Integrated k-means Clustering with Nature Inspired Optimization Algorithm for the Prediction of Disease on High Dimensional Data

As per WHO report, 86.6% of deaths in China is due to chronic disease. Also, on average, 12 million people dies every year due to heart attack. On another side, we have a tremendous amount of heart disease data in the healthcare sector from which we have to find the hidden patterns and extract the knowledge for the prediction of disease. Data mining techniques play a vital role in the healthcare sector for diagnosis of disease and deeper understanding of biological data in healthcare. K-means clustering is an unsupervised, numerical, non-deterministic clustering technique for partitioning the data but converges the results to the local optima. In this paper, we have propose a Hybrid of K-means and Particle swarm optimization clustering algorithms due to their linear time complexity for prediction of disease and global optimization results. In our paper we have taken cholesterol level and age attributes of the patient from heart disease dataset for the prediction of disease in the improved k-means algorithm, which shows more accuracy than traditional k-means clustering algorithm also solve the problem of empty cluster.