Health Big Data Classification Using Improved Radial Basis Function Neural Network and Nearest Neighbor Propagation Algorithm

Congshi Jiang,Yihong Li

doi:10.1109/access.2019.2956751

Abstract

Health Big Data Classification Using Improved Radial Basis Function Neural Network and Nearest Neighbor Propagation Algorithm

Highlights

With the rapid development and application of mobile Internet, computer storage technology and cloud computing technology, human society is entering an era of information explosion
This paper proposes a new method of health big data classification using radial basis function neural network, and proposes an algorithm of constructing variable basis width neural network model
Health big data refers to all data closely related to human health and public health, including data generated from birth, infant health care, vaccine injection, school physical examination, work physical examination, medical treatment, hospitalization, exercise, sleep, death and other life cycles

Summary

INTRODUCTION

With the rapid development and application of mobile Internet, computer storage technology and cloud computing technology, human society is entering an era of information explosion. The main characteristics of medical record data are relatively high dimension and complex type. Health big data has a variety of complex forms of existence, contains rich medical value, and health big data can make the whole medical resources more efficient. Li: Health Big Data Classification Using Improved RBFNN and Nearest Neighbor Propagation Algorithm. Health big data classification can effectively improve the level of medical and health services and management, help medical staff to carry out auxiliary diagnosis, improve the efficiency of doctors and the accuracy of diagnosis, so as to make precision medicine truly possible. This paper proposes a new method of health big data classification using radial basis function neural network, and proposes an algorithm of constructing variable basis width neural network model. On the basis of subtracting clustering algorithm and K - means algorithm to determine clustering center, the maximum distance between sample and clustering center is selected as the base width, and the value of base width is adaptive with the optimization of clustering center the new algorithm improves the classification accuracy and shortens the convergence time

RELATED RESEARCH

ADJUSTMENT OF SIMILARITY MATRIX

THE STEPS OF VARIABLE BASIS WIDTH RBF NEURAL NETWORK CLASSIFIER

CLASSIFICATION EFFECTIVENESS EVALUATION

Findings

CONCLUSION