This is the implementation of a paper entitled Pattern recognition frequency-based feature selection with multi-objective discrete evolution strategy on high_dimentional medical datasets. The proposed feature selection method (MPR-MDES) is a hybrid (filter + wrapper) method which is a direct furture work of Automatic frequency-based feature selection using discrete weighted evolution strategy.
Maximum Pattern Recognition (MPR) is a frequency-based filter ranking method, which belongs to a series of frequency-based rankers:
1- Mutual Congestion (MC). Publication year: 2019
2- Sorted Label Interference (SLI). Publication year: 2021
3- Sorted Label Interference-gamma (SLI-gamma). Publication year: 2022
4- Extended Mutual Congestion (EMC). Publication year: 2022
5- Maximum Pattern Recognition. Publication year: 2024
##################### Instruction #########################
After loading the corresponding dataset from your local drive:
1- Run lines 55-71 to calculate the summation of samples per label
2- Run lines 73-93 for Maximum Pattern Recognition (MPR)
3- Run lines 96-110 to create a dataset with top 20 features of MPR
4- Run lines 115-409 for Multi-objective Discrete Evolution Strategy and its corresponding functions
NOTICE: In Section 5.2 of the paper, we introduced using macro-averaging for the calculation of precision, recall, and F1-score in the multiclass classification. In Section 5.3, during the experimental analysis, we reported the results based on micro-averaging. However, the difference between macro-averaging and micro-averaging in our scenarios is not significant.
Cite this article
Hossein Nematzadeh, José García-Nieto, José F. Aldana-Montes, Ismael Navas-Delgado. Pattern recognition frequency-based feature selection with multi-objective discrete evolution strategy for high-dimensional medical datasets. Expert Systems with Applications, Volume 249, Part A, 2024, 123521, ISSN 0957-4174. https://doi.org/10.1016/j.eswa.2024.123521.