Journal
NEURAL COMPUTING & APPLICATIONS
Volume 29, Issue 4, Pages 925-934Publisher
SPRINGER LONDON LTD
DOI: 10.1007/s00521-016-2473-7
Keywords
Hamming distance; Modified cuckoo search (MCS); Feature selection; Rough sets; Levy flight
Categories
Ask authors/readers for more resources
In this paper, a modified cuckoo search algorithm with rough sets is presented to deal with high dimensionality data through feature selection. The modified cuckoo search algorithm imitates the obligate brood parasitic behavior of some cuckoo species in combination with the L,vy flight behavior of some birds. The modified cuckoo search uses the rough sets theory to build the fitness function that takes the number of features in reduct set and the classification quality into account. The proposed algorithm is tested and validated benchmark on several benchmark datasets drawn from the UCI repository and using different evaluation criteria as well as a further analysis is carried out by means of the Analysis of Variance test. In addition, the proposed algorithm is experimentally compared with the existing algorithms on discrete datasets. Finally, two learning algorithms, namely K-nearest neighbors and support vector machines are used to evaluate the performance of the proposed approach. The results show that the proposed algorithm can significantly improve the classification performance.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available