Research On Data Quality Evaluation Model Of Mobile Medical Big Data

As mobile health care enters the era of big data,smart medical big data is gradually formed with the amount of data increasing,and the data generated by equipment monitoring growing rapidly.The variety of data sources,and the dispersion and heterogeneity of mobile medical data have a serious impact on data mining and statistical analysis.Therefore,how to conduct data quality assessment in the context of medical big data is a topic of practical application.Data quality assessment before data use is a necessary prerequisite for medical big data scenarios and has practical value.At the same time,the rapid development of big data technology provides technical support for analyzing data and mining data value through big data.This paper deeply studies the big data quality assessment process and big data technology,and proposes a mobile medical big data quality assessment method based on trusted analysis.Firstly,combined with the characteristics of current mobile medical electronic records,a credible analysis model is proposed to analyze the credibility of data from indirect,direct and comprehensive perspectives,then eliminating untrustworthy data.Secondly,the data quality is evaluated from five dimensions:accuracy,completeness,consistency,normativeness and timeliness.Thirdly,for the dimensional evaluation of data quality,the analytic hierarchy process is used to obtain the comprehensive data quality assessment results.Finally,as for the operation of the algorithm,a distributed experiment based on spark SQL is designed to improve the data analysis of massive medical big data.In terms of verification,this paper carries out an example analysis through simulation data,and uses electronic medical records as an empirical object.Experiments show that the model can effectively identify untrusted data in the data set,and the trusted analysis improves the data quality assessment results.The data quality assessment results are more obj ective,and meet the actual needs and the big data application scenarios.With the increasing amount of data,the parallelized big data quality assessment method can significantly reduce the running time compared with the stand-alone environment,which proves that the data quality assessment model of medical big data based on trusted analysis has a better data quality assessment capability in the face of massive data.