| Autor: |
Rodrigo Hernandez-Mazariegos, Jose Ortiz-Bejar, Jesus Ortiz-Bejar |
| Jazyk: |
angličtina |
| Rok vydání: |
2023 |
| Předmět: |
|
| Zdroj: |
Engineering Proceedings, Vol 39, Iss 1, p 71 (2023) |
| Druh dokumentu: |
article |
| ISSN: |
2673-4591 |
| DOI: |
10.3390/engproc2023039071 |
| Popis: |
This study compares three methods for optimizing the hyper-parameters m (embedding dimension) and τ (time delay) from Taken’s Theorem for time-series forecasting to train a Support Vector Regression system (SVR). Firstly, we use a method which utilizes Mutual Information for optimizing τ and a technique referred to as “Dimension Congruence” to optimize m. Secondly, we employ a grid search and random search, combined with a cross-validation scheme, to optimize m and τ hyper-parameters. Lastly, various real-world time series are used to analyze the three proposed strategies. |
| Databáze: |
Directory of Open Access Journals |
| Externí odkaz: |
|
