Bio Inspired Algorithms for Dimensionality Reduction and Outlier Detection in Medical Datasets
S. Vijayarani, C. Sivamathi, Mrs.S. Maria Sylviaa · International Journal of Advanced Networking and Applications · 2022
Dimensionality Reduction is one of the useful techniques used in number of applications in order to reduce the number of features to improve the productivity and efficiency of the task. Clustering is one of the influential tasks in data mining. Dimensionality reductions are used in data mining, Image processing, Networking, Mobile computing, etc. The elementary intention of this work is to apply dimensionality reduction algorithms and then cluster the datasets to detect outliers. A bio-inspired ACO (Ant Colony optimization) algorithm has been proposed to reduce dimensionality. Also another bio-inspired algorithm FA (Firefly Algorithm) has been proposed to detect outliers. The three distinct medical datasets: thyroid dataset, Oesophagal dataset and Heart disease dataset are used for experimental results.