Feature selection method based on binary bee foraging algorithm

The invention discloses a feature selection method based on a binary bee foraging algorithm, and relates to the technical field of data mining. For the existing requirement for feature selection of high-dimensional data, the method divides a bee population into reconnaissance bees, collection bees and to-be-worked bees, different sub-populations are updated by adopting different strategies, and a search space is limited between [0, 1]. The reconnaissance bees adopt a random search mode, the worker bees adopt a V-shaped transfer function to perform binary position updating, and the worker bee position updating adopts a single-dimensional position negation strategy. The method has the characteristics of high feature selection accuracy, strong robustness, high search efficiency and the like, and efficient solution of feature selection of the high-dimensional data set is realized. 本发明公开了一种基于二进制蜜蜂觅食算法的特征选择方法，涉及数据挖掘技术领域。针对现有的对高维数据进行特征选择的需求，该方法将蜜蜂种群分为侦察蜂、采蜜蜂和待工蜂，不同的子种群采用不同的策略进行更新，并将搜索空间限制在[0,1]之间。其中，侦察蜂采用随机搜索的方式，采蜜蜂