BP neural network-based drug sales prediction and decision-making method and system

Abstract

The invention provides a BP neural network-based drug sales prediction and decision-making method and system. The method comprises the steps of obtaining related data information of drug historical sales volume, performing preprocessing and abnormal data inspection on the related data information, and performing prediction by adopting a basic BP neural network based on the preprocessed drug historical sales volume data; correcting the short-term prediction result of the basic BP neural network based on the historical data of drug sales and the priori data of drug sales; and performing sales decision based on the output prediction result. According to the invention, the prediction precision is improved through a neural network drug sales prediction algorithm based on big data, and the prediction precision is further improved through each correction step. The drug sales volume of each period in a future period can be accurately predicted; and meanwhile, a series of reasonable decision schemes capable of changing the drug sales situation can be made based on the prediction conclusion in the prior art, so that the total drug sales amount in a future period of time is maximum.

Description

technical field [0001] The invention relates to the field of pharmacy, in particular to a BP neural network-based drug sales forecast and decision-making method and system. Background technique [0002] With the rapid development of the computer industry, data mining technology based on big data and intelligent algorithms is in a period of vigorous development. There are many different types of computing data in big data, and there are certain correlations between various types of data. Data mining technology is a technology that rationally integrates and analyzes all kinds of data in big data to obtain the inherent relationship and laws of the data itself. Data mining needs to take into account both computational efficiency and computational accuracy, which relies on intelligent optimization algorithms with powerful search capabilities. With the in-depth advancement of intelligent algorithm research, many intelligent optimization algorithms have been applied to data minin...

AI Technical Summary

Problems solved by technology

For drug sales decision-making, there are two shortcomings in the existing decision-making methods: 1. The decision-making has nothing to do with sales forecast or is based on completely inaccurate sales forecast results; 2. The decision-making method adopted is too subjective

Although such a decision-making method has a certain scientific and reasonable basis, due to the existence of the above-mentioned shortcomings, the decision-making plan will not match the actual situation, resulting in the implementation of the decision-making plan, and the sales results will greatly deviate from the ideal sales expectations.

Although it may help near-term drug sales, the effect is not obvious

Images