Abstract: A key large data activity, feature selection (FS), reduces the "curse of dimensionality" by choosing a meaningful feature subset to improve classification performance. Search algorithms may be constrained by FS techniques as the number of attributes grows. To achieve equivalent or better classification performance and increase computing efficiency, a subset of pertinent characteristics are chosen from a large number of original features using the feature selection process. Particle swarm optimization (PSO) is a global search metaheuristic that can swiftly and with few presumptions search a space with many dimensions. An FSPSO method known as particle swarm optimization (PSO) has recently attracted a lot of attention from experts in the field. Following a basic explanation of feature selection and PSO, a review of recent PSO for feature selection work is given.
Keywords: Particle Swarm Optimization, Feature Selection, Classification, Mutation Operator.
|
DOI:
10.17148/IJARCCE.2023.12748
