional databases. This work presents a new hybrid methodology to solve this problem. First, we incorporated to PSO-PARSIMONY an aggressive mutation strategy to encourage parsimony. Second, a hybrid method between PSO and genetic algorithms was also implemented. With these changes, particularly with the second one, improvements were observed in the search for more accurate and low-complexity models in high-dimensional databases. |
Year | DOI | Venue |
|---|---|---|
2022 | 10.1007/978-3-031-15471-3_29 | Hybrid Artificial Intelligent Systems |
Keywords | DocType | ISSN |
PSO-PARSIMONY, Hybrid method, Parsimonious modeling, Auto machine learning, GA-PARSIMONY | Conference | 0302-9743 |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
3 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Divasón Jose | 1 | 0 | 0.34 |
| Pernia-Espinoza Alpha | 2 | 0 | 0.34 |
| Martinez-de-Pison Francisco Javier | 3 | 0 | 0.34 |