TY - JOUR
T1 - AISAC
T2 - An artificial immune system for associative classification applied to breast cancer detection
AU - González-Patiño, David
AU - Villuendas-Rey, Yenny
AU - José Argüelles-Cruz, Amadeo
AU - Camacho-Nieto, Oscar
AU - Yáñez-Márquez, Cornelio
N1 - Publisher Copyright:
© 2020 by the authors.
PY - 2020/1/1
Y1 - 2020/1/1
N2 - Early breast cancer diagnosis is crucial, as it can prevent further complications and save the life of the patient by treating the disease at its most curable stage. In this paper, we propose a new artificial immune system model for associative classification with competitive performance for breast cancer detection. The proposed model has its foundations in the biological immune system; it mimics the detection skills of the immune system to provide correct identification of antigens. The Wilcoxon test was used to identify the statistically significant differences between our proposal and other classification algorithms based on the same bio-inspired model. These statistical tests evidenced the enhanced performance shown by the proposed model by outperforming other immune-based algorithms. The proposed model proved to be competitive with respect to other well-known classification models. In addition, the model benefits from a low computational cost. The success of this model for classification tasks shows that swarm intelligence is useful for this kind of problem, and that it is not limited to optimization tasks.
AB - Early breast cancer diagnosis is crucial, as it can prevent further complications and save the life of the patient by treating the disease at its most curable stage. In this paper, we propose a new artificial immune system model for associative classification with competitive performance for breast cancer detection. The proposed model has its foundations in the biological immune system; it mimics the detection skills of the immune system to provide correct identification of antigens. The Wilcoxon test was used to identify the statistically significant differences between our proposal and other classification algorithms based on the same bio-inspired model. These statistical tests evidenced the enhanced performance shown by the proposed model by outperforming other immune-based algorithms. The proposed model proved to be competitive with respect to other well-known classification models. In addition, the model benefits from a low computational cost. The success of this model for classification tasks shows that swarm intelligence is useful for this kind of problem, and that it is not limited to optimization tasks.
KW - Artificial immune systems
KW - Breast cancer
KW - Classification
KW - Swarm intelligence
UR - http://www.scopus.com/inward/record.url?scp=85081238875&partnerID=8YFLogxK
U2 - 10.3390/app10020515
DO - 10.3390/app10020515
M3 - Artículo
SN - 2076-3417
VL - 10
JO - Applied Sciences (Switzerland)
JF - Applied Sciences (Switzerland)
IS - 2
M1 - 515
ER -