Multichannel LSTM neural network influenza epidemic situation prediction method based on attention mechanism

A neural network and prediction method technology, applied in the field of influenza epidemic prediction, can solve the problems of low prediction accuracy of influenza epidemic prediction technology, achieve accurate and effective real-time prediction, solve the problem of influenza prediction, and improve the effect of accuracy

Inactive Publication Date: 2019-08-02
DONGGUAN UNIV OF TECH
View PDF4 Cites 24 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The present invention provides a multi-channel LSTM neural network influenza epidemic prediction method based on the attention mechanism in order to solve the problem of low prediction accuracy of the existing influenza epidemic prediction technology

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Multichannel LSTM neural network influenza epidemic situation prediction method based on attention mechanism
  • Multichannel LSTM neural network influenza epidemic situation prediction method based on attention mechanism
  • Multichannel LSTM neural network influenza epidemic situation prediction method based on attention mechanism

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0024] Specific implementation mode one: combine figure 1 The present embodiment is described, and the multi-channel LSTM neural network influenza epidemic prediction method based on the attention mechanism that the present embodiment provides specifically includes the following steps:

[0025] Step 1. Preprocessing and normalizing the data in the data set; then using model-based ranking for feature selection, and dividing the selected data into weather-related data and influenza epidemic-related data Two categories, generate a training set;

[0026] Step 2. Establish a multi-channel LSTM neural network model including attention mechanism (Attention); here the model is named Att-MCLSTM, namely Attention-based multi-channel LSTM (multi-channel LSTM neural network based on attention mechanism). The multi-channel LSTM neural network model based on the attention mechanism is different from the traditional recurrent neural network RNN ​​(Recurrent Neural Network), and the LSTM neu...

specific Embodiment approach 2

[0032] Embodiment 2: The difference between this embodiment and Embodiment 1 is that the standardization described in Step 1 specifically includes the following processes:

[0033] Perform Min-Max normalization on the preprocessed data (also known as dispersion standardization):

[0034]

[0035] Among them, x is the value in the preprocessed data, x min is the minimum value in the preprocessed data, x max is the maximum value in the preprocessed data, and y is the value of x after Min-Max normalization processing; after data normalization, the data values ​​will be scaled between 0 and 1.

[0036] Other steps and parameters are the same as those in the first embodiment.

specific Embodiment approach 3

[0037] Specific implementation mode three: the difference between this implementation mode and specific implementation mode two is that, as figure 2 As shown, the LSTM memory unit in the multi-channel LSTM neural network model described in step 2 includes an input gate 1, an output gate 2, a forgetting gate 3 and a self-circulating neuron 4; the structure gate structure of the LSTM memory unit controls the data in the LSTM The transmission in the memory unit includes the data transmission between different units and the data transmission within the unit. The input gate 1 controls the state update process of the unit, the output gate 2 controls whether the output sequence of the unit will change the memory state of other units, and the forgetting gate 3 can selectively retain or forget the previous state.

[0038] LSTM memory cells can be represented by the following equations:

[0039]

[0040] Among them, σ(·) is a logistic sigmoid function (mapping variables between 0 a...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention provides a multichannel LSTM (Long-short term memory) neural network influenza epidemic situation prediction method based on an attention mechanism, and belongs to the technical field ofepidemic disease monitoring. The multichannel LSTM neural network influenza epidemic situation prediction method comprises the following steps: firstly, carrying out preprocessing, normalization andfeature selection on data in a data set, dividing the selected data into weather-related data and influenza epidemic situation-related data, and generating a training set; then establishing a multichannel LSTM neural network model comprising an attention mechanism; inputting training set data into the model for training, and performing MAPE (mean absolute percentage) evaluation to obtain a trainedmultichannel LSTM neural network model; processing the test data to obtain a test set; inputting the test set data into the trained LSTM neural network model for testing; and finally, performing inverse standardization processing on a test output result to obtain an influenza epidemic situation prediction value. According to the multichannel LSTM neural network influenza epidemic situation prediction method based on an attention mechanism, the problem of low prediction accuracy of the existing influenza epidemic situation prediction technology is solved. The multichannel LSTM neural network influenza epidemic situation prediction method based on an attention mechanism can be used for influenza prediction of different regions.

Description

technical field [0001] The invention relates to a method for predicting influenza epidemic situation and belongs to the technical field of epidemic monitoring. Background technique [0002] Influenza is an acute respiratory infection caused by influenza virus. After the patient is infected, it is very likely that the primary disease will be aggravated, causing secondary bacterial pneumonia and chronic cardiopulmonary disease. The outbreak of influenza is seasonal, which may cause social panic and have a great impact on human health and social stability (D.N.T.How, C.K.Loo, and K.S.M. Sahari. Behavior recognition for humanoid robots using long short-termmemory. International Journal of Advanced Robotic Systems, 13(6):1729881416663369, 2016.). For example, the H1N1 influenza outbreak in 2009 caused 151,700 to 575,400 deaths worldwide in the first year of the outbreak (S. Yang, M. Santillana, and S. C. Kou. Accurate estimation of influenza epidemics using google search data v...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G16H50/80G06N3/04
CPCG16H50/80G06N3/045
Inventor 郝建业侯韩旭马钰付博峰
Owner DONGGUAN UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap