Symmetrized Feature Selection with Stacked Generalization based Machine Learning Algorithm for the Early Diagnosis of Chronic Diseases

Many healthcare sectors generate huge amount of healthcare data every day. To determine the risk of various chronic conditions, the collected data can be correctly analyzed using machine learning algorithms. Using machine learning algorithms, Health care analytics predicts the presence or absence of disease and detects various diseases. Health care analytics will improve patient care and medical practitioners’ decision-making process by performing early disease detection process. Stacking is a meta-learning-based transfer learning approach that combines various machine learning algorithms. The advantage of layered generalization is that it allows to use the predictive capability of different machine learning algorithms to develop intermediate solutions, which improves the overall performance of the model. The proposed method utilizes a symmetric feature selection algorithm to extract the appropriate features from the chronic disease dataset. The proposed algorithm determines the merit of the feature by considering the entropy and information gain of the individual features. Metaheuristic algorithms can solve a wide range of complex real-world issues with high quality. A stacked generalization-based meta heuristic algorithm is used in the proposed approach to predict the possibility of chronic diseases. The experiment makes use of chronic disease datasets from the kaggle repository. The experimental findings reveal that the proposed symmetrized feature selection algorithm in combination with Stacked generalization-based Metaheuristics technique has obtained a better accuracy than the existing approaches.