Abstract
Binary butterfly optimization approach (bBOA) is a recent high performing feature selection algorithm presented in 2018 which is based on the food foraging behavior of butterflies. This paper tries to improve the structure of the bBOA to enhance its classification accuracy, dimension reduction and reliability in feature selection task for who are interested in the fields of data mining and pattern recognition. The new initialization strategy and differential evolution strategy are applied to reduce the randomness of bBOA's initialization and local search process. Then, a new parameter is added to make the bBOA's transfer function more adaptive to the change of exploration and exploitation. Besides, evolution population dynamics (EPD) mechanism is employed as an extension of bBOA. The new method called optimization and extension of binary butterfly optimization approaches (OEbBOA) is tested with the K nearest neighbor classier in which twenty UCI datasets and seven recent algorithms are utilized to assess the performance of the OEbBOA algorithm. The experimental results and nonparametric Wilcoxons rank sum test confirm the efficiency of the proposed OEbBOA in maximizing classification accuracy while minimizing the number of features selected.
Highlights
Feature selection is a process of selecting some of the most effective features from the original features to reduce the data dimensions [1], [2], which is a key pre-processing step in machine learning, data mining and pattern recognition
Due to the high global search capability, evolutionary computation has gained more and more attention on the field of feature selection in recent years [6], [7]. These algorithms have the ability to exploit useful population information to find the optimal solution [8]. Some of these algorithms are binary grasshopper optimisation algorithm approaches for feature selection [9], whale optimization approaches for wrapper feature selection [10], hybrid whale optimization algorithm with simulated annealing for feature selection [11]
The mechanism of ObBOA_NIS has the best promotion effect on classification accuracy and dimension reduction and its stability is second only to OEbBOA according to Table 4 and Table 6, but its convergence speed is lower than ObBOA_PFV and EbBOA_EPD and its time cost is relatively high according to section V
Summary
Feature selection is a process of selecting some of the most effective features from the original features to reduce the data dimensions [1], [2], which is a key pre-processing step in machine learning, data mining and pattern recognition. Due to the high global search capability, evolutionary computation has gained more and more attention on the field of feature selection in recent years [6], [7] These algorithms have the ability to exploit useful population information to find the optimal solution [8]. Binary butterfly optimization approaches for feature selection (bBOA) [8] is recently proposed to solve feature selection problem. The transfer function is improved, in which the fragrance of butterflies is added as a new parameter, to enable the transfer function to adjust the development and exploration ability of algorithm according to the number of iterations and the health level of butterflies. Butterfly replacement mechanism is proposed to eliminate butterflies whose fitness is relatively small and replace them with new butterflies born
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
