作者: Yi Liu , Xingchun Diao , Jianjun Cao , Lei Zhang
DOI: 10.1007/978-981-10-7179-9_6
关键词: Ant colony optimization algorithms 、 Feature selection 、 Genetic algorithm 、 Evolutionary computation 、 Evolutionary algorithm 、 Computer science 、 Truncation selection 、 Memetic algorithm 、 Evolutionary programming 、 Data mining
摘要: In order to improve the feature selection stability based on evolutionary algorithms, an algorithms’ improvement system is proposed. Three Filter methods’ results are aggregated provide information, and classification accuracy adopted as two optimization objectives. Weighted sum, weighted product biobjective methods together applied system’s models. Ant colony optimization, particle swarm genetic algorithm used testing experiments taken benchmark datasets. The show that proposed can of efficiently their performance simultaneously.