Human body health index predicting algorithm based on genetic algorithm and BP neural network

Abstract

The invention discloses a human body health index predicting algorithm based on a genetic algorithm and a BP neural network. The algorithm comprises the following steps of (1), acquiring sign data andperforming genetic coding for forming initial population, and successively calculating individual adaptability, selecting operator, crossover operator and variation operator of the initial population; (2), setting a highest heredity generation number i of a fusion layer to 100, and inputting the population after respective calculation of the individual adaptability, selecting operator, crossoveroperator and variation operator into the fusion layer respectively; and (3), inputting the population with the fusion layer which satisfies an iteration requirement into a BP neural network, and realizing human body health index predicting through training and learning of the BP neural network. According to the predicting algorithm, the acquired initial data are processed by means of the genetic algorithm, an optimal solution of the data can be realized; the optimal solution is input into the BP neural network for regularizing the data which are input into the BP neural network, improving weight precision, training efficiency, network performance and network approaching capability in the BP neural network.

Description

technical field [0001] The invention relates to the technical field of human health prediction, in particular to a human health index prediction algorithm based on a genetic algorithm and a BP neural network. Background technique [0002] With the improvement of living standards and the development of science and technology, people pay more and more attention to their health, and the application of wearable and portable smart sign data detection devices emerges. At present, most intelligent sign data detection devices can only realize single data detection, such as electronic thermometers, portable blood pressure meters, etc., and the analysis of data is limited to the number of times collected and the current single comparison, and does not have comprehensive analysis capabilities. There are also some specific large-scale data analysis capabilities, but large amounts of data need to be collected for a long time, which is not practical. [0003] The BP neural network uses t...

AI Technical Summary

Problems solved by technology

The invention patent with the application number of 201610543416.7 discloses a method for identifying human health status based on BP neural network. This method directly inputs into the BP neural network by measuring the sign data (age, gender, lung capacity, weight, etc.) The network conducts training and learning to predict the state of human health. Although the prediction of human health can also be realized, the initial data volume is huge and there are differences in the number, which makes the training efficiency low, resulting in a decline in network performance, which directly affects the approximation ability of the network. Thus affecting the prediction results of the final output human health indicators

Images