Clustering algorithm of novel distribution function for dimensionality reduction using big data of OMICS: Health, clinical and biology research information

摘要

This paper defines the problem and design of the appropriate similarity with distribution function of the omics data is a critical objective. Data mining integrate methodical section at the large explosion of huge amount data that can be obtained to utilize and innovative knowledge. Researchers present and future the omics technologies permit to imitate as highly dimensional of omics data. This paper main objective to distance measure is using to concern as clustering algorithms. In order to particular tasks based on reduced high-dimensional omics data of dimensional reduction applying proposed distance measure is designed with distribution function based on PDF and CDF using designed average function and distance measure. It is using training and testing reduce data based on clusters. Reduced data using with class and without class of the OMICS data with accurate results.