Two-step Technique for Prediction Analysis using K-Means Clustering Algorithm

Abstract

The technique that is utilized for analyzing the complex data is known as data mining technique. As per the input dataset provided, the predictions are made for the data with the help of prediction analysis method. There are various new techniques proposed for the execution of prediction analysis technique. In this paper, the k-mean algorithm is utilized for categorizing the data. Further, for the classification of this data, the SVM classifier is applied. For improving the performance of prediction analysis in terms of accuracy the back propagation algorithm is used along with the k-mean clustering algorithm. For executing this proposed technique, the MATLAB tool is used. As per the experimental results it is concluded that the accuracy of the clustering algorithm is improved as well as the execution time utilized for prediction analysis is decreased.

Complex predictive analysis for health care: a comprehensive review

“…Clustering is a frequent evocative activity that involves identifying a limited collection of categories or clusters to characterise the data. In the clustering method, the technology of vector quantization was applied to predict readmissions in advanced medicine [7]. Classification is the discovery of a prediction characteristic of learning that classifies a data item into one of several specified classes [8]: -Statistical Taguchi and Jugulum [9] used the mahalanobis Taguchi system (MTS) approach for multivariable statistical assessment.…”