Human Health Evaluation Based On Improved AP Clustering Algorithm

The acceleration of the pace of life has increased the pressure on people,and the number of sub-healthy people has also increased.People's health concept has gradually changed from "treatment with disease" to "prevention without disease".In order to achieve "prevention before disease”,a health assessment is an essential link.At present,the main methods of health assessment include subjective survey methods and physiological parameter detection methods,but the problems faced are the lack of objectivity and uniform assessment standards,and the inability to comprehensively consider the fluctuation of multi-source physiological parameters,resulting in inaccuracy and instability of the assessment.In view of this,how to select multiple physiological parameters and choose appropriate machine learning algorithms to comprehensively and scientifically evaluate human health is the focus of this article.In order to solve the above problems,this paper chooses the AP clustering algorithm to cluster multiple physiological parameters to achieve health assessment.The AP clustering algorithm is simple and fast,and it can obtain better clustering results than traditional clustering algorithms in solving clustering problems of many data sets.Since its inception,many areas have used this algorithm to implement clustering,such as business intelligence,digital healthcare,and so on.However,the AP clustering algorithm still has problems such as the limitation of biasing parameters and damping factors on the clustering effect of the algorithm and the high complexity of the algorithm.In view of this,this article gives a corresponding solution.This article selects six basic physiological parameters of human blood pressure,pulmonary arterial pressure,heart rate,blood oxygen saturation,body temperature,and respiratory rate.The improved AP clustering algorithm is used to evaluate human health.Finally,technologies such as Java,JSP,and database are combined.Designed and developed a human health assessment system to achieve human health assessment.The main research work of this article is:(1)Research on improved AP clustering algorithm.The first is to improve the AP clustering algorithm based on the genetic algorithm to determine the values of the bias parameters and damping factors that will enable the algorithm to produce optimal clustering results and to automatically eliminate and converge after the algorithm has oscillated.Aiming at the problem of high algorithm complexity,and improved AP clustering algorithm based on the density peak clustering algorithm is proposed;finally,these two algorithms are combined to obtain the final improved AP clustering algorithm.This algorithm can not only obtain the best biasing parameters and damping factor values,which can further improve the clustering effect of the algorithm,but also shorten the running time of the algorithm to a certain extent.(2)Research on human health assessment algorithm based on improved AP clustering algorithm.It is mainly the acquisition and extraction of data sources,data pre-processing,establishment of human health assessment models,testing,and results analysis.(3)Design and implementation of a human health assessment system.Using Java,JSP,Echarts,CCS,and My SQL databases,functional modules such as user registration and login,user basic information management,physiological information input and management,health assessment and guidance,and historical data visualization are implemented.The research in this topic can not only evaluate people's health status but also provide medical personnel with a relatively complete electronic health record of the human body.The results of simulation experiments show that the proposed method is not only better than other methods in terms of accuracy but also shortens the running time to a certain extent.Therefore,the research of this topic can not only make some contributions to digitally-driven electronic health management but also provide new ideas in the improvement and application of AP clustering algorithms.