Adaptive feature selection method based on particle swarm optimization for gastric cancer prediction
L. Thara, Ramalingam Gunasundari · 2017
Medical / Clinical data mining attracted many researchers to work in that area. Bio medical engineers, computer scientists, analytic professionals are involved in such research area to develop an outcome which is nothing but a decision support system. Gastric cancer is one of the deadliest diseases that have more potential research scope. Only few research articles are published on gastric cancer prediction using computing and analytics. This research paper aims to propose an adaptive feature selection method based on particle swarm optimization for gastric cancer prediction. Performance metrics such as accuracy, hit rate and time taken for classification are taken into account for comparing the proposed AFS-PSO with the existing algorithms. 1127 real-time patients' records were obtained and the implementation of AFS-PSO is carried out using MATLAB and the results portray that AFS-PSO outperforms the existing algorithms in terms of accuracy, hit rate and elapsed time.