Identification and Analysis of Breast Cancer Disease using Swarm and Evolutionary Algorithm
Gyanaranjan Shial, Sabita Sahoo, Sibarama Panigrahi
Abstract
Medical data analysis is an important field of research, among which analysis of disease detection or patient's survival after a specific treatment is of prime importance. The recent survey suggests that breast cancer is the most common life-threatening disease among women in the age group (35–50). Therefore, the focus of this study aims on breast cancer disease detection and survival prediction using UCI machine learning standard benchmark breast cancer datasets. The research is conducted through simulation work that includes five meta-heuristic-based clustering models. To access the performances of each models accuracy, specificity, Precision, F-score and MCC performance measures are used. To check superiority among each clustering model, Duncan's multiple range test is conducted with a 95 % significance level. In addition, the Friedman and Nymenyi hypothesis test is applied to access the ranks and verify the stability of each model for breast cancer disease analysis.