Feature selection based on attributes clustering

Feature selection is an important preprocessing step in machine learning; it aims to remove redundant and irrelevant features, improve the efficiency of machine-learning algorithms, and increase the model’s interpretability. In this paper, we proposed a new feature selection algorithm based on attributes clustering. Our method consists of two main steps. In the first step, we separate all attributes into K groups (clusters). The next step consists of choosing the best subset of features from the clusters. The performance of the proposed method is evaluated on seven standard benchmark datasets from the UCI repository.