WO2021139336A1

WO2021139336A1 - Epidemic prevention and control effect prediction method and apparatus, and server and storage medium

Info

Publication number: WO2021139336A1
Application number: PCT/CN2020/124702
Authority: WO
Inventors: 郭建影
Original assignee: 平安科技（深圳）有限公司
Priority date: 2020-09-28
Filing date: 2020-10-29
Publication date: 2021-07-15
Also published as: CN111933300B; CN111933300A

Abstract

Disclosed are an epidemic prevention and control effect prediction method and apparatus, and a server and a storage medium, wherein same are applied to the field of medical technology. The method comprises: compiling statistics on new media data and epidemic data of each date in a first preset date range; calling an epidemic prevention and control effect prediction model, and predicting an epidemic prevention and control effect according to the new media data and the epidemic data to obtain a prediction result, wherein the prediction result comprises negative news data and epidemic infection data of each date in a second preset date range; and sending the prediction result to a terminal device for display. By using the method, the epidemic prevention and control effect can be predicted on the basis of new media information, so as to carry out epidemic prevention and control. The method relates to blockchain technology, for example, a prediction result can be written into a blockchain.

Description

Epidemic prevention and control effect prediction method, device, server and storage medium

This application claims the priority of a Chinese patent application filed with the Chinese Patent Office on September 28, 2020 with the application number 202011043912.9 and the invention title "Prediction method, device, server and storage medium for epidemic prevention and control effect", all of which are approved The reference is incorporated in this application.

Technical field

This application relates to the field of artificial intelligence technology, and in particular to a method, device, server and storage medium for predicting the effect of epidemic prevention and control.

Background technique

The outbreak and spread of the epidemic will have a serious impact on the economy and people's lives in various places. Recently, COVID-19 has caused a worldwide outbreak, causing great loss of life and economic losses. People can obtain daily real-time epidemic data through the Internet to keep abreast of the latest developments of the epidemic. With the development of new media, people can not only serve as publishers of epidemic-related news, but also as viewers of epidemic-related news. As a medium for the spread of epidemic-related content, new media affects people's lives.

The inventor realizes that the existing products in the industry mainly predict the development of the epidemic based on the epidemic data itself, but do not predict the effect of epidemic prevention and control based on new media information. The same pre-epidemic development trend will result in completely different development directions of the epidemic due to different epidemic prevention and control progress information, rumors, and expert comments released on new media, and will also have different impacts on the effectiveness of epidemic prevention and control. Therefore, how to predict the effect of epidemic prevention and control based on new media information and use it for epidemic prevention and control has become an urgent problem to be solved.

technical problem

The embodiments of the present application provide a method, device, server, and storage medium for predicting an epidemic prevention and control effect, which can predict an epidemic prevention and control effect based on new media information for use in epidemic prevention and control.

Technical solutions

In the first aspect, the embodiment of the present application provides a method for predicting the epidemic prevention and control effect, which includes: counting new media data and epidemic data on each date within a first preset date range, the new media data including negative news data and others New media data, the epidemic data including epidemic infection data; call the epidemic prevention and control effect prediction model, and predict the epidemic prevention and control effect based on the new media data and the epidemic data, and obtain the prediction result, the prediction result including Negative news data and epidemic infection data on each date within a second preset date range, where the second preset date range is after the first preset date range; the prediction result is sent to the terminal device so that the terminal device Show the predicted results.

In the second aspect, an embodiment of the present application provides an epidemic prevention and control effect prediction device, including: a statistics module, configured to count new media data and epidemic data on each date within a first preset date range, and the new media data includes Negative news data and other new media data, the epidemic data including epidemic infection data; a prediction module, used to call the epidemic prevention and control effect prediction model, and predict the epidemic prevention and control effect based on the new media data and the epidemic data , Obtaining a prediction result, the prediction result including negative news data and epidemic infection data on each date within a second preset date range, the second preset date range being after the first preset date range; the output module, It is used to send the prediction result to the terminal device so that the terminal device can display the prediction result.

In a third aspect, an embodiment of the present application provides a server, including a processor, an output device, and a memory, the processor, the output device, and the memory are connected to each other, wherein the memory is used to store a computer program, The computer program includes program instructions, and the processor is configured to call the program instructions to execute the following method: statistics new media data and epidemic data of each date within a first preset date range, the new media data includes Negative news data and other new media data, the epidemic data including epidemic infection data; call the epidemic prevention and control effect prediction model, and predict the epidemic prevention and control effect based on the new media data and the epidemic data, and obtain the prediction result, The prediction result includes negative news data and epidemic infection data on each date within a second preset date range, the second preset date range is after the first preset date range; the prediction result is sent to the terminal Device so that the terminal device can display the prediction result.

In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the following method: counting the first preset date range The new media data and epidemic data of each date within the country, the new media data includes negative news data and other new media data, and the epidemic data includes epidemic infection data; the epidemic prevention and control effect prediction model is invoked and based on the new media data The epidemic prevention and control effect is predicted with the epidemic data, and the forecast result is obtained. The forecast result includes the negative news data and the epidemic infection data on each date within the second preset date range. After the first preset date range; sending the prediction result to a terminal device, so that the terminal device can display the prediction result.

Beneficial effect

In summary, the embodiments of the present application can predict the epidemic prevention and control effect based on new media information for epidemic prevention and control.

Description of the drawings

FIG. 1 is a schematic flowchart of a method for predicting an epidemic prevention and control effect provided by an embodiment of the present application.

FIG. 2 is a schematic flowchart of another method for predicting an epidemic prevention and control effect provided by an embodiment of the present application.

FIG. 3 is a schematic diagram of a network architecture of an epidemic prevention and control effect prediction system provided by an embodiment of the present application.

FIG. 4 is a schematic structural diagram of a device for predicting an epidemic prevention and control effect provided by an embodiment of the present application.

Fig. 5 is a schematic structural diagram of a server provided by an embodiment of the present application.

Embodiments of the present invention

The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application.

The technical solution of this application can be applied to the fields of artificial intelligence, digital medicine, smart city, blockchain and/or big data technology to predict the effect of epidemic prevention and control and realize smart medical treatment. Optionally, the data involved in this application, such as epidemic data and/or prediction results, can be stored in a database, or can be stored in a blockchain, or can be stored in other ways, and this application is not limited.

Please refer to FIG. 1, which is a schematic flowchart of a method for predicting an epidemic prevention and control effect provided by an embodiment of this application. This method can be applied to the server. The server can be a server or a server cluster. Specifically, the method may include the following steps.

S101. Count new media data and epidemic data on each date within the first preset date range, where the new media data includes negative news data and other new media data, and the epidemic data includes epidemic infection data.

Among them, negative news data can include at least one of the following: the number of negative news during the epidemic and the growth rate of negative news during the epidemic. The number of negative news during the epidemic period can include the number of daily negative news and the total number of negative news. The growth rate of negative news during the epidemic period includes the average daily growth rate of negative news.

Among them, other new media data includes at least one of the following: the release method of epidemic-related data, the audience category of the new media platform corresponding to each release method, the activity data of the new media platform, and the epidemic situation released for the new media platform The labels of the relevant data and the number of news corresponding to each category of the epidemic-related data released by the new media platform. The publishing method can be public account, short video, for example. The audience category can be divided according to age and education level, and the audience category can also be divided according to other forms, and there is no restriction here. For example, the activity data may include at least one of the following: daily activity and hourly activity. In one embodiment, the activity data may also include the number of views, the number of likes, or the number of comments on the target data in the epidemic-related data. For example, labels can be divided into positive labels (which can be further subdivided) and negative labels (which can be further subdivided), or can be divided according to the content of epidemic prevention and transmission, and labels can also be divided according to other forms, and there is no restriction here.

Among them, the epidemic data may include epidemic infection data, which includes at least one of the following: the number of people infected by the epidemic and the growth rate of the number of people infected by the epidemic. The number of people infected by the epidemic can be the number of people infected by the epidemic in the target area. The target area may be the area to be predicted. The growth rate of the number of people infected by the epidemic may be the growth rate of the number of people infected by the epidemic in the target area. The infection data of the epidemic may include at least one of the following: daily increase in the number of infections and the total number of infections, and the increase in the number of infections in the epidemic may be the average daily increase in the number of infections. In one embodiment, the epidemic data may also include the number of deaths from the epidemic, and the death toll of the epidemic may include at least one of the following: daily increase in deaths and total deaths.

S102. Call the epidemic prevention and control effect prediction model, and predict the epidemic prevention and control effect based on the new media data and the epidemic data to obtain a prediction result, the prediction result including the negative of each date within the second preset date range For news data and epidemic infection data, the second preset date range is after the first preset date range.

In the embodiment of the present application, the server may call the epidemic prevention and control effect prediction model, and predict the epidemic prevention and control effect based on the new media data and the epidemic data, and obtain the prediction result. The embodiments of this application can aim at the epidemic prevention situation of large-scale regional infectious diseases that occur, and combine the processing and dissemination of information related to epidemic prevention and control by various new media, and dig out the details of the operation of the new media industry to improve the effectiveness of the prevention and control of infectious diseases in the region. From the perspective of media operations, it provides more effective assistance for the government’s epidemic prevention decision-making and the health department’s epidemic prevention work.

In one embodiment, the epidemic prevention and control effect prediction model may be a pre-trained first machine learning model. The first machine learning model may be a first deep learning model, and the first machine learning model may be constructed based on a deep neural network. Or, the epidemic prevention and control effect prediction model is a pre-trained second machine learning model. The second machine learning model may be a second deep learning model, and the second machine learning model may be constructed based on a convolutional neural network, a deep neural network, and a long and short-term memory network.

In one embodiment, when the epidemic prevention and control effect is the pre-trained first machine learning model, the pre-trained first machine learning model can be obtained in the following manner: the server counts the new data of each date within the third preset date range Media data and epidemic data; the server uses the new media data and epidemic data of each date within the third preset date range as the input data of the original first machine learning model, and the original first machine learning model is based on the new media data And the epidemic data output negative news data and epidemic infection data for each date within the fourth preset date range; the server obtains the real negative news data and epidemic infection data for each date within the fourth preset date range, and according to the output first 4. The negative news data and epidemic infection data of each date within the preset date range, as well as the real negative news data and epidemic infection data of each date within the fourth preset date range, construct a loss function; the server uses the loss function to train the original The first machine learning model obtains the pre-trained first machine learning model. The fourth preset date range is after the third preset date range.

In one embodiment, when the epidemic prevention and control effect prediction model is a pre-trained second machine learning model, the pre-trained second machine learning model can be obtained in the following manner: the server counts the dates within the third preset date range New media data and epidemic data; the server uses the new media data and epidemic data of each date within the third preset date range as the input data of the original second machine learning model, and the original second machine learning model is based on the new The media data and the epidemic data output negative news data and epidemic infection data for each date within the fourth preset date range; the server obtains the real negative news data and epidemic infection data for each date within the fourth preset date range, and outputs according to the output The negative news data and epidemic infection data of each date within the fourth preset date range, and the real negative news data and epidemic infection data of each date within the fourth preset date range are used to construct a loss function; the server uses the loss function to train the The original second machine learning model obtains the pre-trained second machine learning model.

S103. Send the prediction result to a terminal device, so that the terminal device can display the prediction result.

In the embodiment of the present application, the server may output the prediction result through the terminal device.

In one embodiment, the server may use an attribution analysis model to filter out new media data and epidemic data of each date within the first preset date range from Internet data.

In one embodiment, the server may obtain new media data and epidemic data of each date within the target time range before the first preset time range, and negative news data and data of each date within the next time range of the target time range. Epidemic infection data; the server uses the attribution analysis model based on the new media data and epidemic data of each date in the time range before the first preset time range, and the negative news data and epidemic data of each date in the next time range of the target time range Infection data, determine the target evaluation index for epidemic prevention and control effect prediction from the preset multiple evaluation indexes; the server selects new media with the target evaluation index on each date within the first preset date range from Internet data Data and epidemic data. Among them, the attribution analysis method of the attribution analysis model can be any of the following: g-formula method, inverse probability weighting method, and probability-score method.

It can be seen that, in the embodiment shown in FIG. 1, the server can count the new media data and epidemic data of each date within the first preset date range, and call the epidemic prevention and control effect prediction model, based on the new media data and the epidemic data The effect of epidemic prevention and control is predicted, and the prediction result is obtained and sent to the terminal device for display, so that the epidemic prevention and control effect can be predicted based on new media information for epidemic prevention and control.

Please refer to FIG. 2, which is a schematic flowchart of another method for predicting an epidemic prevention and control effect provided by an embodiment of this application. This method can be applied to the server. The server can be a server or a server cluster. Specifically, the method may include the following steps.

S201. Count new media data and epidemic data on each date within the first preset date range, where the new media data includes negative news data and other new media data, and the epidemic data includes epidemic infection data.

Among them, step S201 may refer to step S101 in the embodiment of FIG. 1, and details are not described in this embodiment of the present application.

S202: Extract spatial features according to the new media data and the epidemic data through the convolutional neural network in the second machine learning model that is pre-trained.

In the embodiment of this application, the server can input the new media data and the epidemic data into the convolutional neural network in the second machine learning model pre-trained, and the convolutional neural network outputs space according to the new media data and the epidemic data feature. In one embodiment, the server may extract the new media information data and epidemic data of each date within the first preset time range in chronological order using days as the time unit to form a feature vector with a total length of N; the server will The feature vector of each time unit is used as the input of the convolutional neural network, and the convolutional neural network outputs spatial features according to the feature vector of each time unit.

S203: Extract time series features according to the new media data and the epidemic data through the deep neural network and the long-short-term memory network in the pre-trained second machine learning model.

In this embodiment of the application, the server uses the deep neural network and the long-short-term memory network in the pre-trained second machine learning model to extract time series features from the new media data and the epidemic data. The process may be: the server applies the new media The data and the epidemic data are input to the first layer of deep neural network in the pre-trained second machine learning model, the output of the first layer of deep neural network is obtained, and the output of the first layer of deep neural network is used as the pre-training The input of the second layer of long and short-term memory network in the second machine learning model of, the output of the second layer of long and short-term memory network is obtained; the server uses the output of the second layer of long and short-term memory network as the pre-trained second machine The output of the third-layer long-term and short-term memory network in the learning model is output by the third-layer long- and short-term memory network for timing characteristics. In one embodiment, the long short-term memory network may be a double-layer long short-term memory network.

In the embodiment of this application, the server may sequentially use the new media data and epidemic data of each date in the first preset date range as the input of the first layer of deep neural network in chronological order to obtain the output of the first layer of deep neural network . The server can use the output of the first layer of deep neural network as the input of the second layer of two-way long and short-term memory network, and use days as the time unit, and correspond to the new media information data and epidemic data at two adjacent times to the second layer. The output of the two-way long and short-term memory network, as the new media data and epidemic data at two adjacent time points, corresponds to the input of the third layer of the two-way long and short-term memory network, so that the third layer of bidirectional long and short-term output timing characteristics.

S204. Perform fusion processing on the spatial feature and the time sequence feature to obtain a temporal feature.

S205. Predict the negative news growth trend and the epidemic trend of infectious diseases according to the temporal and spatial characteristics, and obtain a prediction result, where the prediction result includes negative news data and epidemic infection data on each date within a second preset date range, and the first 2. The preset date range is after the first preset date range.

In step S204-step S205, the server may perform fusion processing on the spatial feature and the temporal feature to obtain the temporal and spatial features, and predict the growth trend of negative news and the trend of infectious diseases based on the temporal and spatial features, and obtain the prediction result.

In one embodiment, the server predicts the growth trend of negative news and the epidemic trend of infectious diseases according to the spatiotemporal characteristics, and obtains the prediction result. Specifically, the server uses the spatiotemporal characteristics as the fourth layer in the pre-trained second machine learning model. The input of the deep neural network, the output of the fourth layer of deep neural network is obtained, and the output of the fourth layer of deep neural network is input to the output layer of the pre-trained second machine learning model, and the output layer predicts result.

In one embodiment, the server predicts the growth trend of negative news and the epidemic trend of infectious diseases according to the spatiotemporal characteristics, and the process of obtaining the prediction results may be: the server uses the spatiotemporal characteristics as the output in the pre-trained second machine learning model The input of the layer, and the prediction result is output by the output layer. The structure of the second machine learning model here is simpler and easier to implement than the former.

S206. Send the prediction result to a terminal device, so that the terminal device can display the prediction result.

Wherein, step S206 may refer to step S103 in the embodiment of FIG. 1, and details are not described in the embodiment of the present application.

In one embodiment, when the epidemic prevention and control effect prediction model is a pre-trained second machine learning model, the method of acquiring the pre-trained second machine learning model may be specifically as follows: the server counts each item within the third preset date range Date of new media data and epidemic data; the server uses the new media data and epidemic data of each date within the third preset date range as the input data of the original second machine learning model, and passes the data in the original second machine learning model. The convolutional neural network extracts target spatial features based on the new media data and epidemic data of each date within the third preset date range, and uses the deep neural network and long-short-term memory network in the pre-trained second machine learning model, According to the new media data and epidemic data of each date within the third preset date range, the target time series feature is extracted; the server performs fusion processing on the target space feature and the target time series feature to obtain the target time and space feature, and compare the target time and space feature according to the target time and time feature. The negative news growth trend and the epidemic trend of infectious diseases are predicted, and the negative news data and epidemic infection data of each date within the fourth preset date range are obtained; the server obtains the real negative news data and epidemic situation of each date within the fourth preset date range Infection data, and construct a loss function based on the output negative news data and epidemic infection data of each date within the fourth preset date range, and the real negative news data and epidemic infection data of each date within the fourth preset date range; Use the loss function to train the original second machine learning model to obtain a pre-trained second machine learning model.

It can be seen that in the embodiment shown in FIG. 2, the server can extract spatial features based on the new media data and epidemic data of each date within the first preset time range through the convolutional neural network in the second machine learning model pre-trained ; The server uses the deep neural network and the long-short-term memory network in the pre-trained second machine learning model to extract time series features based on the new media data and the epidemic data; the server merges the spatial features and the time series features to obtain space-time According to the temporal and spatial characteristics, the negative news growth trend and the epidemic trend of infectious diseases are predicted, and the prediction results can be obtained. The pre-trained second machine learning model can be used to predict the epidemic prevention and control effect based on new media information. Epidemic prevention and control.

This application can be applied to the field of medical technology. This application relates to blockchain technology. For example, prediction results can be written into the blockchain or encrypted data of the prediction results can be written into the blockchain.

Please refer to FIG. 3, which is a schematic diagram of a network architecture of an epidemic prevention and control effect prediction system provided by an embodiment of this application. The epidemic prevention and control effect prediction system may include a server 10 and a terminal device 20. Wherein, the server 10 can perform step S101 and step S102 to predict the epidemic prevention and control effect based on new media data and epidemic data to obtain a prediction result, and can perform step S103 to display the prediction result through the terminal device 30.

Please refer to FIG. 4, which is a schematic structural diagram of an epidemic prevention and control effect prediction device provided by an embodiment of this application. The device can be applied to a server. Specifically, the device may include the following modules.

The statistics module 401 is configured to count new media data and epidemic data on each date within the first preset date range. The new media data includes negative news data and other new media data, and the epidemic data includes epidemic infection data.

The prediction module 402 is used to call the epidemic prevention and control effect prediction model, and predict the epidemic prevention and control effect based on the new media data and the epidemic data to obtain a prediction result, the prediction result including the second preset date range For the negative news data and epidemic infection data of each date, the second preset date range is after the first preset date range.

The output module 403 is configured to send the prediction result to the terminal device so that the terminal device can display the prediction result.

In an optional implementation manner, the negative news data includes at least one of the following: the number of negative news during the epidemic and the growth rate of negative news during the epidemic; the other new media data includes at least one of the following: The release method, the audience category of the new media platform corresponding to each release method, the activity data of the new media platform, the labels for the epidemic-related data released by the new media platform, and the epidemic situation released by the new media platform The relevant data is the number of news corresponding to each type of label; the epidemic data includes epidemic infection data, and the epidemic infection data includes at least one of the following: the number of people infected by the epidemic and the growth rate of the number of people infected by the epidemic.

In an optional embodiment, the epidemic prevention and control effect prediction model is a pre-trained first machine learning model, and the first machine learning model is constructed based on a deep neural network; or, the epidemic prevention and control The effect prediction model is a pre-trained second machine learning model, and the second machine learning model is constructed based on a convolutional neural network, a deep neural network, and a long and short-term memory network.

In an optional implementation, when the epidemic prevention and control effect prediction model is a pre-trained second machine learning model, the prediction module 402 invokes the epidemic prevention and control effect prediction model, and based on the new media data and information The epidemic data predicts the epidemic prevention and control effect, and obtains the prediction result, specifically through the convolutional neural network in the pre-trained second machine learning model, extracting spatial features based on the new media data and the epidemic data; The deep neural network and the long short-term memory network in the pre-trained second machine learning model extract time series features according to the new media data and the epidemic data; perform fusion processing on the spatial features and the time series features to obtain Temporal and spatial characteristics; predict the growth trend of negative news and the epidemic trend of infectious diseases according to the temporal and spatial characteristics, and obtain the prediction result.

In an optional implementation manner, the prediction module 402 extracts time series features from the new media data and the epidemic data through the deep neural network and the long-short-term memory network in the pre-trained second machine learning model, Specifically, the new media data and the epidemic data are input into the first layer of deep neural network in the pre-trained second machine learning model to obtain the output of the first layer of deep neural network; The output of the layered deep neural network is used as the input of the second layer of long and short-term memory network in the pre-trained second machine learning model to obtain the output of the second layer of long- and short-term memory network; The output of the memory network is used as the output of the third-layer long- and short-term memory network in the pre-trained second machine learning model, and the third-layer long- and short-term memory network outputs timing characteristics.

In an optional implementation manner, the prediction module 402 predicts the negative news growth trend and the epidemic trend of infectious diseases according to the spatio-temporal characteristics, and obtains the prediction result, specifically using the spatio-temporal characteristics as the second pre-training. Input of the fourth layer of deep neural network in the machine learning model to obtain the output of the fourth layer of deep neural network; input the output of the fourth layer of deep neural network into the pre-trained second machine learning model The output layer of which outputs the prediction result.

In an optional implementation manner, the prediction module 402 predicts the growth trend of negative news and the epidemic trend of infectious diseases according to the temporal and spatial characteristics, and obtains the prediction result, specifically using the temporal and spatial characteristics as the second pre-training. The output layer in the machine learning model is an input, and the output layer outputs a prediction result.

It can be seen that in the embodiment shown in FIG. 4, the epidemic prevention and control effect prediction device can count the new media data and epidemic data of each date within the first preset date range, and call the epidemic prevention and control effect prediction model, and according to the new media The data and the epidemic data predict the epidemic prevention and control effect, and obtain the prediction result to send the forecast result to the terminal device for display. The epidemic prevention and control effect can be predicted based on the new media information for epidemic prevention and control.

Please refer to FIG. 5, which is a schematic structural diagram of a server provided in an embodiment of this application. The server described in this embodiment may include: one or more processors 1000, one or more input devices 2000, one or more output devices 3000, and a memory 4000. The processor 1000 and the memory 4000 may be connected by a bus. The input device 2000 included in the server is an optional device, that is, the server may only include one or more processors 1000, one or more output devices 3000, and a memory 4000. In an embodiment, the input device 2000 and the output device 3000 may be standard wired or wireless communication interfaces.

The processor 1000 may be a central processing unit (Central Processing Unit, CPU), and the processor may also be other general-purpose processors, digital signal processors (digital signal processors, DSP), and application-specific integrated circuits (Application Specific Integrated Circuits). Specific Integrated Circuit, ASIC), ready-made programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor or the like.

The memory 4000 can be a high-speed RAM memory or a non-volatile memory (non-volatile memory), such as disk storage. Wherein, the memory 4000 is used to store a computer program, and the computer program includes program instructions. The processor 1000 is configured to call the program instructions to perform the following steps: to count new media data and epidemic data on each date within the first preset date range, where the new media data includes negative news data and other new media data, The epidemic data includes epidemic infection data; the epidemic prevention and control effect prediction model is called, and the epidemic prevention and control effect is predicted based on the new media data and the epidemic data, and the prediction result is obtained, and the prediction result includes the second preset The negative news data and epidemic infection data of each date within the date range, the second preset date range is after the first preset date range; the prediction result is sent to the terminal device so that the terminal device can display the prediction result.

In one embodiment, the negative news data includes at least one of the following: the number of negative news during the epidemic and the growth rate of negative news during the epidemic.

The other new media data includes at least one of the following: the release method of epidemic-related data, the audience category of the new media platform corresponding to each release method, the activity data of the new media platform, and the release for the new media platform The number of news related to each type of label in the labels labeled with the epidemic-related data of the new media platform and the epidemic-related data released by the new media platform.

The epidemic data includes epidemic infection data, and the epidemic infection data includes at least one of the following: the number of people infected by the epidemic and the growth rate of the number of people infected by the epidemic.

In one embodiment, the epidemic prevention and control effect prediction model is a pre-trained first machine learning model, and the first machine learning model is constructed based on a deep neural network; or, the epidemic prevention and control effect prediction model is A pre-trained second machine learning model, the second machine learning model is constructed based on a convolutional neural network, a deep neural network, and a long and short-term memory network.

In one embodiment, when the epidemic prevention and control effect prediction model is a pre-trained second machine learning model, the epidemic prevention and control effect prediction model is invoked, and the epidemic prevention and control effect prediction model is used based on the new media data and the epidemic data. When the prediction result is obtained, the processor 1000 is configured to call the program instructions to perform the following steps: through the pre-trained convolutional neural network in the second machine learning model, according to the new media data and Extracting spatial features from the epidemic data; extracting time series features based on the new media data and the epidemic data through the deep neural network and the long- and short-term memory network in the pre-trained second machine learning model; extracting the temporal features from the new media data and the epidemic data; and comparing the spatial features Performing fusion processing with the time sequence feature to obtain the spatio-temporal feature; predict the growth trend of negative news and the epidemic trend of infectious diseases according to the spatio-temporal feature, and obtain the prediction result.

In one embodiment, when using the deep neural network and the long short-term memory network in the pre-trained second machine learning model to extract time series features from the new media data and the epidemic data, the processor 1000 is configured Used to call the program instructions to perform the following steps: input the new media data and the epidemic data into the first-layer deep neural network in the pre-trained second machine learning model to obtain the first-layer depth The output of the neural network; the output of the first layer of deep neural network is used as the input of the second layer of long and short-term memory network in the pre-trained second machine learning model to obtain the second layer of long- and short-term memory network Output; use the output of the second-layer long-short-term memory network as the output of the third-layer long-short-term memory network in the pre-trained second machine learning model, and the third-layer long- and short-term memory network outputs timing features .

In one embodiment, when the negative news growth trend and the epidemic trend of infectious diseases are predicted according to the temporal and spatial characteristics, and the prediction result is obtained, the processor 1000 is configured to call the program instructions to perform the following steps: The spatio-temporal features are used as the input of the fourth layer of deep neural network in the pre-trained second machine learning model to obtain the output of the fourth layer of deep neural network; the output of the fourth layer of deep neural network is input to all In the output layer of the pre-trained second machine learning model, the output layer outputs a prediction result.

In one embodiment, when the negative news growth trend and the epidemic trend of infectious diseases are predicted according to the temporal and spatial characteristics, and the prediction result is obtained, the processor 1000 is configured to call the program instructions to perform the following steps: The spatio-temporal feature is used as the input of the output layer in the pre-trained second machine learning model, and the output layer outputs the prediction result.

In specific implementation, the processor 1000, the input device 2000, and the output device 3000 described in the embodiment of the present application can perform the implementation described in the embodiment of FIG. 1 and the embodiment of FIG. 2, as well as the implementation described in the embodiment of the present application. The implementation method will not be repeated here.

The functional modules in each embodiment of the present application may be integrated into one processing module, or each module may exist alone physically, or two or more modules may be integrated into one module. The above-mentioned integrated modules can be implemented in the form of sampling hardware, and can also be implemented in the form of sampling software functional modules.

The embodiments of the present application also provide a computer-readable storage medium with a computer program stored on the computer-readable storage medium. When the computer program is executed by a processor, the steps of the method in the above-mentioned embodiment can be realized, or the computer program is processed. The function of each module/unit of the device in the above-mentioned embodiment is realized when the device is executed, which will not be repeated here. Optionally, the storage medium involved in this application, such as a computer-readable storage medium, may be non-volatile or volatile.

A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing relevant hardware through a computer program. The program can be stored in a computer readable storage medium, and the program can be stored in a computer readable storage medium. During execution, it may include the procedures of the above-mentioned method embodiments. Wherein, the computer-readable storage medium may be volatile or non-volatile. For example, the computer storage medium may be a magnetic disk, an optical disk, a read-only memory (Read-Only Memory, ROM), or a random access memory (Random Access Memory). Memory, RAM) etc. The computer-readable storage medium may mainly include a storage program area and a storage data area, where the storage program area may store an operating system, an application program required by at least one function, etc.; the storage data area may store information based on the blockchain node Use the created data, etc.

Among them, the blockchain referred to in this application is a new application mode of computer technology such as distributed data storage, point-to-point transmission, consensus mechanism, and encryption algorithm. Blockchain, essentially a decentralized database, is a series of data blocks associated with cryptographic methods. Each data block contains a batch of network transaction information for verification. The validity of the information (anti-counterfeiting) and generation of the next block. The blockchain can include the underlying platform of the blockchain, the platform product service layer, and the application service layer.

What is disclosed above is only a preferred embodiment of this application. Of course, it cannot be used to limit the scope of rights of this application. A person of ordinary skill in the art can understand all or part of the process of implementing the above-mentioned embodiments and follow the rights of this application. The equivalent changes required are still within the scope of this application.

Claims

A method for predicting the effect of epidemic prevention and control, including:

Collect statistics on new media data and epidemic data on each date within the first preset date range, where the new media data includes negative news data and other new media data, and the epidemic data includes epidemic infection data;

Call the epidemic prevention and control effect prediction model, and predict the epidemic prevention and control effect based on the new media data and the epidemic data, and obtain the prediction result. The prediction result includes the negative news data on each date within the second preset date range And epidemic infection data, the second preset date range is after the first preset date range;

The prediction result is sent to the terminal device so that the terminal device can display the prediction result.
The method of claim 1, wherein:

The negative news data includes at least one of the following: the number of negative news during the epidemic and the growth rate of negative news during the epidemic;

The other new media data includes at least one of the following: the release method of epidemic-related data, the audience category of the new media platform corresponding to each release method, the activity data of the new media platform, and the release for the new media platform The number of labels marked on the epidemic-related data of the new media platform, and the number of news items corresponding to each category of the epidemic-related data released by the new media platform;

The epidemic data includes epidemic infection data, and the epidemic infection data includes at least one of the following: the number of people infected by the epidemic and the growth rate of the number of people infected by the epidemic.
The method according to claim 1 or 2, wherein the epidemic prevention and control effect prediction model is a pre-trained first machine learning model, and the first machine learning model is constructed based on a deep neural network; or, the The epidemic prevention and control effect prediction model is a pre-trained second machine learning model, and the second machine learning model is constructed based on a convolutional neural network, a deep neural network, and a long and short-term memory network.
The method according to claim 2, wherein, when the epidemic prevention and control effect prediction model is a pre-trained second machine learning model, the epidemic prevention and control effect prediction model is invoked and based on the new media data and the The epidemic data is used to predict the effect of epidemic prevention and control, and the prediction results are obtained, including:

Extracting spatial features based on the new media data and the epidemic data through the convolutional neural network in the pre-trained second machine learning model;

Extracting time series features according to the new media data and the epidemic data through the deep neural network and the long-short-term memory network in the pre-trained second machine learning model;

Performing fusion processing on the spatial feature and the time sequence feature to obtain a temporal feature;

According to the temporal and spatial characteristics, the negative news growth trend and the epidemic trend of infectious diseases are predicted, and the prediction result is obtained.
The method according to claim 4, wherein the deep neural network and the long-short-term memory network in the pre-trained second machine learning model extract time series features according to the new media data and the epidemic data, include:

Input the new media data and the epidemic data into the first layer of deep neural network in the pre-trained second machine learning model to obtain the output of the first layer of deep neural network;

Using the output of the first-layer deep neural network as the input of the second-layer long-short-term memory network in the pre-trained second machine learning model to obtain the output of the second-layer long- and short-term memory network;

The output of the second-layer long- and short-term memory network is used as the output of the third-layer long- and short-term memory network in the pre-trained second machine learning model, and the third-layer long- and short-term memory network outputs timing characteristics.
The method according to claim 4, wherein the predicting the growth trend of negative news and the epidemic trend of infectious diseases according to the temporal and spatial characteristics to obtain the prediction result comprises:

Using the spatiotemporal features as the input of the fourth layer of deep neural network in the pre-trained second machine learning model to obtain the output of the fourth layer of deep neural network;

The output of the fourth layer of deep neural network is input to the output layer of the pre-trained second machine learning model, and the output layer outputs the prediction result.
The method according to claim 4, wherein the predicting the growth trend of negative news and the epidemic trend of infectious diseases according to the temporal and spatial characteristics to obtain the prediction result comprises:

The spatio-temporal feature is used as the input of the output layer in the pre-trained second machine learning model, and the output layer outputs a prediction result.
An epidemic prevention and control effect prediction device, including:

The statistics module is used to count new media data and epidemic data on each date within the first preset date range, the new media data includes negative news data and other new media data, and the epidemic data includes epidemic infection data;

The prediction module is used to call the epidemic prevention and control effect prediction model, and predict the epidemic prevention and control effect based on the new media data and the epidemic data, and obtain the prediction result. The prediction result includes each item within the second preset date range. Negative news data and epidemic infection data of the date, the second preset date range is after the first preset date range;

The output module is used to send the prediction result to the terminal device so that the terminal device can display the prediction result.
A server includes a processor, an output device, and a memory, the processor, the output device, and the memory are connected to each other, wherein the memory is used to store a computer program, and the computer program includes program instructions. The processor is configured to call the program instructions and execute the following methods:

Collect statistics on new media data and epidemic data on each date within the first preset date range, where the new media data includes negative news data and other new media data, and the epidemic data includes epidemic infection data;

Call the epidemic prevention and control effect prediction model, and predict the epidemic prevention and control effect based on the new media data and the epidemic data, and obtain the prediction result. The prediction result includes the negative news data on each date within the second preset date range And epidemic infection data, the second preset date range is after the first preset date range;

The prediction result is sent to the terminal device so that the terminal device can display the prediction result.
The server according to claim 9, wherein:

The negative news data includes at least one of the following: the number of negative news during the epidemic and the growth rate of negative news during the epidemic;

The other new media data includes at least one of the following: the release method of epidemic-related data, the audience category of the new media platform corresponding to each release method, the activity data of the new media platform, and the release for the new media platform The number of labels marked on the epidemic-related data of the new media platform, and the number of news items corresponding to each category of the epidemic-related data released by the new media platform;

The epidemic data includes epidemic infection data, and the epidemic infection data includes at least one of the following: the number of people infected by the epidemic and the growth rate of the number of people infected by the epidemic.
The server according to claim 9 or 10, wherein the epidemic prevention and control effect prediction model is a pre-trained first machine learning model, and the first machine learning model is constructed based on a deep neural network; or, the The epidemic prevention and control effect prediction model is a pre-trained second machine learning model, and the second machine learning model is constructed based on a convolutional neural network, a deep neural network, and a long and short-term memory network.
The server according to claim 10, wherein, when the epidemic prevention and control effect prediction model is a pre-trained second machine learning model, the epidemic prevention and control effect prediction model is invoked, and the epidemic prevention and control effect prediction model is invoked based on the new media data and all information The epidemic data is used to predict the effect of epidemic prevention and control. When the prediction result is obtained, the specific implementation is as follows:

Extracting spatial features based on the new media data and the epidemic data through the convolutional neural network in the pre-trained second machine learning model;

Extracting time series features according to the new media data and the epidemic data through the deep neural network and the long-short-term memory network in the pre-trained second machine learning model;

Performing fusion processing on the spatial feature and the time sequence feature to obtain a temporal feature;

According to the temporal and spatial characteristics, the negative news growth trend and the epidemic trend of infectious diseases are predicted, and the prediction result is obtained.
The server according to claim 12, wherein the deep neural network and the long-short-term memory network in the second pre-trained machine learning model are used to extract time series features based on the new media data and the epidemic data , The specific implementation:

Input the new media data and the epidemic data into the first layer of deep neural network in the pre-trained second machine learning model to obtain the output of the first layer of deep neural network;

Using the output of the first-layer deep neural network as the input of the second-layer long-short-term memory network in the pre-trained second machine learning model to obtain the output of the second-layer long- and short-term memory network;

The output of the second-layer long- and short-term memory network is used as the output of the third-layer long- and short-term memory network in the pre-trained second machine learning model, and the third-layer long- and short-term memory network outputs timing characteristics.
The server according to claim 12, wherein the prediction of the growth trend of negative news and the epidemic trend of infectious diseases according to the temporal and spatial characteristics, and when the prediction result is obtained, specifically executes:

Using the spatiotemporal features as the input of the fourth layer of deep neural network in the pre-trained second machine learning model to obtain the output of the fourth layer of deep neural network;

The output of the fourth layer of deep neural network is input to the output layer of the pre-trained second machine learning model, and the output layer outputs the prediction result.
A computer-readable storage medium in which a computer program is stored, and the computer program is executed by a processor to implement the following method:

Collect statistics on new media data and epidemic data on each date within the first preset date range, where the new media data includes negative news data and other new media data, and the epidemic data includes epidemic infection data;

Call the epidemic prevention and control effect prediction model, and predict the epidemic prevention and control effect based on the new media data and the epidemic data, and obtain the prediction result. The prediction result includes the negative news data on each date within the second preset date range And epidemic infection data, the second preset date range is after the first preset date range;

The prediction result is sent to the terminal device so that the terminal device can display the prediction result.
The computer-readable storage medium according to claim 15, wherein:

The negative news data includes at least one of the following: the number of negative news during the epidemic and the growth rate of negative news during the epidemic;

The other new media data includes at least one of the following: the release method of epidemic-related data, the audience category of the new media platform corresponding to each release method, the activity data of the new media platform, and the release for the new media platform The number of labels marked on the epidemic-related data of the new media platform, and the number of news items corresponding to each category of the epidemic-related data released by the new media platform;

The epidemic data includes epidemic infection data, and the epidemic infection data includes at least one of the following: the number of people infected by the epidemic and the growth rate of the number of people infected by the epidemic.
The computer-readable storage medium according to claim 15 or 16, wherein the epidemic prevention and control effect prediction model is a pre-trained first machine learning model, and the first machine learning model is constructed based on a deep neural network; Or, the epidemic prevention and control effect prediction model is a pre-trained second machine learning model, and the second machine learning model is constructed based on a convolutional neural network, a deep neural network, and a long and short-term memory network.
The computer-readable storage medium according to claim 16, wherein when the epidemic prevention and control effect prediction model is a pre-trained second machine learning model, the epidemic prevention and control effect prediction model is invoked, and according to the new Media data and the said epidemic data predict the effect of epidemic prevention and control, and when the prediction result is obtained, the specific realization:

Extracting spatial features based on the new media data and the epidemic data through the convolutional neural network in the pre-trained second machine learning model;

Extracting time series features according to the new media data and the epidemic data through the deep neural network and the long-short-term memory network in the pre-trained second machine learning model;

Performing fusion processing on the spatial feature and the time sequence feature to obtain a temporal feature;

According to the temporal and spatial characteristics, the negative news growth trend and the epidemic trend of infectious diseases are predicted, and the prediction result is obtained.
The computer-readable storage medium according to claim 18, wherein the deep neural network and the long-short-term memory network in the second machine learning model that are pre-trained are based on the new media data and the epidemic data When extracting timing features, the specific implementation is as follows:

Input the new media data and the epidemic data into the first layer of deep neural network in the pre-trained second machine learning model to obtain the output of the first layer of deep neural network;

Using the output of the first-layer deep neural network as the input of the second-layer long-short-term memory network in the pre-trained second machine learning model to obtain the output of the second-layer long- and short-term memory network;

The output of the second-layer long- and short-term memory network is used as the output of the third-layer long- and short-term memory network in the pre-trained second machine learning model, and the third-layer long- and short-term memory network outputs timing characteristics.
18. The computer-readable storage medium according to claim 18, wherein the prediction of the growth trend of negative news and the epidemic trend of infectious diseases according to the temporal and spatial characteristics, and when the prediction result is obtained, the specific realization is implemented:

Using the spatiotemporal features as the input of the fourth layer of deep neural network in the pre-trained second machine learning model to obtain the output of the fourth layer of deep neural network;

The output of the fourth layer of deep neural network is input to the output layer of the pre-trained second machine learning model, and the output layer outputs the prediction result.