Research on Drug Quality Risk Early Warning Model Based on Multiple Improved Optimizing Algorithm

摘要

The problem we try to solve in this paper is to improve the effectiveness of drug supervision and inspection by using large quantities of drug inspection data for early warning of quality risk. The approach we adopt to solve the problem is characteristics of drug testing data Select Apriori algorithm, Markov chain, 3a(three-sigma)-AHP algorithm, CV-SES (coefficient of variation, single exponential smoothing) Algorithm, through the research on drug quality there are many demerits in the application of drug testing and some improving plans are put forward. Such as to reduces original data dimension, redesign create frequent dataset, scope normalization, correction function for coefficients, and finally establish a drug quality risk early warning model. The impacts on our obtained results are validated by using the 2013-2016 Year drug sampling and testing data of Guangdong institute for drug control, the early warning information generated by this model have important sense to improve the efficiency and accuracy of drug quality risk early warning.