Electronic Government an International Journal · 2020 · 13 citations · 0 references
ComputingHealth InformaticsNephrologyClinical Decision SupportMachine LearningOptimizationDecision Support In HealthcareFeature SelectionGenetic AlgorithmsOptimization For Machine LearningPredictive Health ModelingArtificial Bee ColonyBenchmark DatasetsFeature Selection For ClassificationFeature Selection In DiagnosisGrey Wolf OptimizerHarmony SearchHybrid Feature SelectionParticle Swarm Optimization
Prediction of kidney disease (KD) gains more importance in the medical decision support systems. As the medical dataset are massive in size, effective techniques are required to produce accurate results. This paper proposes a hybrid harmony search (HM-L) algorithm with Levi distribution to properly predict KD at appropriate time. In this research work, correlation-based feature selection (CFS) is used as a feature selection technique. The effectiveness of hybrid harmony search (HS) algorithm is validated by employing it against a set of dataset. The obtained results of applied datasets without and with feature selection are compared to one another. The experimental results imply that the HM-L algorithm attains significant results than existing methods such as HS algorithm, biogeography optimisation algorithm (BBO), grey wolf optimisation (GWO), AL particle swarm optimisation algorithm (ALPSO) and artificial bee colony (ABC) algorithm. The presented HM-L model attains a sensitivity of 96, specificity of 93.33, accuracy of 95, F-score of 96 and kappa value of 0.89.