Feature selection is a crucial method for discovering relevant features in high-dimensional data. However, most studies primarily focus on completely labeled data, ignoring the frequent occurrence of missing labels in real-world problems. To address high-dimensional and label-missing problems in data classification simultaneously, we proposed a semisupervised bacterial heuristic feature selection algorithm. To track the label-missing problem, a k-nearest neighbor semisupervised learning strategy is designed to reconstruct missing labels. In addition, the bacterial heuristic algorithm is improved using hierarchical population initialization, dynamic learning, and elite population evolution strategies to enhance the search capacity for various feature combinations. To verify the effectiveness of the proposed algorithm, three groups of comparison experiments based on eight datasets are employed, including two traditional feature selection methods, four bacterial heuristic feature selection algorithms, and two swarm-based heuristic feature selection algorithms. Experimental results demonstrate that the proposed algorithm has obvious advantages in terms of classification accuracy and selected feature numbers.

No abstract

The information transfer mechanism within the population is an essential factor for population‐based multiobjective optimization algorithms. An efficient leader selection strategy can effectively help the population to approach the true Pareto front. However, traditional population‐based multiobjective optimization algorithms are restricted to a single global leader and cannot transfer information efficiently. To overcome those limitations, in this paper, a multiobjective bacterial colony optimization with dynamic multi‐leader co‐evolution (MBCO/DML) is proposed, and a novel information transfer mechanism is developed within the group for adaptive evolution. Specifically, to enhance convergence and diversity, a multi‐leaders learning mechanism is designed based on a dynamically evolving elite archive via direction‐based hierarchical clustering. Finally, adaptive bacterial elimination is proposed to enable bacteria to escape from the local Pareto front according to convergence status. The results of numerical experiments show the superiority of the proposed algorithm in comparison with related population‐based multiobjective optimization algorithms on 24 frequently used benchmarks. This paper demonstrates the effectiveness of our dynamic leader selection in information transfer for improving both convergence and diversity to solve multiobjective optimization problems, which plays a significant role in information transfer of population evolution. Furthermore, we confirm the validity of the co‐evolution framework to the bacterial‐based optimization algorithm, greatly enhancing the searching capability for bacterial colony.