Binary Archimedes Optimization Algorithm based Feature Selection for Regression Problem

The use of datasets became paramount in many searches in one hand, on the other hand the rapidly growth of data size involves computational complexity and reduces model performances, this encourage us to find new methods to deal with this problem. Features Selection is the one of the main task used...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2022 4th International Conference on Pattern Analysis and Intelligent Systems (PAIS) s. 1 - 7
Hlavní autoři: Amine, Djermane, Hichem, Haouassi, Soumia, Zertal
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 12.10.2022
Témata:
Archimedes optimization algorithm
Classification algorithms
Data mining
Data models
Feature extraction
feature selection
Metaheuristics
Prediction algorithms
Predictive models
regression
Search problems
swarm-based optimization
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:The use of datasets became paramount in many searches in one hand, on the other hand the rapidly growth of data size involves computational complexity and reduces model performances, this encourage us to find new methods to deal with this problem. Features Selection is the one of the main task used to resolve this issue. In this paper we propose a novel features selection method for regression task based on AOA (Archimedes Optimization Algorithm), experimental results shows that the proposed method can efficiently reduce dataset size and improve model performance.
DOI:10.1109/PAIS56586.2022.9946903