Feature Selection Using Particle Swarm Optimization for Microarray Data Classification

Microarray data have large dimensions, redundancy, and noise. Feature selection (FS) has recently become an important approach in data preparation, particularly for high-dimensional data. The number of characteristics gathered in many machine learning applications is growing due to the rapid advancement of data collection technology. We know all features are not required to obtain high accuracy. The main motive of feature selection is to find the optimal subset of features. Here, we’ll combine classification methods like decision tree, and logistic regression and random forest with particle swarm optimization as a feature selection method. Additionally, we will display a brief comparison of each algorithmic model’s accuracy as well as each model’s fitness error plot. Here, the performance of the classifier changes depending on the dataset. Overall, LR and RF have shown the best performance when used with PSO.