CN111430035B - Method, device, electronic device and medium for predicting number of infectious diseases - Google Patents
Method, device, electronic device and medium for predicting number of infectious diseases Download PDFInfo
- Publication number
- CN111430035B CN111430035B CN202010194757.4A CN202010194757A CN111430035B CN 111430035 B CN111430035 B CN 111430035B CN 202010194757 A CN202010194757 A CN 202010194757A CN 111430035 B CN111430035 B CN 111430035B
- Authority
- CN
- China
- Prior art keywords
- data
- unit time
- time
- target
- prediction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Public Health (AREA)
- Human Resources & Organizations (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Game Theory and Decision Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Primary Health Care (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Epidemiology (AREA)
- Development Economics (AREA)
- Pathology (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Biomedical Technology (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The embodiment of the disclosure provides a method and a device for predicting the number of people with infectious diseases, electronic equipment and a computer-readable storage medium, belonging to the technical field of medical data, wherein the method for predicting the number of people with infectious diseases comprises the following steps: correcting original medical data related to a target disease to obtain corrected data corresponding to the original medical data; determining multidimensional feature data by combining the correction data and the associated data corresponding to the target disease; and predicting according to multi-dimensional characteristic data corresponding to reference duration before the target time so as to determine the number of cases belonging to the target disease aiming at the target time. The embodiment of the disclosure can improve the accuracy of case quantity prediction.
Description
Technical Field
The embodiment of the disclosure relates to the technical field of medical data, in particular to an infectious disease people number prediction method, an infectious disease people number prediction device, electronic equipment and a computer-readable storage medium.
Background
After a country or a region has an infectious epidemic, the accurate prediction of the infected people and the future tendency of the epidemic is of great significance for controlling diseases.
In the related art, fixed parameters are generally used to build a model and predict the number of cases. In this way, fixed parameters are difficult to describe in a complex situation and therefore have certain limitations. In addition, in the related art, only historical data can be used for prediction, and certain hysteresis exists, so that missing recognition can be caused, and the accuracy and reliability of a prediction result are poor.
It is to be noted that the information disclosed in the above background section is only for enhancement of understanding of the background of the present invention and therefore may include information that does not constitute prior art known to a person of ordinary skill in the art.
Disclosure of Invention
An object of the embodiments of the present disclosure is to provide a method for predicting the number of persons with infectious diseases, an apparatus for predicting the number of persons with infectious diseases, an electronic device, and a computer-readable storage medium, thereby overcoming, at least to some extent, the problems of low accuracy and poor reliability of data prediction.
Additional features and advantages of the disclosed embodiments will be set forth in the detailed description which follows, or may be learned by practice of the invention.
According to an aspect of an embodiment of the present disclosure, there is provided a method for predicting the number of persons with infectious diseases, including: correcting original medical data related to a target disease to obtain corrected data corresponding to the original medical data; determining multidimensional feature data by combining the correction data and the associated data corresponding to the target disease; and predicting according to multi-dimensional characteristic data corresponding to reference duration before the target time so as to determine the number of cases belonging to the target disease aiming at the target time.
In an exemplary embodiment of the disclosure, the modifying original medical data related to a target disease, and acquiring modified data corresponding to the original medical data, includes: and performing logical operation on the original medical data of the unit time to be processed and the corrected data of the next unit time adjacent to the unit time to be processed based on the time-from-back exponential averaging technology to obtain the corrected data of the unit time to be processed.
In an exemplary embodiment of the disclosure, the modifying the original medical data related to the target disease and obtaining modified data corresponding to the original medical data include: and carrying out average processing on increment of the original medical data in unit time to obtain correction data corresponding to the original medical data.
In an exemplary embodiment of the present disclosure, the predicting according to the multi-dimensional feature data corresponding to a reference time length before a target time to determine the number of cases belonging to the target disease for the target time includes: and inputting the multidimensional characteristic data corresponding to the reference duration into a trained machine learning model for prediction, and determining the number of cases at the target time.
In an exemplary embodiment of the present disclosure, the method further comprises: and training the machine learning model through the historical data and the real result of the historical data to obtain the trained machine learning model.
In an exemplary embodiment of the disclosure, the training a machine learning model through historical data and a real result of the historical data to obtain the trained machine learning model includes: inputting historical data into a machine learning model for prediction to obtain a prediction result of the historical data; and adjusting the weight parameters of the machine learning model according to the comparison result of the prediction result and the real result until the prediction result is consistent with the real result so as to obtain the trained machine learning model.
In an exemplary embodiment of the disclosure, the inputting the historical data into a machine learning model for prediction to obtain a prediction result of the historical data includes: correcting historical data related to the target disease to obtain corrected data corresponding to the historical data; determining reference multi-dimensional characteristic data by combining the correction data corresponding to the historical data and the historical associated characteristic data corresponding to the target disease; and predicting the quantity of the reference multidimensional correction data through the machine learning model to obtain a prediction result of the historical data.
According to an aspect of an embodiment of the present disclosure, there is provided an infectious disease person prediction apparatus including: the data correction module is used for correcting original medical data related to the target disease and acquiring corrected data corresponding to the original medical data; the characteristic extraction module is used for determining multi-dimensional characteristic data by combining the correction data and the associated data corresponding to the target disease; and the quantity prediction module is used for predicting according to the multi-dimensional characteristic data corresponding to the reference duration before the target time so as to determine the quantity of the cases belonging to the target disease aiming at the target time.
According to an aspect of the embodiments of the present disclosure, there is provided a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the infectious disease person number prediction method as described in any one of the above.
According to an aspect of an embodiment of the present disclosure, there is provided an electronic device including: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute any one of the above methods of predicting the number of infectious diseases via execution of the executable instructions.
In the infectious disease population prediction method, the infectious disease population prediction device, the electronic device and the computer-readable storage medium, correction data corresponding to original medical data are obtained by correcting the original medical data, multi-dimensional feature data used for expressing the characteristics of the original medical data are obtained according to the correction data, and the number of cases in a target time is determined by predicting the number of the cases according to the multi-dimensional feature data of the correction data in a reference time length. On one hand, the original medical data can be corrected, so that data deviation caused by hysteresis is reduced, and the accuracy of the data is improved. On the other hand, the number of cases can be predicted by referring to the multi-dimensional characteristic data corresponding to the duration, the case data can be predicted from multiple dimensions, the limitation caused by prediction only based on fixed parameters in the related technology is avoided, the characteristics of the multiple dimensions can be comprehensively considered, the error is reduced, the number of cases corresponding to the target time can be accurately predicted, dynamic prediction can be achieved, and the reliability of the prediction result is improved.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention. It is obvious that the drawings in the following description are only some embodiments of the invention, and that for a person skilled in the art, other drawings can be derived from them without inventive effort. In the drawings:
FIG. 1 schematically illustrates a system architecture diagram for performing a method for predicting the number of persons with an infectious disease according to an embodiment of the present disclosure;
fig. 2 schematically illustrates a flow chart of an infectious disease population prediction method according to an embodiment of the present disclosure;
FIG. 3 schematically illustrates a flow diagram of model training of an embodiment of the present disclosure;
FIG. 4 schematically shows a diagram of determining the number of cases for a target time in an embodiment of the disclosure;
FIG. 5 schematically illustrates a block diagram of an infectious disease person prediction apparatus according to an embodiment of the present disclosure;
fig. 6 schematically shows a block diagram of an electronic device for implementing the infectious disease person number prediction method described above.
Detailed Description
Example embodiments will now be described more fully with reference to the accompanying drawings. Example embodiments may, however, be embodied in many different forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of example embodiments to those skilled in the art.
Furthermore, the described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of embodiments of the invention. One skilled in the relevant art will recognize, however, that the invention may be practiced without one or more of the specific details, or with other methods, components, devices, steps, and so forth. In other instances, well-known methods, devices, implementations or operations have not been shown or described in detail to avoid obscuring aspects of the invention.
The block diagrams shown in the figures are functional entities only and do not necessarily correspond to physically separate entities. I.e. these functional entities may be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different networks and/or processor means and/or microcontroller means.
The flow charts shown in the drawings are merely illustrative and do not necessarily include all of the contents and operations/steps, nor do they necessarily have to be performed in the order described. For example, some operations/steps may be decomposed, and some operations/steps may be combined or partially combined, so that the actual execution sequence may be changed according to the actual situation.
A system architecture diagram for performing the infectious disease person number prediction method in an embodiment of the present disclosure is schematically shown in fig. 1.
As shown in fig. 1, the system architecture 100 may include a first end 101, a network 102, and a second end 103. The first end 101 may be a client, and may be, for example, various handheld devices (smart phones) with display functions, desktop computers, vehicle-mounted devices, and the like. The network 102 is used as a medium for providing a communication link between the first end 101 and the second end 103, the network 102 may include various connection types, such as a wired communication link, a wireless communication link, and the like, and in the embodiment of the present disclosure, the network 102 between the first end 101 and the second end 103 may be a wired communication link, such as a communication link provided by a serial connection line, or a wireless communication link, such as a communication link provided by a wireless network. The second terminal 103 may be a client, for example, a terminal device with a data processing function, such as a portable computer, a desktop computer, a smart phone, and the like, for predicting the number of cases of the target disease. When the first end 101 and the second end 103 are both clients, they may be the same client. The second end 103 may also be a server, which is not limited herein.
It should be understood that the number of first ends, networks and second ends in fig. 1 is merely illustrative. There may be any number of clients, networks, and servers, as desired for an implementation.
It should be noted that the method for predicting the number of persons in an infectious disease provided by the embodiment of the present disclosure may be completely executed by the second end 103, and accordingly, the device for predicting the number of persons in an infectious disease may be disposed in the second end 103.
Based on the system architecture, the embodiment of the present disclosure provides an infectious disease people number prediction method, which can be applied to an application scenario in which medical data is processed to predict the data. The main body of the infectious disease people number prediction method can be a server or a terminal with computing capability, and as shown in fig. 2, the infectious disease people number prediction method includes steps S210 to S230, which are described in detail as follows:
in step S210, original medical data related to a target disease is corrected, and correction data corresponding to the original medical data is acquired;
in step S220, determining multidimensional feature data by combining the corrected data and the associated data corresponding to the target disease;
in step S230, prediction is performed according to the multi-dimensional feature data corresponding to a reference time length before the target time, so as to determine the number of cases belonging to the target disease for the target time.
In the infectious disease people number prediction method provided by the embodiment of the disclosure, on one hand, the original medical data can be corrected, so that data deviation caused by hysteresis is reduced, and the accuracy of the data is improved. On the other hand, the number of cases can be predicted through the multi-dimensional feature data, the case data can be predicted from multiple dimensions, the limitation caused by prediction only based on fixed parameters in the related technology is avoided, the features of the multiple dimensions can be considered comprehensively, errors are reduced, the number of cases corresponding to the target time is predicted accurately, dynamic prediction can be achieved, and the reliability of the prediction result is improved.
Next, the method for predicting the number of infectious diseases in the embodiment of the present disclosure will be explained in detail with reference to the drawings.
In step S210, original medical data related to a target disease is corrected, and correction data corresponding to the original medical data is acquired.
In the embodiment of the present disclosure, the target disease may be an infectious disease (infectious disease), for example, various epidemic situations or various infectious influenza, infectious diseases, and the like. The target disease may be for a certain region or all regions, and is not limited herein. The original medical data refers to data about the target disease that has been published before the current time, and may be historical confirmed data, i.e., the number of cases belonging to the target disease. For data processing, raw medical data related to a target disease may be acquired from a database of a hospital or an institution such as a medical treatment and medical center that can confirm the target disease, or may be acquired from a network. In order to avoid the problem that the number of determined cases is less than the actual number due to the lag of the initial medical response of the target disease discovery, the original medical data related to the target disease may be corrected in the embodiment of the present disclosure so that the original medical data more conforms to the actual data.
In the embodiment of the present disclosure, the original medical data may be modified by smoothing the original medical data. Specifically, the manner of correcting the original medical data related to the target disease may include: and performing logical operation on the original medical data of the unit time to be processed and the corrected data of the next unit time adjacent to the unit time to be processed based on the time-from-back exponential averaging technology to obtain the corrected data of the unit time to be processed. The unit of time may be day, so the unit time to be processed may be any day before the target time, for example, may be every day. The next unit time is used to represent the next day of each day. It should be noted that the unit time to be processed is a time when the original medical data has been determined, and the unit time to be processed may be a time when the original medical data has been determined on any day of the next day and the following days. The exponential averaging technique is a technique often used in a gradient descent algorithm, and the exponential averaging achieves smoothing of data by calculating a dynamic average of a history value and a current value so that noise data can be well suppressed, and the algorithm is more stable.
Specifically, the raw medical data may be assumed to be [ y1,y2,...,yi,...,yt]The correction data is [ n ]1,n2,...,ni,...,nt]. In this case, the correction data corresponding to the original medical data may be determined by the following method and formula (1), in which:
let yt＝ntThe loop is completed with i as input, starting with i as t-1, and ending until i equals 1.
ni＝β*ni+1+(1-β)*yiFormula (1)
Wherein β is a constant less than 1.
In addition, other smoothing techniques may be used to modify the raw medical data. Specifically, the increment of the original medical data in unit time may be averaged to obtain the correction data corresponding to the original medical data. The increment refers to an amount of increase of case data per unit time with respect to the last unit time. However, due to the hysteresis of the initial stage, the increment per unit time may not be generated by the one unit time but may be generated by a plurality of unit times in common. The amount of increase in the raw medical data per unit time can be distributed to all the unit times. All unit times herein refer to the unit times for which the original medical data has been determined, i.e. the number of cases has been determined for all days. The allocation may be performed by an average allocation or a ratio allocation, and the allocation may be performed by various logical operations, which is not particularly limited.
For example, if the target disease starts on day 1/3 and the increment of day 3/3 is 900, the increments of 900 may be equally distributed to all days including day 3/3, and the correction data per unit time may be the sum of the original medical data and the increment distributed per unit time.
In the embodiment of the disclosure, the original medical data of each unit time is corrected according to the data of the later unit time, so that the problem that the number of confirmed cases is far lower than the actual number due to the hysteresis of medical response in the initial stage of epidemic outbreak can be avoided, the deviation of case data can be corrected, the corrected data is more accurate, the learned and trained model is more accurate, and the model is more in line with the actual situation.
With continued reference to fig. 2, in step S220, multi-dimensional feature data is determined in combination with the corrected data and the associated data corresponding to the target disease.
In the embodiment of the present disclosure, after obtaining the correction data, feature data indicating the development state of the target disease may be obtained on the basis of the correction data, where the feature data may be used to uniquely indicate a corresponding feature of the target disease. In order to avoid the problem of limitation caused by prediction from fixed parameters in the related art, in the embodiment of the present disclosure, the multidimensional feature data may be determined by combining the correction data that meets the actual situation and the associated data corresponding to the target disease. The relevant data corresponding to the target disease may be data (e.g., clinical data, treatment data, input data, etc.) related to the development status of the target disease, and is particularly used for assisting in predicting the target disease. The associated data can be derived from data of multiple dimensions in order to provide assistance comprehensively. The plurality of dimensions of the associated data may be at least one or more of treatment dimensions, user dimensions, input dimensions, and strength of governance. The treatment dimension can be hospital bed number, the user dimension can be the number of days (latent time length) of confirmed diagnosis of onset, the input dimension is the number of inflows of foreign population, and the control degree dimension can be the degree of activity of people stream and the like. The associated data may be dynamic data, i.e. the associated data is updated in real time each day. It should be noted that the associated data may also be derived from more dimensions, and may be specifically determined according to actual requirements. The multidimensional feature data refers to multidimensional feature data per unit time, and the multidimensional feature data per unit time may be different, and is determined according to actual correction data and associated data.
In the embodiment of the present disclosure, the description will be given by taking an example in which the relevant data is derived from the number of hospital beds, the number of days until diagnosis is confirmed (i.e., the latency time), and the number of incoming foreign population. The associated data may be obtained from statistical information in a database. Specifically, after the correction data and the associated data are acquired, the correction data and the associated data may be spliced and combined. When the correction data and the related data are combined, they may be spliced in any order as long as they can include all of the correction data and the related data, and the present invention is not particularly limited thereto.
For example, the multidimensional feature data obtained by combining the correction data and the associated data can be represented as xiAnd multi-dimensional feature data xi＝[ni,mi,bi,di]. Wherein n isiFor indicating the number of historical infections, where the number of historical infections refers to the correction data, m, described aboveiFor indicating the number of incoming foreign bodies, biFor indicating the number of hospital beds, diUsed to indicate the number of days from onset to diagnosis.
In the embodiment of the disclosure, the correction data and the associated data corresponding to the target disease are combined to obtain the multi-dimensional characteristic data for representing the development state of the target disease, so that the problem of inaccurate prediction result caused by prediction only according to one dimension of historical infection quantity in the related art is solved, the function of comprehensive prediction from multiple dimensions can be realized, the complex situation can be dynamically described through multiple dimensions, and the accuracy and the usability are improved.
Continuing to refer to fig. 2, in step S230, prediction is performed according to the multi-dimensional feature data corresponding to the reference time length before the target time to determine the number of cases belonging to the target disease for the target time.
In the embodiment of the present disclosure, the reference time length refers to a time length which is located before the target time and in which the unit time composition of the original medical data and the corrected data has been determined. The reference time duration may include one unit time or a plurality of unit times, and the reference time duration includes several unit times, which may be specifically determined according to the position of the target time in the whole development process of the target disease, and the position may be described by the several unit times. The target time refers to a unit time (i.e., a future day) in which an undetermined number of current times are located. Specifically, if the target time is earlier than day 4, all the unit times before day 4 may be taken as the reference time length. If the target time is later than the 4 th day (e.g., the 5 th day), a preset number of unit times before the 5 th day may be used as the reference time period. Wherein the unit time of the preset number can be 3 days or 4 days, etc. Tests show that when the unit time of the preset quantity is 4 days, the obtained result is more accurate. On the basis, if the target time is day 2, the reference time length can be determined according to a unit time before day 2; if the target time is day 3, the reference time length is the previous 2 days; if the target time is day 6, the reference time period is 4 days which are adjacent to and before day 6. In the embodiment of the disclosure, the reference duration is determined by the position of the target time, so that the determined data has a reference value, and the result is more accurate.
After the reference duration is determined, the multidimensional feature data corresponding to each unit time included in the reference duration can be obtained, the multidimensional feature data of the unit time of the reference duration can be combined to obtain one piece of multidimensional feature data, and data prediction is performed based on the multidimensional feature data. For example, the multidimensional feature data [ x ] of the first 4 days adjacent to the i-th day may be usedi-4,xi-3,xi-2,xi-1]To predict the number of cases n at the target time (day i)i. If the multidimensional feature data of the previous 4 days is not determined, the multidimensional feature data of the previous 4 days needs to be estimated according to the method, and the number of cases at the target time (day i) needs to be estimated based on the estimated multidimensional feature data of the previous 4 days. For example, if only the data for 3 days is determined, the number of cases on the 4 th day can be estimated from the multidimensional feature data for the previous 3 days, and further, if the i th day is the 5 th day, the number of cases on the target time (the 5 th day) can be estimated from the multidimensional feature data for the previous 4 days on the 5 th day.
When the number of cases at the target time is predicted, the multi-dimensional feature data corresponding to the reference duration can be input into a trained machine learning model for prediction, and the number of cases at the target time is determined. The number of cases refers to the number of infectious diseases. The machine learning model may be any suitable model, such as a long-short term memory network or a tree model, and the long-short term memory network is exemplified in the embodiments of the present disclosure.
Before predicting the multidimensional feature data by using the trained machine learning model, in order to improve the accuracy of data prediction, the machine learning model may be trained first to obtain the trained machine learning model.
Fig. 3 schematically illustrates a flowchart of model training, and referring to fig. 3, the method specifically includes step S310 and step S320, where:
in step S310, inputting historical data into a machine learning model for prediction to obtain a prediction result of the historical data;
in step S320, a weight parameter of a machine learning model is adjusted according to a comparison result between the predicted result and the real result until the predicted result is consistent with the real result, so as to obtain the trained machine learning model.
In the embodiment of the disclosure, the historical data can be used as an input of the machine learning model, and a prediction result of the historical data can be determined through the machine learning model, and the prediction result here can be the number of cases corresponding to the historical data. In order to ensure the accuracy of the data, the historical data can be corrected to obtain corrected data corresponding to the historical data, and then the machine learning model is trained on the basis of the corrected data. Specifically, predicting historical data may include the steps of: correcting historical data related to the target disease to obtain corrected data corresponding to the historical data; determining reference multi-dimensional characteristic data by combining the correction data corresponding to the historical data and the historical associated characteristic data corresponding to the target disease; and predicting the quantity of the reference multidimensional correction data through the machine learning model to obtain a prediction result of the historical data. The historical data may be, among other things, raw medical data for all units of time prior to the target time. The historical data may be corrected by the same correction method as the original medical data. Specifically, the original medical data and the correction data of the next unit time adjacent to the unit time to be processed may be logically operated based on a time-backward exponential averaging technique to obtain the correction data of the unit time to be processed. Or averaging the increment of the original medical data in unit time to acquire the correction data corresponding to the original medical data contained in the historical data.
After the correction data of the history data is obtained, the correction data of the history data and the reference related data of the target disease in the unit time to which the history data belongs may be combined to obtain the reference multidimensional characteristic data. The reference associated data may here be the same as the associated data part and the reference multi-dimensional feature data may also be the same as the multi-dimensional feature data part.
Still further, the reference multi-dimensional feature data may be input into a machine learning model to predict the number of cases per unit time to which the history data belongs by a process such as convolution operation.
And further, parameters of the machine learning model can be adjusted and trained according to the real result and the prediction result of the historical data until the prediction result of the historical data obtained through the machine learning model is consistent with the real result, and the parameters with consistent results are used as the final parameters of the machine learning model to form the trained machine learning model. The history data refers to correction data for all unit times before the target time. The true result refers to the true number of units of time to which the history data belongs.
After the trained machine learning model is obtained, the multidimensional feature data corresponding to the reference time length before the reference target time can be input into the trained machine learning model for prediction, so as to determine the number of cases at the target time. Next, a machine learning model will be described as an example of a long-term and short-term memory network.
The long-term and short-term memory network in the embodiment of the present disclosure may include a plurality of gates, and specifically may include, but is not limited to, a forgetting gate, an updating gate, and an output gate. Wherein the forgetting gate is used to determine the discarded information. The forgetting gate reads the value transmitted by the weight block last time and the value input this time, and outputs a number between 0 and 1, wherein 1 represents complete retention, and 0 represents complete rejection. This function may help dynamically determine whether to remember the last entered value, and whether to consider the interaction of the two entered values.
The update gate may also be referred to as an input gate. For determining how much new information to add to the model. This process requires two specific steps: firstly, a sigmoid function layer needs to decide which input information needs to be updated; a tanh layer generates a vector, which is alternatively the content to be updated. Next, the two parts are combined to update the state of the model.
An output gate for determining what value to output, the output being based on the entire model. Firstly, a sigmoid function layer is used for determining which part of the model state is output. The model state is then processed through tanh to obtain a value between-1 and 1, which is multiplied by the output of the sigmoid gate, and only a portion of the determined output will be output.
For the long and short term memory network, a cell state can be included. Inputting multidimensional characteristic data x at time ttAnd hidden state h at time t-1t-1The state quantities of the refresh gate, the output gate, and the forgetting gate at time t are generated by the combined action of the weight W, the weight U, and the offset vector b, and U may be usedt、otAnd ftAnd (4) showing. Further generating a cell state c at time t under the action of the cell state at time t-1t. Finally, at time t, cell state ctAnd output gate otUnder the action of (3), generating a hidden state h at the time ttFurther, the internal changes of the neurons of the long-term and short-term memory network at the time t +1 are affected, and the specific process can refer to the formula (2):
wherein, the sum of the weights representing the cell state, the multi-dimensional feature data at the time t, the hidden state at the time t-1, the offset vector representing the cell state, the weight representing a forgetting gate, the offset vector representing a forgetting gate, the weight representing an updating gate, the offset vector representing an updating gate, the weight representing an output gate, the offset vector representing an output gate, the cell state at the time t-1, the cell state representing the time t, the forgetting gate representing the time t, the updating gate representing the time t, the output gate representing the time t, the hidden state representing the time t, and the number of predicted persons (namely, the number of predicted cases).
In the embodiment of the disclosure, on the basis of the long-short term memory network, the multi-dimensional characteristic data of a plurality of unit times in the reference duration is subjected to linear conversion, so that the number of cases at the target time is calculated through the trained long-short term memory network. For example, the first 4 days with the reference time length as the target time are taken as an example for explanation. The multi-dimensional feature data from day 1 to day 4 of the first 4 days may be sequentially represented as x i-4，xi-3，xi-2，x i-1. Referring to the schematic diagram of the prediction process shown in FIG. 4, the multi-dimensional feature data x of day 1 is obtainedi-4By hidden state hi-4Performing non-linear processing (non-linear transformation) to obtain the multi-dimensional feature data x of day 2i-3By hidden state hi-3Performing nonlinear processing to obtain 3 rd day multidimensional characteristic data xi-2By hidden state hi-2Performing nonlinear processing until the multi-dimensional characteristic data x of the 4 th dayi-1By hidden state hi-1Performing nonlinear processing to obtain output result of long-short term memory network
And the output result of the long-short term memory network is used as the number of cases belonging to the target disease corresponding to the target time. The hidden states are updated in real time, so that the hidden states corresponding to each unit time may be different.
In the technical scheme of fig. 4, the prediction is performed by referring to the multidimensional feature data of a plurality of unit times included in the duration and the hidden state of each unit time, so that the quantity prediction can be performed from a plurality of dimensions in combination with the historical data, and the dynamic prediction can be performed more accurately. Moreover, the method in the embodiment of the disclosure can be applied to various regions (for example, a city, a province or a country, etc.), so that the application range is increased, and the universality is improved.
In the embodiment of the disclosure, when the number of cases at the target time is predicted by using the long-short term memory network which can be used for time series analysis, the prediction can be performed according to the multidimensional feature data of a plurality of unit times contained in the reference time length, and the multidimensional feature data comprise data of a plurality of dimensions, so that the parameters of the model can be dynamically learned based on the data to obtain a well-trained machine learning model with better performance, and further, the prediction can be accurately performed based on the model, so that the prediction of the number of cases at the target time can be comprehensively and accurately performed, the obtained predicted number of cases is more accurate and more in line with the actual situation, and is not influenced by data hysteresis, and the reliability of the prediction result can be improved. Through the prediction by the characteristics of multiple dimensions, the more complex situation can be described, the following crowd flow and the dynamic prediction of the change of protective measures are realized, and the application range is wider. The LSTM (Long Short-Term Memory) method can automatically learn the parameters of the model through historical data, can depict more complex conditions in the real world, has stronger practicability, avoids limitation and increases usability.
In an embodiment of the present disclosure, there is also provided an infectious disease person prediction apparatus, and referring to fig. 5, the infectious disease person prediction apparatus 500 may include the following modules:
the data correction module 501 is configured to correct original medical data related to a target disease, and obtain correction data corresponding to the original medical data;
a feature extraction module 502, configured to determine multidimensional feature data by combining the correction data and the associated data corresponding to the target disease;
a quantity predicting module 503, configured to perform prediction according to the multi-dimensional feature data corresponding to the reference duration before the target time, so as to determine the quantity of cases belonging to the target disease for the target time.
In an exemplary embodiment of the present disclosure, the data modification module includes: the first correction module is used for performing logical operation on the original medical data of the unit time to be processed and the correction data of the next unit time adjacent to the unit time to be processed based on the time-backward exponential averaging technology to obtain the correction data of the unit time to be processed.
In an exemplary embodiment of the present disclosure, the data modification module includes: and the second correction module is used for carrying out average processing on the increment of the original medical data in unit time so as to obtain correction data corresponding to the original medical data.
In an exemplary embodiment of the present disclosure, the quantity prediction module includes: and the data prediction module is used for inputting the multidimensional characteristic data corresponding to the reference duration into a trained machine learning model for prediction and determining the number of cases at the target time.
In an exemplary embodiment of the present disclosure, the apparatus further includes: and the model training module is used for training the machine learning model according to the historical data and the real result of the historical data so as to obtain the trained machine learning model.
In an exemplary embodiment of the present disclosure, the model training module includes: the historical data prediction module is used for inputting historical data into a machine learning model for prediction so as to obtain a prediction result of the historical data; and the parameter adjusting module is used for adjusting the weight parameters of the machine learning model according to the comparison result of the prediction result and the real result until the prediction result is consistent with the real result so as to obtain the trained machine learning model.
In an exemplary embodiment of the disclosure, the historical data prediction module is configured to: correcting historical data related to the target disease to obtain corrected data corresponding to the historical data; determining reference multi-dimensional characteristic data by combining the correction data corresponding to the historical data and the historical associated characteristic data corresponding to the target disease; and predicting the quantity of the reference multidimensional correction data through the machine learning model to obtain a prediction result of the historical data.
It should be noted that, the functional modules of the infectious disease person number prediction apparatus according to the embodiment of the present disclosure are the same as the steps of the above described exemplary embodiment of the infectious disease person number prediction method, and therefore, the description thereof is omitted.
It should be noted that although in the above detailed description several modules or units of the device for action execution are mentioned, such a division is not mandatory. Indeed, the features and functions of two or more modules or units described above may be embodied in one module or unit according to an embodiment of the invention. Conversely, the features and functions of one module or unit described above may be further divided into embodiments by a plurality of modules or units.
Moreover, although the steps of the methods of the present invention are depicted in the drawings in a particular order, this does not require or imply that the steps must be performed in this particular order, or that all of the depicted steps must be performed, to achieve desirable results. Additionally or alternatively, certain steps may be omitted, multiple steps combined into one step execution, and/or one step broken into multiple step executions, etc.
In an exemplary embodiment of the present invention, there is also provided an electronic device capable of implementing the above method.
As will be appreciated by one skilled in the art, aspects of the present invention may be embodied as a system, method or program product. Thus, various aspects of the invention may be embodied in the form of: an entirely hardware embodiment, an entirely software embodiment (including firmware, microcode, etc.) or an embodiment combining hardware and software aspects that may all generally be referred to herein as a "circuit," module "or" system.
An electronic device 600 according to this embodiment of the invention is described below with reference to fig. 6. The electronic device 600 shown in fig. 6 is only an example and should not bring any limitations to the function and scope of use of the embodiments of the present disclosure.
As shown in fig. 6, the electronic device 600 is in the form of a general purpose computing device. The components of the electronic device 600 may include, but are not limited to: the at least one processing unit 610, the at least one memory unit 620, a bus 630 connecting different system components (including the memory unit 620 and the processing unit 610), and a display unit 640.
Wherein the memory unit stores program code that may be executed by the processing unit 610 to cause the processing unit 610 to perform steps according to various exemplary embodiments of the present invention as described in the above section "exemplary method" of this specification. For example, the processing unit 610 may perform the steps as shown in fig. 2.
The storage unit 620 may include readable media in the form of volatile memory units, such as a random access memory unit (RAM)6201 and/or a cache memory unit 6202, and may further include a read-only memory unit (ROM) 6203.
The memory unit 620 may also include a program/utility 6204 having a set (at least one) of program modules 6205, such program modules 6205 including, but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which, or some combination thereof, may comprise an implementation of a network environment.
The electronic device 600 may also communicate with one or more external devices 700 (e.g., keyboard, pointing device, bluetooth device, etc.), with one or more devices that enable a user to interact with the electronic device 600, and/or with any devices (e.g., router, modem, etc.) that enable the electronic device 600 to communicate with one or more other computing devices. Such communication may occur via an input/output (I/O) interface 650. Also, the electronic device 600 may communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network such as the Internet) via the network adapter 660. As shown, the network adapter 660 communicates with the other modules of the electronic device 600 over the bus 630. It should be appreciated that although not shown in the figures, other hardware and/or software modules may be used in conjunction with the electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, to name a few.
Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein may be implemented by software, or by software in combination with necessary hardware. Therefore, the technical solution according to the embodiment of the present invention may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (which may be a CD-ROM, a usb disk, a mobile hard disk, etc.) or on a network, and includes instructions to enable a computing device (which may be a personal computer, a server, a mobile terminal, or a network device, etc.) to execute the method for predicting the number of people with infectious diseases according to the embodiment of the present invention.
In an exemplary embodiment of the present invention, there is also provided a computer-readable storage medium having stored thereon a program product capable of implementing the above-described method of the present specification. In some possible embodiments, aspects of the invention may also be implemented in the form of a program product comprising program code means for causing a terminal device to carry out the steps according to various exemplary embodiments of the invention described in the above section "exemplary methods" of the present description, when said program product is run on the terminal device.
According to the program product for realizing the method, the portable compact disc read only memory (CD-ROM) can be adopted, the program code is included, and the program product can be operated on terminal equipment, such as a personal computer. However, the program product of the present invention is not limited in this respect, and in this document, a readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
A computer readable signal medium may include a propagated data signal with readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A readable signal medium may also be any readable medium that is not a readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Program code for carrying out operations of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server. In situations involving remote computing devices, the remote computing devices may be connected to the user computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to external computing devices (e.g., through the internet using an internet service provider).
Furthermore, the above-described figures are merely schematic illustrations of processes involved in methods according to exemplary embodiments of the invention, and are not intended to be limiting. It will be readily appreciated that the processes illustrated in the above figures are not intended to indicate or limit the temporal order of the processes. In addition, it is also readily understood that these processes may be performed synchronously or asynchronously, e.g., in multiple modules.
Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the invention being indicated by the following claims.
Claims (8)
1. A method for predicting the number of persons with an infectious disease, comprising:
distributing the increment of the original medical data of each unit time related to the target disease to all unit times, and acquiring the correction data of each unit time corresponding to the original medical data, wherein the correction data of each unit time is the sum of the original medical data and the increment distributed to each unit time;
combining the correction data and the associated data corresponding to the target disease to obtain multi-dimensional characteristic data; the relevant data corresponding to the target disease is data related to the development state of the target disease, and the relevant data comprises at least one or more of treatment dimension, user dimension, input dimension and control strength;
acquiring multi-dimensional characteristic data corresponding to each adjacent unit time in a reference time length before the target time; the reference duration is the reference duration which is located before the target time and consists of unit time of the original medical data and the corrected data, and the number of the unit time of the reference duration is determined according to the position of the target time in the development process of the target disease;
combining the multidimensional characteristic data of the unit time of the reference duration to obtain multidimensional characteristic data corresponding to the reference duration;
predicting according to multi-dimensional characteristic data corresponding to reference duration before the target time to determine the number of cases belonging to the target disease aiming at the target time;
the predicting according to the multi-dimensional feature data corresponding to the reference duration before the target time to determine the number of cases belonging to the target disease for the target time includes:
inputting the multidimensional characteristic data corresponding to the reference duration into a trained machine learning model for prediction, and determining the number of cases at the target time, wherein the trained machine learning model is obtained by training the machine learning model by using reference multidimensional characteristic data, and the reference multidimensional characteristic data is obtained by combining and determining correction data of historical data and reference associated data of the target disease in unit time to which the historical data belongs;
wherein the method further comprises:
if the multi-dimensional characteristic data corresponding to the unit time is not determined, estimating the number of cases in the unit time by using the multi-dimensional characteristic data corresponding to each adjacent unit time in the reference time length before the unit time, and obtaining the original medical data in the unit time.
2. An infectious disease person number prediction method according to claim 1, further comprising:
and performing logical operation on the original medical data of the unit time to be processed and the corrected data of the next unit time adjacent to the unit time to be processed based on the time-from-back exponential averaging technology to obtain the corrected data of the unit time to be processed.
3. An infectious disease person number prediction method according to claim 1, further comprising:
and training the machine learning model through the historical data and the real result of the historical data to obtain the trained machine learning model.
4. A method for predicting the number of persons having an infectious disease according to claim 3, wherein the training of the machine learning model based on the history data and the actual results of the history data to obtain the trained machine learning model comprises:
inputting historical data into a machine learning model for prediction to obtain a prediction result of the historical data;
and adjusting the weight parameters of the machine learning model according to the comparison result of the prediction result and the real result until the prediction result is consistent with the real result so as to obtain the trained machine learning model.
5. An infectious disease person number prediction method according to claim 4, wherein the inputting historical data into a machine learning model for prediction to obtain a prediction result of the historical data comprises:
correcting historical data related to the target disease to obtain corrected data corresponding to the historical data;
determining reference multi-dimensional characteristic data by combining the correction data corresponding to the historical data and the historical associated characteristic data corresponding to the target disease;
and predicting the quantity of the reference multi-dimensional feature data through the machine learning model to obtain a prediction result of the historical data.
6. An infectious disease person number prediction device, comprising:
the data correction module is used for distributing the increment of the original medical data of each unit time related to the target disease to all the unit times and acquiring the correction data of each unit time corresponding to the original medical data, wherein the correction data of each unit time is the sum of the original medical data and the increment distributed to each unit time;
the characteristic extraction module is used for combining the correction data and the associated data corresponding to the target disease to obtain multi-dimensional characteristic data; the relevant data corresponding to the target disease is data related to the development state of the target disease, and the relevant data comprises at least one or more of treatment dimension, user dimension, input dimension and control strength;
the data acquisition module is used for acquiring the multidimensional characteristic data reference time length corresponding to each adjacent unit time in the reference time length before the target time, wherein the multidimensional characteristic data reference time length is positioned before the target time, the reference time length formed by the unit time of the original medical data and the unit time of the corrected data is determined, and the number of the unit time included in the reference time length is determined according to the position of the target time in the development process of the target disease; combining the multidimensional characteristic data of the unit time of the reference duration to obtain multidimensional characteristic data corresponding to the reference duration;
the quantity prediction module is used for predicting according to multi-dimensional characteristic data corresponding to reference duration before the target time so as to determine the quantity of cases belonging to the target disease aiming at the target time;
the quantity prediction module is configured to input the multidimensional feature data corresponding to the reference duration into a trained machine learning model for prediction, and determine the quantity of cases at the target time, where the trained machine learning model is obtained by training a machine learning model using reference multidimensional feature data, and the reference multidimensional feature data is determined by combining correction data of historical data and reference associated data of a target disease in unit time to which the historical data belongs;
and the data estimation module is used for estimating the number of cases in the unit time by using the multi-dimensional characteristic data corresponding to each adjacent unit time in the reference time length before the unit time if the multi-dimensional characteristic data corresponding to the unit time is not determined, so as to obtain the original medical data in the unit time.
7. A computer-readable storage medium, on which a computer program is stored, wherein the computer program, when executed by a processor, implements the infectious disease person number prediction method according to any one of claims 1 to 5.
8. An electronic device, comprising:
a processor; and
a memory for storing executable instructions of the processor;
wherein the processor is configured to perform the infectious disease person number prediction method of any one of claims 1-5 via execution of the executable instructions.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010194757.4A CN111430035B (en) | 2020-03-19 | 2020-03-19 | Method, device, electronic device and medium for predicting number of infectious diseases |
| CN202210709133.0A CN115101211A (en) | 2020-03-19 | 2020-03-19 | Method, device, electronic device and medium for predicting number of infectious diseases |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010194757.4A CN111430035B (en) | 2020-03-19 | 2020-03-19 | Method, device, electronic device and medium for predicting number of infectious diseases |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210709133.0A Division CN115101211A (en) | 2020-03-19 | 2020-03-19 | Method, device, electronic device and medium for predicting number of infectious diseases |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111430035A CN111430035A (en) | 2020-07-17 |
| CN111430035B true CN111430035B (en) | 2022-07-12 |
Family
ID=71546515
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010194757.4A Active CN111430035B (en) | 2020-03-19 | 2020-03-19 | Method, device, electronic device and medium for predicting number of infectious diseases |
| CN202210709133.0A Pending CN115101211A (en) | 2020-03-19 | 2020-03-19 | Method, device, electronic device and medium for predicting number of infectious diseases |
Family Applications After (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210709133.0A Pending CN115101211A (en) | 2020-03-19 | 2020-03-19 | Method, device, electronic device and medium for predicting number of infectious diseases |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN111430035B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112201361A (en) * | 2020-09-01 | 2021-01-08 | 浙江大学山东工业技术研究院 | COVID-19 epidemic situation prediction method based on LSTM model |
| CN113257426B (en) * | 2021-06-30 | 2021-09-21 | 杭州华网信息技术有限公司 | Aggregated group flu prediction system, storage medium and device |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2022008719A (en) * | 2016-11-23 | 2022-01-14 | セルヴァス エーアイ インコーポレイテッド | Method and apparatus for predicting disease onset |
| CN108281198A (en) * | 2017-12-19 | 2018-07-13 | 昆明理工大学 | A kind of method of early warning infectious disease early period |
| CN109036579B (en) * | 2018-08-22 | 2021-11-26 | 泰康保险集团股份有限公司 | Information prediction method, device, medium and electronic equipment based on block chain |
| CN109616205A (en) * | 2018-12-04 | 2019-04-12 | 泰康保险集团股份有限公司 | Data processing method, device, medium and electronic equipment |
| CN109859854A (en) * | 2018-12-17 | 2019-06-07 | 中国科学院深圳先进技术研究院 | Prediction Method of Communicable Disease, device, electronic equipment and computer-readable medium |
-
2020
- 2020-03-19 CN CN202010194757.4A patent/CN111430035B/en active Active
- 2020-03-19 CN CN202210709133.0A patent/CN115101211A/en active Pending
Also Published As
| Publication number | Publication date |
|---|---|
| CN115101211A (en) | 2022-09-23 |
| CN111430035A (en) | 2020-07-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Jin et al. | Provably efficient reinforcement learning with linear function approximation | |
| CN110852438B (en) | Model generation method and device | |
| US11222262B2 (en) | Non-Markovian control with gated end-to-end memory policy networks | |
| CN112289442B (en) | Method and device for predicting disease end point event and electronic equipment | |
| CN111430035B (en) | Method, device, electronic device and medium for predicting number of infectious diseases | |
| Birhanu et al. | Doubly robust and multiple-imputation-based generalized estimating equations | |
| CN112382406B (en) | Method, apparatus, medium, and device for estimating basic regeneration number of infectious disease | |
| EP3916641A1 (en) | Continuous time self attention for improved computational predictions | |
| CN115496285B (en) | Power load prediction method and device and electronic equipment | |
| CN115101061A (en) | Training method and device of voice recognition model, storage medium and electronic equipment | |
| EP3895080A1 (en) | Regularization of recurrent machine-learned architectures | |
| EP4287182A1 (en) | Method and apparatus for training a speech recognition model, electronic device and storage medium | |
| CN109993638A (en) | Method, apparatus, medium and the electronic equipment of Products Show | |
| CN111489830A (en) | Method and device for predicting epidemic situation data in sections, medium and electronic equipment | |
| CN113484882B (en) | GNSS sequence prediction method and system of multi-scale sliding window LSTM | |
| CN113642635B (en) | Model training method and device, electronic equipment and medium | |
| CN114118570A (en) | Service data prediction method and device, electronic equipment and storage medium | |
| CN117351299A (en) | Image generation and model training method, device, equipment and storage medium | |
| EP4134873A1 (en) | Neurosymbolic data imputation using autoencoder and embeddings | |
| CN117035842A (en) | Model training method, traffic prediction method, device, equipment and medium | |
| CN110991661A (en) | Method and apparatus for generating a model | |
| CN115312151A (en) | Data information generation method and device, computer equipment and storage medium | |
| JP7349811B2 (en) | Training device, generation device, and graph generation method | |
| CN110020738B (en) | Method, apparatus, electronic device and computer readable medium for data processing | |
| US11308431B2 (en) | Hierarchical optimization for processing objectives sequentially and/or iteratively |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |