A new paper published by SDA-Lab in Journal of Structural Integrity and Maintenance. The paper is authored by Dr. Ramin Ghiasi and Dr. Abdollah Malekjafarian from SDA-Lab.
In this paper, a novel wrapper method proposed based on an Advanced version of the Binary Slime Mould Algorithm (ABSMA) for feature selection in structural/substructural health monitoring datasets. Feature selection is the automatic process of reducing the number of inputs from the set of features by choosing the relevant variables before feeding them to the machine learning model. The motive is to reduce the irrelevant data(noise) and choose relevant features. Wrapper methods measure the importance of a feature based on its usefulness while training the Machine Learning model on it. In this paper, the capability of the proposed ABSMA for Feature selection is compared with 5 well-known wrapper-based methods and 4 filter-based methods. Furthermore, the classification accuracy of the neural network as a learning model of wrapper-based Feature selection methods is compared with 5 other classifier algorithms.
Paper can be accessed for free at the following link:
https://www.tandfonline.com/doi/full/10.1080/24705314.2023.2230398