Influenza prediction system and device for optimizing LSTM (Long Short Term Memory) and LightGBM parameters and storage medium

Abstract

The invention discloses an influenza prediction system for optimizing LSTM (Long Short Term Memory) and LightGBM parameters. The system comprises an LSTM module, a LightGBM module and a prediction module; the LSTM module is configured to calculate a predicted value n of the number of people infected with influenza in an aggregated group by using LSTM based on historical factor data and external factor data; the LightGBM module is configured to calculate the influenza infection probability of each person in the aggregation group by adopting the LightGBM based on health condition data and surrounding environment data, and an infection probability sequence is obtained by ranking from large to small; the prediction module is configured to select the top n individuals with the highest probability of infecting influenza as high-risk population. The system combines the influenza outbreak trend of the aggregated group with the influenza infection probability of each individual in the group, so that the susceptible individuals in the aggregated group can be accurately predicted; meanwhile, the system adjusts parameters of an LSTM model and a LightGBM algorithm so that the accuracy of a prediction result can be further improved.

Description

technical field [0001] The invention belongs to the fields of artificial intelligence, data statistics, medical informatization, etc., and relates to a multivariable LSTM and LightGBM parameter adjustment system, storage medium and device for pre-influenza. Background technique [0002] At present, some progress has been made in the prediction of influenza trends. For example, the prediction of influenza outbreak trends mainly uses linear regression models, time series models, etc., and these prediction methods use historical influenza population data to train the models. The external characteristics such as environmental factors and weather factors that have a certain degree of influence on the percentage of influenza cases, but it is only a prediction of the outbreak trend and cannot accurately identify susceptible persons in a certain range of people. [0003] In the prior art, a technical solution for disease prediction through Long Short Term Memory networks (LSTM) has ...

AI Technical Summary

Problems solved by technology

[0006] In order to solve the problem that the LSTM model is large and the calculation speed is slow and difficult to converge when processing data with many dimensions, which leads to the inability to take into account many types of factors, and the problem that lightGBM is easy to overfit and cause inaccurate calculations, the present invention provides the following technical solutions:

Images