Place classification method and device, storage medium and electronic equipment

Abstract

The embodiment of the invention relates to a place classification method and device, a storage medium and electronic equipment, and relates to the technical field of medical big data, and the place classification method comprises the steps: obtaining a to-be-evaluated place; determining an influence index of the to-be-evaluated place, and generating feature data of the to-be-evaluated place based on the influence index; and fitting the feature data according to a trained place classification model to obtain index parameters of infectious diseases corresponding to the to-be-evaluated place, and determining level information of the to-be-evaluated place through the index parameters of the infectious diseases. According to the technical scheme provided by the embodiment of the invention, the level information of the to-be-evaluated place can be accurately determined, the identification efficiency and accuracy are improved, and guidance is provided for place management.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] The present disclosure claims priority from application number: 202311667338.8, application date: December 6, 2023, and invention name: Place classification method and device, storage medium, electronic device. The entire contents of the Chinese patent application are incorporated herein by reference in their entirety. Technical Field

[0003] The present disclosure relates to the field of medical big data technology, and in particular to a place classification method, a place classification device, a computer-readable storage medium, and an electronic device. Background Art

[0004] Various types of infectious diseases have the characteristics of short incubation period and rapid transmission speed, so it is necessary to conduct point investigation and identification of key places as soon as possible.

[0005] In related technologies, all places where infectious diseases occur are generally subject to the same type of control; in addition, they can be differentiated according to the type of place, or they can be scored based on multi-dimensional indicators and places that meet the indicator conditions can be managed. In the above methods, there may be problems of missed identification or wrong identification, the accuracy of classifying places is low, the coverage is not comprehensive, and the versatility is poor.

[0006] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0007] The purpose of the present disclosure is to provide a place classification method and device, a storage medium, and an electronic device, thereby overcoming the problem of poor place classification accuracy caused by the limitations and defects of related technologies to at least a certain extent.

[0008] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.

[0009] According to a first aspect of the present disclosure, a place classification method is provided, comprising: obtaining a place to be evaluated; determining an impact index of the place to be evaluated, and generating feature data of the place to be evaluated based on the impact index; fitting the feature data according to a trained place classification model to obtain index parameters of the infectious disease corresponding to the place to be evaluated, and determining the level information of the place to be evaluated through the index parameters of the infectious disease.

[0010] In an exemplary embodiment of the present disclosure, determining the impact index of the place to be evaluated includes: determining the impact index of the place to be evaluated based on the environmental state corresponding to the place to be evaluated; wherein the environmental state includes a first type of state or a second type of state; determining the impact index of the place to be evaluated based on the environmental state corresponding to the place to be evaluated includes: if the environmental state is the first type of state, using the objective environmental indicators of each place to be evaluated, the case indicators of infectious disease patients, and the place personnel indicators of reference personnel existing in the place to be evaluated as the impact indicators; if the environmental state is the second type of state, using the objective environmental indicators of each place to be evaluated as the impact indicators.

[0011] In an exemplary embodiment of the present disclosure, generating the characteristic data of the place to be evaluated based on the impact indicator includes: numerically mapping the impact indicator based on the type of the impact indicator to obtain the characteristic data of the place to be evaluated.

[0012] In an exemplary embodiment of the present disclosure, the characteristic data is fitted according to the trained place classification model to obtain the indicator parameters of the infectious disease, including: determining the target characteristic data according to one or more of the attribute characteristics or prediction requirements of the characteristic data, and fitting the target characteristic data based on the trained place classification model to obtain the indicator parameters.

[0013] In an exemplary embodiment of the present disclosure, determining the target feature data based on one or more of the attribute characteristics or predicted requirements of the feature data includes: if the attribute characteristics or predicted requirements of the feature data meet the data screening conditions, determining all feature data as the target feature data; if the attribute characteristics or predicted requirements of the feature data do not meet the data screening conditions, screening all feature data based on feature importance to determine the target feature data.

[0014] In an exemplary embodiment of the present disclosure, the method further includes: acquiring sample epidemiological survey data, and determining a place classification model from multiple candidate place classification models based on the sample epidemiological survey data; and training the place classification model to obtain the trained place classification model.

[0015] In an exemplary embodiment of the present disclosure, the method of determining a place classification model from multiple candidate place classification models based on sample epidemic survey data includes: segmenting the sample epidemic survey data to obtain K sub-sample data, using each of the K sub-sample data as a test set, and using the remaining sub-sample data as a training set, training multiple candidate place classification models through the training set to obtain multiple trained candidate place classification models; evaluating the multiple trained candidate place classification models based on multiple evaluation parameters, and determining the place classification model according to the evaluation results.

[0016] In an exemplary embodiment of the present disclosure, the multiple trained candidate place classification models are evaluated based on multiple evaluation parameters, and the place classification model is determined according to the evaluation results, including: obtaining the sensitivity and specificity of each trained candidate place classification model, and determining an evaluation curve based on the specificity and the sensitivity; determining one or more intermediate candidate classification models based on the area parameter of the evaluation curve, and screening the intermediate candidate classification models according to sensitivity and / or specificity to obtain the place classification model.

[0017] In an exemplary embodiment of the present disclosure, the training of the place classification model to obtain the trained place classification model includes: obtaining sample data, obtaining historical impact indicator data from the sample data, and obtaining historical feature data based on the historical impact indicator data; inputting the historical feature data into the place classification model to obtain prediction indicator parameters corresponding to the sample data, and comparing the predicted indicator parameters with the actual indicator parameters of the sample data to determine the difference between the two; determining a loss function based on the difference, and training the place classification model based on the loss function to obtain the trained place classification model.

[0018] According to a second aspect of the present disclosure, a place classification device is provided, comprising: a place acquisition module, used to acquire a place to be evaluated; an indicator determination module, used to determine the impact indicator of the place to be evaluated, and generate characteristic data of the place to be evaluated based on the impact indicator; a level information determination module, used to fit the characteristic data according to a trained place classification model, obtain the indicator parameters of the infectious disease corresponding to the place to be evaluated, and determine the level information of the place to be evaluated through the indicator parameters of the infectious disease.

[0019] According to a third aspect of the present disclosure, there is provided a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements any one of the above-mentioned methods for classifying places.

[0020] According to a fourth aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute any one of the above-mentioned place classification methods by executing the executable instructions.

[0021] In the place classification method, place classification device, electronic device and computer-readable storage medium provided in the embodiments of the present disclosure, on the one hand, the characteristic data of the place to be evaluated is fitted based on the trained place classification model to obtain the indicator parameters of the infectious disease, and then the level information of the place to be evaluated is determined according to the indicator parameters, which can accurately determine the level of the place to be evaluated, avoid the limitations of classification in related technologies, and can perform comprehensive and accurate classification, improve the accuracy of place classification and the efficiency of place classification. On the other hand, by determining the influencing indicators of multiple dimensions according to the environmental state, and then generating accurate characteristic data, the place level information can be accurately evaluated for different environmental states, which increases the scope of application, can improve versatility, and improves recognition efficiency.

[0022] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the specification are used to explain the principles of the present disclosure. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure, and for ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0024] Figure 1 A schematic diagram schematically illustrates a method for classifying places in an embodiment of the present disclosure.

[0025] Figure 2 The following is a schematic diagram of a process for determining feature data in an embodiment of the present disclosure.

[0026] Figure 3 A schematic diagram schematically illustrates a flow chart of evaluation curves of multiple candidate venue classification models in an embodiment of the present disclosure.

[0027] Figure 4 The flowchart of training the place classification model in the embodiment of the present disclosure is schematically shown.

[0028] Figure 5 A schematic diagram of the process of performing site assessment in an infectious disease environment in an embodiment of the present disclosure is shown.

[0029] Figure 6A block diagram of a place classification device in an embodiment of the present disclosure is schematically shown.

[0030] Figure 7 A block diagram of an electronic device in an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0031] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as being limited to the examples set forth herein; on the contrary, these embodiments are provided so that the present disclosure will be more comprehensive and complete, and the concepts of the example embodiments are fully conveyed to those skilled in the art. The described features, structures, or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced while omitting one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known technical solutions are not shown or described in detail to avoid obscuring various aspects of the present disclosure.

[0032] In addition, the accompanying drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.

[0033] In the related art, when managing places, it is generally unified management or differentiated according to the type of place to determine several types of places that need to be focused on. In addition, it can also be scored according to multiple indicators, and when a place meets multiple indicators, it will be given priority management.

[0034] In order to solve the technical problems in the related art, a place classification method is provided in the embodiment of the present disclosure, which can be executed by a server or a client. The place classification method can be applied to any environment state and any type of place for management or attention application scenario.

[0035] Next, refer to Figure 1 The figure shows a detailed description of the place classification method in the embodiment of the present disclosure.

[0036] In step S110, a place to be evaluated is obtained.

[0037] In the embodiments of the present disclosure, the place to be evaluated may be any place. The type of the place to be evaluated may be determined according to the application scenario or actual needs, and may be various types of indoor places or outdoor places, etc. The type of the place may be, for example, a market, a medical institution, a public service place, etc.

[0038] In some embodiments, the place to be evaluated in the environmental state can be obtained. The environmental state can be the environmental state corresponding to the place to be evaluated, that is, the environmental state at the geographical location corresponding to the place to be evaluated, and the environmental state is specifically used to indicate whether any type of infectious disease exists in the environment. The environmental state can be a first type of state and a second type of state. The first type of state can be an environmental state where an infectious disease exists, and the second type of state can be an environmental state where no infectious disease exists, which can be determined based on the medical treatment data of each medical institution and the health status of the personnel obtained by other platforms.

[0039] In step S120, the impact index of the site to be evaluated is determined, and characteristic data of the site to be evaluated is generated based on the impact index.

[0040] In the disclosed embodiment, the influencing index refers to a factor associated with the level information of the place to be evaluated. In order to improve the compatibility with the environment, the influencing index can be determined based on the environmental state. Different environmental states may have different corresponding influencing indicators. Specifically, when the environmental state is a first type of state, the objective environmental indicators of each place to be evaluated, the case indicators of infectious disease patients, and the place personnel indicators of the reference personnel in the place to be evaluated can be used together as the influencing indicators of the place to be evaluated.

[0041] Among them, the objective environmental indicators of the venue refer to the environmental characteristics of the venue to be evaluated, which may specifically include the fixed characteristics of the venue itself, as well as the variable characteristics that can be adjusted due to the people in the venue. Fixed characteristics may include, for example, at least the type of venue, the area of ​​the venue, the height of the venue, whether it is ventilated, whether there are hand washing facilities, whether there are portable disinfection facilities, and the frequency of disinfection of the venue. Variable characteristics may be the protection of the personnel in the venue, etc. In addition, case indicators of infectious disease patients can also be obtained. Case indicators refer to factors related to infectious diseases possessed by infectious disease patients. For example, case indicators may specifically include at least: the interval between onset of cases, the health status of cases, the first viral load detected by cases, the wearing of protective items by cases, and the length of stay of cases.

[0042] Reference personnel refer to other persons in the places involved in the activity trajectory of infectious disease patients. For a certain place to be evaluated, reference personnel refer to persons who have contact with infectious disease patients in the place to be evaluated. The place personnel index may include at least the movement characteristics and behavior characteristics of the reference personnel. The movement characteristics may include but are not limited to the frequency of mutual contact, the length of time people stay in the place, the protective measures taken, and the flow of personnel. In the embodiment of the present disclosure, the place personnel index is taken as the flow of personnel in the place to be evaluated as an example for explanation.

[0043] In the disclosed embodiment, after determining the influencing index, multiple data sources can be obtained, and data corresponding to the influencing index can be extracted from the multiple data sources. Specifically, when the environmental state is the first type of state, the case characteristics of the infectious disease patient, the activity trajectory of the infectious disease patient within a preset time before the onset of illness, and the corresponding health status, the protection situation taken, and the contact with other reference persons can be obtained based on the epidemiological survey report. By consulting the on-site epidemiological survey report and visiting the site, the site characteristics of the places involved in the activity trajectory of the infectious disease patient before the onset of illness are collected: including the type of place, the area of ​​the place, the ventilation of the place, etc. The movement characteristics of other reference persons in the place involved in the activity trajectory of the infectious disease patient can be obtained from the monitoring data of the place to be evaluated through artificial intelligence machine vision and face recognition algorithms as the place personnel indicators of the reference personnel, or the personnel flow of the place to be evaluated can be directly used as the place personnel indicators of the reference personnel. Specifically, the personnel flow can be determined by combining the code scanning data of the place to be evaluated with consumption data and other information.

[0044] When the environmental state is of the second type, that is, in a normalized environment where there is no infectious disease, the objective environmental indicators corresponding to each place to be evaluated can be used as its corresponding impact indicators without considering case indicators related to infectious diseases and site personnel indicators.

[0045] In the embodiment of the present disclosure, when the environmental state is the first type, by obtaining case indicators of infectious disease patients, objective environmental indicators of the venue, and venue personnel indicators of reference personnel, the impact indicators of the venue can be obtained from multiple dimensions, which can improve the comprehensiveness and accuracy of the impact indicators of the venue. By selecting different impact indicators according to the environmental state to evaluate the venue, the venue under different environmental states can be accurately evaluated.

[0046] It should be noted that before obtaining the impact indicators, multiple optional impact indicators can be determined based on historical infectious disease data, and the above impact indicators can be further screened from the optional impact indicators.

[0047] Among them, the Delphi method or other scoring methods can be used for scoring and screening, and the Delphi method is used as an example for explanation here. The steps of the Delphi method may include: given rules, such as the scores of each indicator when it is located in different positions, etc., according to the sum of the score of each position and the number of experts, and the sum of the total number of experts and the sum of all scores, determine the ratio, so as to determine the weight or proportion of any indicator, so as to determine the indicator with a weight greater than the weight threshold as the influencing indicator. In the embodiment of the present disclosure, the final influencing indicator can be shown in Table 1. The corresponding influencing indicator can be obtained according to the determined primary indicator and secondary indicator.

[0048] Table 1

[0049]

[0050]

[0051] After obtaining the impact indicators, the impact indicators may be preprocessed to improve accuracy because the collected data may contain missing data or inappropriate dimensions. The preprocessing may include filling in missing impact indicators and classifying and transforming the impact indicators.

[0052] If there is a missing in any influencing indicator, the missing influencing indicator can be filled. Specifically, the influencing indicator can include numerical influencing indicators and non-numerical influencing indicators. Numerical influencing indicators refer to influencing indicators that can be expressed by a certain numerical value, and non-numerical influencing indicators refer to influencing indicators that cannot be expressed by numerical values. Different methods can be used to fill in numerical influencing indicators and non-numerical influencing indicators. Exemplarily, in response to the presence of a missing influencing indicator, and the influencing indicator is a numerical influencing indicator, the missing influencing indicator is filled with the mean value of the influencing indicator; in response to the presence of a missing influencing indicator, and the influencing indicator is a non-numerical influencing indicator, the missing influencing indicator is filled with the variable value of the highest frequency of the influencing indicator to improve the accuracy of the influencing indicator.

[0053] In addition, in order to facilitate model training, the health status of infectious disease patients in the influencing indicators can be integrated from the various health statuses shown in Table 1 into "symptomatic" and "asymptomatic", and a new variable "whether the first case has symptoms" can be added. Based on this, multiple discrete variables can be converted into one-hot encoding, which can avoid the problem of processing difficulties caused by expanding all discrete values, reduce the processing difficulty, and improve efficiency. Furthermore, the number of cases of close contacts in the venue risk metric venue can be converted into two types: "close contact turns positive" and "no close contact turns positive" to achieve preprocessing of the influencing indicators.

[0054] After the influencing indicators are preprocessed, characteristic data of the place to be evaluated can be generated based on the preprocessed influencing indicators. Specifically, after obtaining the preprocessed influencing indicators, the preprocessed influencing indicators can be numerically mapped to convert the influencing indicators into numerical values, where the numerical values ​​can be discrete values ​​or floating point values. That is, numerical influencing indicators and non-numerical influencing indicators can be converted into discrete values ​​or floating point values. For non-numerical influencing indicators, the mapping relationship between the non-numerical influencing indicators and the numerical values ​​can be determined in advance, and the influencing indicators can be converted according to the mapping relationship. For example, the place type can be converted into 1-1, 1-2, etc. according to the mapping relationship. In addition, for influencing indicators with actual numerical values, their actual numerical values ​​can be used to convert them into floating point values, or they can be converted into discrete values ​​according to whether they are greater than a certain threshold.

[0055] After converting the impact indicators into numerical values, the characteristic data of the place to be evaluated can be constructed based on the converted numerical values. Specifically, the numerical values ​​corresponding to all the impact indicators can be combined to generate a vector, and the vector is used as the characteristic data corresponding to the impact indicators. Among them, for each place to be evaluated, characteristic data for describing the place to be evaluated can be generated based on its impact indicators. Different places to be evaluated may have different corresponding characteristic data. For the same place to be evaluated, since different environmental states correspond to different impact indicators, the characteristic data generated for the same place to be evaluated under different environmental states are also different.

[0056] For the first type of environmental status, characteristic data can be obtained based on the impact indicators and the attribute characteristics of the infectious disease. Figure 2 As shown in , the influencing indicators such as the objective environmental indicators of the site, the case indicators of the infectious disease patients, and the site personnel indicators of the reference personnel can be converted to obtain corresponding numerical values, and the numerical values ​​corresponding to the influencing indicators can be further combined with the attribute characteristics of the infectious disease to form a vector with multiple dimensions to obtain characteristic data of multiple dimensions. The attribute characteristics of infectious diseases refer to the basic reproduction number of infectious diseases, which is an indicator used to reflect the transmission capacity of an infectious disease. Specifically, it refers to the number of new cases that a case can infect during its entire infection process. For the second type of environmental state, refer to Figure 2 As shown in , the impact indicators represented by the objective environmental indicators of the site can be directly converted into numerical impact indicators and non-numerical impact indicators to obtain characteristic data. Among them, the impact indicators can also be combined with the attribute characteristics of the site to be evaluated, and then the combined characteristics are numerically converted to obtain characteristic data, which is not specifically limited here.

[0057] In the disclosed embodiment, the feature data is generated by influencing indicators of multiple dimensions, which can improve the accuracy and comprehensiveness of the feature data. For different environmental states, feature data can be generated based on different dimensions, which can improve the pertinence of the feature data and the matching between the scenes.

[0058] In step S230, the characteristic data is fitted according to the trained venue classification model to obtain the indicator parameters of the infectious disease corresponding to the venue to be evaluated, and the level information of the venue to be evaluated is determined based on the indicator parameters of the infectious disease.

[0059] In the disclosed embodiment, after obtaining the characteristic data, the characteristic data of the place to be evaluated can be input into the trained place classification model, and the characteristic data can be fitted by the trained place classification model to obtain the indicator parameters of the infectious disease of the place to be evaluated. The indicator parameters of the infectious disease can be the probability of occurrence of the infectious disease in the place to be evaluated, and the probability of occurrence can be the secondary probability, that is, the probability of the infectious disease occurring again in the place to be evaluated.

[0060] In some embodiments, in order to improve accuracy, the place classification model needs to be trained to obtain a trained place classification model. For example, the place classification model can be first determined based on the sample flow survey data and multiple candidate place classification models, and then the place classification model can be further trained based on the sample data to obtain a trained place classification model.

[0061] Among them, the sample flow survey data refers to the historical flow survey data corresponding to multiple places, and the number of places can be determined according to actual needs. The candidate place classification model can be any type of model that can perform classification and identification, for example, it can include but is not limited to logistic regression LR (Logistic Regression), random forest RForest (RandomForest), gradient boosted decision tree GBDT (Gradient Boosted Decision Trees), extreme gradient boosted decision tree XGBoost (eXtreme Gradient Boosting) and support vector machine, etc. Next, the effects of multiple candidate place classification models can be compared to select the final place classification model from them. In the candidate place classification model, the characteristic value and weight of each influencing indicator and other parameters can be determined based on historical data or other reference information.

[0062] In some embodiments, multiple candidate place classification models can be trained based on sample flow survey data to obtain multiple trained candidate place classification models; further, multiple candidate place classification models are evaluated based on multiple evaluation parameters, and one candidate place classification model is selected from multiple candidate place classification models as a place classification model according to the evaluation results. Exemplarily, the sample flow survey data can be divided into a training set and a test set to train the candidate place classification model based on the training set and the test set. Specifically, the sample flow survey data can be segmented by a K-fold cross-validation method to obtain K sub-sample data; further, each of the K sub-sample data is used as a test set, and the remaining sub-sample data is used as a training set, and each candidate place classification model is trained based on the training set until training K times, so as to obtain multiple trained candidate place classification models. Among them, K can be any value, which is determined according to actual needs. In addition, each trained candidate place classification model can also be tested based on the test set to determine the accuracy of each trained candidate place classification model.

[0063] By cross-validating K times, it is ensured that each sub-sample data is verified once, and the results of K verifications are averaged to obtain the test result. The above method can train and test the model through randomly generated sub-sample data. Each sample data will be used as training or test data, which can improve accuracy.

[0064] Furthermore, the sensitivity, specificity and area parameter of the evaluation curve of each candidate place classification model can be obtained as evaluation parameters, so as to screen from multiple candidate place classification models based on the sensitivity and specificity of the trained candidate place classification model and the area parameter of the evaluation curve to obtain a place classification model. Exemplarily, the sensitivity and specificity of each trained candidate place classification model can be obtained, and the evaluation curve can be determined based on the specificity and sensitivity; the area parameter can be determined based on the evaluation curve, one or more intermediate candidate place classification models can be determined based on the area parameter, and the intermediate candidate place classification models can be screened according to the sensitivity and / or specificity to obtain a place classification model.

[0065] Among them, sensitivity (Sensitivity, also known as true positive rate, recall rate) refers to the proportion of samples that are actually positive that are judged to be positive (for example, the proportion of people who are actually sick and are judged to be sick). The sum of true positives and false negatives can be calculated, and the ratio between true positives and the summed result can be further calculated to determine the sensitivity. Among them, false negatives refer to samples that are actually positive but judged to be negative. Specificity (Specificity, also known as true negative rate) refers to the proportion of samples that are actually negative in the epidemiological survey data that are judged to be negative (for example, the proportion of people who are actually not sick and are judged to be healthy by the hospital). The sum of true negatives and false positives can be calculated, and the ratio of true negatives to the summed result can be calculated to determine the specificity. False positives refer to samples that are actually negative but are judged to be positive.

[0066] Sensitivity refers to the extent to which true positives are not ignored (with very few false negatives), while specificity refers to the extent to which true negatives are actually identified (with very few false positives). When analyzing risk points, a high recall rate is required, that is, even if low-risk points are classified as high-risk points, in order to reduce the omission of high-risk points. Therefore, it is necessary to focus on measuring sensitivity. In addition, the classification effects of multiple candidate venue classification models need to be evaluated. Specifically, an evaluation curve can be generated by sensitivity and specificity, where the ordinate can be sensitivity (true positive rate) and the abscissa can be 1-specificity (false positive rate) to generate an evaluation curve, and the evaluation curve can be a ROC (Receiver operating characteristic) curve. After determining the evaluation curve, the area parameter AUC (Area Under Curve) under the curve (that is, the curve close to the horizontal axis) can be determined based on the evaluation curve. The closer the evaluation curve is to the upper left corner, the closer the value of the area parameter AUC is to 1.0, indicating that the classification effect of the candidate venue classification model is better. For details, refer to Figure 3 Based on this, the index comparison results of multiple candidate venue classification models are shown in Table 2.

[0067] Table 2

[0068]

[0069] After obtaining the sensitivity, specificity and area parameter of the evaluation curve of multiple candidate place classification models based on Table 2, one or more intermediate candidate place classification models can be determined based on the area parameter. Specifically, all candidate place classification models whose area parameters are greater than the area threshold can be used as intermediate candidate place classification models. The area threshold can be set according to actual needs, such as 0.5 or other values. Furthermore, among the intermediate candidate place classification models, the one with the highest sensitivity can be determined as the place classification model. If the sensitivity of multiple intermediate candidate place classification models is the same, they can be screened in combination with the specificity. For example, when the specificity meets the specificity threshold, the one with the highest sensitivity can be determined as the place classification model.

[0070] For example, from the ROC curve, the area parameter AUCs of the above candidate venue classification models are all greater than 0.5 (random classification results). The area parameter AUC of GBDT is 0.78, and the classification effect is the best, followed by RForest (its area parameter AUC is 0.71), and LR has the worst effect (its area parameter AUC is 0.581). From the perspective of sensitivity, the GBDT model has the best effect, followed by XGBoost, and LR is still the worst model. However, the risk indicators of the venue do not have linear independence, so the LR model is a linear model, and the effect of the evaluation is difficult to achieve the expected effect, while the tree model through residual fitting generally achieves good results. From the perspective of task objectives, based on the perspective of venue risk analysis, sensitivity is more important, so the GBDT model can be used as the venue classification model.

[0071] After determining the venue classification model, the venue classification model can be trained based on the sample data to obtain a trained venue classification model. The sample data can be sample data for each venue, and each venue can correspond to a sample data, and then the model training is performed based on the sample data of all venues.

[0072] In some embodiments, after determining the sample data, for an environmental state of the first type, historical impact indicators of the corresponding places can be obtained from the sample data, and attribute characteristics of the infectious disease can be obtained. Based on the historical impact indicators and the attribute characteristics of the infectious disease, historical characteristic data can be determined, and the historical characteristic data can be further input into the place classification model to determine the predicted indicator parameters corresponding to the sample data, and compared with the actual indicator parameters of the sample data to determine the difference between the two; a loss function is determined based on the difference, and the place classification model is trained based on the loss function to obtain a trained place classification model.

[0073] The historical impact index can still be determined according to the scene state. For example, when the scene state is the first type state, the objective environment index of each place, the case index of the infectious disease patient, and the place personnel index of the reference person in the trajectory place corresponding to the infectious disease patient are used as the historical impact index. When the scene state is the second type state, the objective environment index of the place corresponding to the place can be used as the historical impact index.

[0074] Next, the historical impact index can be converted into a numerical value. Specifically, for the first type of state, the historical impact index can be numerically mapped to obtain the corresponding numerical value, and the numerical value corresponding to the historical impact index can be further combined with the attribute characteristics of the infectious disease to obtain historical characteristic data in multiple dimensions. Among them, the attribute characteristics of the infectious disease still refer to the basic reproduction number, which is an indicator used to reflect the transmission ability of an infectious disease.

[0075] For the second type of status, the historical impact indicators can be directly converted into numerical values ​​to obtain historical characteristic data.

[0076] Furthermore, the historical feature data can be input into the venue classification model to fit the venue classification model and obtain the prediction index parameters corresponding to the sample data. The index parameters can be used to represent the probability of subsequent outbreak of infectious diseases, that is, the probability that the infectious disease will occur in any venue in the future.

[0077] At the same time, the predicted indicator parameters can be compared with the actual indicator parameters of the sample data to obtain the difference between the two. For sample data, the cases appearing in the sample data and the intergenerational transmission between people who come into contact with infectious disease patients can be determined based on the results of epidemiological surveys and gene sequencing methods to determine the molecular biological source of the infectious disease, and further determine the actual indicator parameters of the sample data based on the molecular biological source. The actual indicator parameter refers to the actual secondary probability. The actual indicator parameter can be the difference between the number of people who come into contact with infectious disease patients and the number of molecular biological sources, and the ratio of the number of people who come into contact with infectious disease patients in the venue.

[0078] In some embodiments, the molecular biological source can be determined by the following methods: through field epidemiological surveys and case specimen gene sequencing, the intergenerational transmission chain of infectious diseases in patients with infectious diseases and related personnel in contact with them is determined. Samples are collected using inactivated virus sampling tubes and sent to the laboratory for viral nucleic acid detection within 4 hours. A magnetic bead method nucleic acid extraction kit is used for nucleic acid extraction, and the remaining unextracted specimens are promptly packaged and placed at -80°C for long-term storage. The nucleic acid is detected for viral nucleic acid using the fluorescent quantitative PCR method, and the CT values ​​of the two targets (viral ORF1ab gene and N gene detection) in the specimen and the positive control are recorded at the same time. According to the preliminary screening results, specimens with lower CT values ​​are selected for viral nucleic acid extraction and purification using silica gel membrane filtration to obtain viral nucleic acid samples with higher purity. Using the viral whole genome targeted amplification kit (QIAGEN, QIAseq SARS-CoV-2 Primer Panel), cDNA synthesis and target fragment targeted amplification are performed on the viral nucleic acid sample, and the amplified product is purified, recovered and quantified. The recovered products were end-repaired, Y-type adapters were added, and the library was amplified using a library construction kit (QIAGEN, QIAseq FXDNA Library UDI-A Kit). The obtained library was purified and recovered by magnetic beads and accurately quantified. After quantitative dilution and sufficient denaturation of the purified library, a kit with a suitable flux was selected, loaded into the ILLUMINA MINISEQ gene sequencer, the program was set, and the library sequence was read on the machine. After confirming that the data volume of each sample library was normal, the obtained data was copied to a dedicated data analysis computer for analysis; the obtained raw data was evaluated for quality, data repaired, referenced, and spliced ​​using a professional data analysis software package (QIAGEN, QIAGEN CLC genomics WB, deskstop plus), and the viral genome sequence was preliminarily obtained. The local software package and online analysis tool were used to calculate the integrity of the obtained genome data, the location of the missing region, the single-site nucleotide mutation, and the amino acid substitution. The single-nucleotide site mutation was used to build a database in combination with the reference strain sequence, and the gene evolution tree was reconstructed to obtain the molecular biological origin of the virus in the sample.

[0079] The accurate calculation of the number of cases of reference personnel associated with the activity places of cases with infectious diseases has a great impact on the evaluation results of the places. In the related technology, the investigation results of the epidemiological survey personnel are generally used for judgment, which has certain limitations. Based on this, in the embodiments of the present disclosure, the gene sequencing results between cases are combined to assist in the judgment based on the epidemiological survey results, which effectively improves the accuracy of the judgment.

[0080] After determining the difference between the predicted index parameters and the actual index parameters of the sample data, the loss function can be determined based on the difference between the two, and then the model parameters of the place classification model can be adjusted with the minimum loss function as the goal, so as to iteratively train the place classification model and obtain the trained place classification model.

[0081] It should be noted that, for the first type of status and the second type of status, the historical feature data can be determined based on the corresponding historical influencing factors, and then the place classification model corresponding to the first type of status can be trained, as well as the place classification model corresponding to the second type of status can be trained to improve the accuracy of model training.

[0082] Figure 4 The flowchart for model training is shown schematically in Figure 4 As shown in , it mainly includes the following steps:

[0083] In step S402, a venue classification model based on machine learning, namely, a venue risk amplifier model, is obtained;

[0084] In step S404, case indicators of infectious disease patients are collected through the infectious disease reporting system, on-site epidemiological survey forms, and investigation reports; case indicators include, for example, onset time, clinical symptoms, time of symptom onset, viral load, etc., and activity trajectory of the case within a preset time period before onset and health status during activities, protection measures taken, and contact with close contacts, etc.;

[0085] In step S406, the site indicators involved in the activity trajectory of the infectious disease patients within a preset time period are analyzed by analyzing the epidemiological survey report and collecting on-site; the site indicators include, for example, the site type, site area, site ventilation conditions, etc.;

[0086] In step S408, through machine vision, face recognition technology and survey results, site personnel indicators such as the frequency of mutual contact, mobility, protection, length of stay, and personnel flow of reference personnel in the site are collected;

[0087] In step S410, the site risk impact indicators are sorted out to construct features for quantifying site risks;

[0088] In step S412, a feature value table is configured;

[0089] In step S414, the intergenerational transmission chain of the case and its related persons is determined through on-site epidemiological investigation and basic case sequencing;

[0090] In step S416, the indicator parameter of the infectious disease is selected as the result variable;

[0091] In step S418, the attribute characteristics of the infectious disease are determined;

[0092] In step S420, a trained machine learning-based place classification model is obtained;

[0093] In step S422, the place level information is calculated using the trained machine learning-based place classification model.

[0094] In the disclosed embodiment, historical feature data is determined by obtaining historical impact indicators of multiple dimensions and attribute characteristics of infectious diseases from sample data, and then model training is performed based on the historical feature data, which can improve the accuracy and comprehensiveness of the model.

[0095] In the disclosed embodiment, after determining the trained place classification model, all influencing indicators of the place classification model can be analyzed to determine the feature importance of each influencing indicator of the trained place classification model. Referring to Table 3, it can be seen that the first CT (min) of the first case, the scan code record query, the mask wearing condition of the first case and the stay time (hours) of the first case are the indicator independent variables with high feature importance in these four models. Since the original data of the ventilation level only distinguishes between indoor and outdoor environments, a large amount of point data are all indoor environments. Therefore, the classification effect is not obvious.

[0096] Table 3

[0097]

[0098]

[0099] In the disclosed embodiment, the impact of the places visited by infectious disease patients on secondary transmission is determined through the characteristics of the places to assess the risk level of the places. It can be seen that the first CT (min) of the first case in the epidemic-related places, the number of scan code record inquiries, the wearing of masks by the first case and the stay time (hours) of the first case are important indicators of epidemic control in the epidemic-related places. The technical solution in the disclosed embodiment provides a basis for assessing the risk of secondary transmission in the place, and can assist disease control and prevention workers in determining prevention and control priorities when resources are tight.

[0100] In the disclosed embodiment, after obtaining the trained place classification model, the obtained feature data can be input into the trained place classification model for fitting to obtain indicator parameters of infectious diseases.

[0101] In some embodiments, the indicator parameters of infectious diseases can be determined based on all or part of the feature data. Exemplarily, the target feature data can be determined from the feature data based on the attribute characteristics of the feature data and one or more of the prediction requirements. The target feature data can be all feature data or part of the feature data determined based on the feature importance. The attribute feature can be the difficulty of acquisition or other types of features. For example, if there are many missing feature data, the attribute feature of the difficulty of acquisition is used as an example for explanation.

[0102] For example, when the difficulty of acquiring feature data is the first level or the prediction requirement is the first degree, it can be considered that the data screening condition is met, and all feature data can be used as target feature data; when the difficulty of acquiring feature data is the second level or the prediction requirement is the second degree, it can be considered that the data screening condition is not met, and in this case, feature importance can be used to screen part of the feature data as target feature data. Among them, the difficulty of the first level is lower than the difficulty represented by the second level; the prediction requirement of the first degree is higher than the prediction requirement represented by the second degree, and the prediction requirement can be, for example, accuracy.

[0103] After determining the target feature data, the target feature data can be input into the trained site assessment model to obtain the indicator parameters of the infectious disease in the site to be assessed, that is, the probability of occurrence. It should be noted that one or more sites to be assessed can be predicted at the same time to improve operational efficiency.

[0104] Based on this, for the first type of state, the characteristic data can be obtained based on the impact index composed of the objective environment index of each place to be evaluated, the case index of the infectious disease patient and the place personnel index of the reference personnel in the place to be evaluated, and the attribute characteristics of the infectious disease converted into a numerical value. Based on this, the indicator parameters of the infectious disease in each place to be evaluated under the infectious disease environment can be obtained based on the characteristic data.

[0105] For the second type of status, the characteristic data can be obtained by converting the impact index represented by the objective environmental index of the venue into a numerical value. Based on this, the characteristic data determined by the objective environmental index of the venue can be directly input into the model to obtain the indicator parameters of infectious diseases in each venue to be evaluated under a normalized environment.

[0106] Based on the determined indicator parameters of infectious diseases in each place to be evaluated, the level information of the place to be evaluated can be determined. The indicator parameter refers to the probability of secondary outbreak of infectious diseases, and the indicator parameter is positively correlated with the level information. That is, the larger the indicator parameter, the higher the level information.

[0107] Figure 5 A flow chart for conducting a site assessment in an infectious disease setting is shown schematically in Figure 5As shown in , it mainly includes the following steps:

[0108] In step S502, the impact index of the place to be evaluated is obtained;

[0109] In step S504, the attribute characteristics of the infectious disease are obtained;

[0110] In step S506, characteristic data is obtained based on the influencing index and the attribute characteristics;

[0111] In step S508, target feature data is selected from the feature data;

[0112] In step S510, the target feature data is input into the trained venue classification model to obtain the indicator parameters of the infectious disease;

[0113] In step S512, the level information of the place is determined based on the index parameters of the infectious disease.

[0114] In the disclosed embodiment, the above analysis shows that the higher the rate of people wearing masks in public places, the lower the risk of secondary infectious diseases. The first CT scan of the first case in the epidemic-related places, the number of scan code record inquiries, the wearing of masks by the first case, and the length of stay (hours) of the first case are important concerns for place management.

[0115] The technical solution in the disclosed embodiment determines the indicator parameters of the infectious disease of the place to be evaluated according to the characteristic data, and can quickly and accurately determine the level information of the place to be evaluated, so that the prevention and control priority of the place to be evaluated can be determined based on the level information of the place to be evaluated. For the infectious disease environment, it can accurately determine the places that need to be controlled first under limited resources, improve operational efficiency and accuracy, and avoid the limitations in the relevant technology. For the normalized environment, it can quickly distinguish the places that need to be focused on, which is of guiding significance for the place management. It avoids the limitation that all places can only be managed in the same way in the relevant technology, improves the comprehensiveness and accuracy, and also improves the pertinence. In addition, since the characteristic data can be predicted to determine the level information of each place, the versatility is improved and the scope of application is increased. It provides a basis for evaluating the risk of secondary transmission in the place, and can assist disease control personnel in determining the priority of place management when medical resources are insufficient, thereby improving the accuracy of place management.

[0116] The present disclosure provides a place classification device, referring to Figure 6 As shown in , the place classification device 600 may include:

[0117] A site acquisition module 601 is used to acquire a site to be evaluated;

[0118] An index determination module 602 is used to determine the impact index of the site to be evaluated, and generate characteristic data of the site to be evaluated based on the impact index;

[0119] The level information determination module 603 is used to fit the characteristic data according to the trained place classification model to obtain the indicator parameters of the infectious disease corresponding to the place to be evaluated, and determine the level information of the place to be evaluated through the indicator parameters of the infectious disease.

[0120] In an exemplary embodiment of the present disclosure, determining the impact index of the place to be evaluated includes: determining the impact index of the place to be evaluated based on the environmental state corresponding to the place to be evaluated; wherein the environmental state includes a first type of state or a second type of state; determining the impact index of the place to be evaluated based on the environmental state corresponding to the place to be evaluated includes: if the environmental state is the first type of state, using the objective environmental indicators of each place to be evaluated, the case indicators of infectious disease patients, and the place personnel indicators of reference personnel existing in the place to be evaluated as the impact indicators; if the environmental state is the second type of state, using the objective environmental indicators of each place to be evaluated as the impact indicators.

[0121] In an exemplary embodiment of the present disclosure, generating the characteristic data of the place to be evaluated based on the impact indicator includes: numerically mapping the impact indicator based on the type of the impact indicator to obtain the characteristic data of the place to be evaluated.

[0122] In an exemplary embodiment of the present disclosure, the characteristic data is fitted according to the trained place classification model to obtain the indicator parameters of the infectious disease, including: determining the target characteristic data according to one or more of the attribute characteristics or prediction requirements of the characteristic data, and fitting the target characteristic data based on the trained place classification model to obtain the indicator parameters.

[0123] In an exemplary embodiment of the present disclosure, determining the target feature data based on one or more of the attribute characteristics or predicted requirements of the feature data includes: if the attribute characteristics or predicted requirements of the feature data meet the data screening conditions, determining all feature data as the target feature data; if the attribute characteristics or predicted requirements of the feature data do not meet the data screening conditions, screening all feature data based on feature importance to determine the target feature data.

[0124] In an exemplary embodiment of the present disclosure, the device also includes: a place classification model determination module, used to obtain sample flow survey data, and determine a place classification model from multiple candidate place classification models based on the sample flow survey data; a model training module, used to train the place classification model to obtain the trained place classification model.

[0125] In an exemplary embodiment of the present disclosure, the method of determining a place classification model from multiple candidate place classification models based on sample epidemic survey data includes: segmenting the sample epidemic survey data to obtain K sub-sample data, using each of the K sub-sample data as a test set, and using the remaining sub-sample data as a training set, training multiple candidate place classification models through the training set to obtain multiple trained candidate place classification models; evaluating the multiple trained candidate place classification models based on multiple evaluation parameters, and determining the place classification model according to the evaluation results.

[0126] In an exemplary embodiment of the present disclosure, the multiple trained candidate place classification models are evaluated based on multiple evaluation parameters, and the place classification model is determined according to the evaluation results, including: obtaining the sensitivity and specificity of each trained candidate place classification model, and determining an evaluation curve based on the specificity and the sensitivity; determining one or more intermediate candidate classification models based on the area parameter of the evaluation curve, and screening the intermediate candidate classification models according to sensitivity and / or specificity to obtain the place classification model.

[0127] In an exemplary embodiment of the present disclosure, the training of the place classification model to obtain the trained place classification model includes: obtaining sample data, obtaining historical impact indicator data from the sample data, and obtaining historical feature data based on the historical impact indicator data; inputting the historical feature data into the place classification model to obtain prediction indicator parameters corresponding to the sample data, and comparing the predicted indicator parameters with the actual indicator parameters of the sample data to determine the difference between the two; determining a loss function based on the difference, and training the place classification model based on the loss function to obtain the trained place classification model.

[0128] It should be noted that the specific details of each module in the above-mentioned place classification device have been described in detail in the corresponding place classification method, so they will not be repeated here.

[0129] The exemplary embodiment of the present disclosure also provides an electronic device. The electronic device may be the above-mentioned client or a server. Generally, the electronic device may include a processor and a memory, the memory is used to store executable instructions of the processor, and the processor is configured to execute the above-mentioned place classification method by executing the executable instructions.

[0130] Refer to the following Figure 7 An electronic device 700 according to this embodiment of the present disclosure is described. Figure 7 The electronic device 700 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0131] like Figure 7 As shown, the electronic device 700 is in the form of a general computing device. The components of the electronic device 700 may include, but are not limited to: the at least one processing unit 710, the at least one storage unit 720, a bus 730 connecting different system components (including the storage unit 720 and the processing unit 710), and a display unit 740.

[0132] The storage unit stores program codes, which can be executed by the processing unit 710, so that the processing unit 710 performs the steps according to various exemplary embodiments of the present disclosure described in the above “Exemplary Method” section of this specification. For example, the processing unit 710 can perform the following steps: Figure 2 Follow the steps shown in .

[0133] The storage unit 720 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 7201 and / or a cache storage unit 7202 , and may further include a read-only storage unit (ROM) 7203 .

[0134] The storage unit 720 may also include a program / utility 7204 having a set (at least one) of program modules 7205, such program modules 7205 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 may include an implementation of a network environment.

[0135] Bus 730 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, a graphics acceleration interface, a processing unit, or a local bus using any of a variety of bus architectures.

[0136] The electronic device 700 may also communicate with one or more external devices 800 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 700, and / or communicate with any device that enables the electronic device 700 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 750. Furthermore, the electronic device 700 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 760. As shown, the network adapter 760 communicates with other modules of the electronic device 700 via a bus 730. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 700, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0137] In an embodiment of the present disclosure, a computer-readable storage medium is also provided, on which a program product capable of implementing the above method of the present specification is stored. In some possible implementations, various aspects of the present disclosure can also be implemented in the form of a program product, which includes a program code, and when the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps according to various exemplary implementations of the present disclosure described in the above "Exemplary Method" section of the present specification.

[0138] According to the program product for implementing the above method in the embodiment of the present disclosure, it can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can be run on a terminal device, such as a personal computer. However, the program product of the present disclosure is not limited thereto. In this document, a readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, an apparatus or a device.

[0139] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with 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 disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0140] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, in which readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Readable signal media may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0141] The program code contained on the readable medium can be transmitted with any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the above. The program code for performing the disclosed operation can be written in any combination of one or more programming languages, including object-oriented programming languages-such as Java, C++, etc., and also including conventional procedural programming languages-such as "C" language or similar programming languages. The program code can be executed completely on the user computing device, partially on the user device, as an independent software package, partially on the user computing device, partially on the remote computing device, or completely on the remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect through the Internet).

[0142] In addition, the above-mentioned figures are only schematic illustrations of the processes included in the method according to the embodiment of the present disclosure, and are not intended to be limiting. It is easy to understand that the processes shown in the above-mentioned figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.

[0143] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary technical means in the art that are not disclosed in the present disclosure. The specification and examples are to be considered as exemplary only, and the true scope and spirit of the present disclosure are indicated by the claims.

[0144] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A method for classifying places, characterized in that: include: Obtaining the site to be assessed; Determining the impact index of the place to be evaluated, and generating characteristic data of the place to be evaluated based on the impact index; The characteristic data is fitted according to the trained venue classification model to obtain the indicator parameters of the infectious disease corresponding to the venue to be evaluated, and the level information of the venue to be evaluated is determined through the indicator parameters of the infectious disease.

2. The place classification method according to claim 1, characterized in that: Determining the impact indicators of the site to be evaluated includes: Determining the impact index of the place to be evaluated based on the environmental status corresponding to the place to be evaluated; Wherein, the environmental state includes a first type state or a second type state; and determining the impact index of the place to be evaluated based on the environmental state corresponding to the place to be evaluated includes: If the environmental state is of the first type, the objective environmental index of each site to be evaluated, the case index of infectious disease patients, and the site personnel index of the reference personnel in the site to be evaluated are used as the impact index; If the environmental state is of the second type, the objective environmental index corresponding to each site to be evaluated is used as the impact index.

3. The place classification method according to claim 1, characterized in that: The generating the characteristic data of the place to be evaluated based on the impact index includes: The influencing indicators are numerically mapped based on the types of the influencing indicators to obtain characteristic data of the place to be evaluated.

4. The place classification method according to claim 1, characterized in that: The step of fitting the characteristic data according to the trained place classification model to obtain the indicator parameters of infectious diseases includes: The target feature data is determined according to one or more of the attribute characteristics or predicted requirements of the feature data, and the target feature data is fitted based on the trained place classification model to obtain the indicator parameters.

5. The place classification method according to claim 4, characterized in that: The determining the target feature data according to one or more of the attribute features or predicted requirements of the feature data includes: If the attribute characteristics or prediction requirements of the feature data meet the data screening conditions, all feature data are determined as the target feature data; If the attribute characteristics or prediction requirements of the feature data do not meet the data screening conditions, all feature data are screened based on feature importance to determine the target feature data.

6. The place classification method according to claim 1, characterized in that: The method further comprises: Acquire sample epidemiological survey data, and determine a venue classification model from a plurality of candidate venue classification models based on the sample epidemiological survey data; The place classification model is trained to obtain the trained place classification model.

7. The place classification method according to claim 6, characterized in that: The determining of a venue classification model from a plurality of candidate venue classification models based on the sample flow survey data includes: The sample flow survey data is segmented to obtain K sub-sample data, each of the K sub-sample data is used as a test set, and the remaining sub-sample data is used as a training set, and multiple candidate venue classification models are trained by the training set to obtain multiple trained candidate venue classification models; A plurality of trained candidate venue classification models are evaluated based on a plurality of evaluation parameters, and the venue classification model is determined according to the evaluation results.

8. The place classification method according to claim 7, characterized in that: The step of evaluating a plurality of trained candidate venue classification models based on a plurality of evaluation parameters and determining the venue classification model according to the evaluation results includes: Obtaining the sensitivity and specificity of each trained candidate site classification model, and determining an evaluation curve based on the specificity and the sensitivity; One or more intermediate candidate classification models are determined based on the area parameter of the evaluation curve, and the intermediate candidate classification models are screened according to sensitivity and / or specificity to obtain the venue classification model.

9. The place classification method according to claim 6, characterized in that: The step of training the place classification model to obtain the trained place classification model includes: Acquire sample data, acquire historical impact indicator data from the sample data, and obtain historical feature data based on the historical impact indicator data; Inputting the historical feature data into a venue classification model to obtain prediction index parameters corresponding to the sample data, and comparing the prediction index parameters with the actual index parameters of the sample data to determine the difference between the two; A loss function is determined according to the difference, and the place classification model is trained based on the loss function to obtain a trained place classification model.

10. A place classification device, characterized in that: include: A site acquisition module is used to acquire sites to be evaluated; An indicator determination module, used to determine the impact indicator of the place to be evaluated, and generate characteristic data of the place to be evaluated based on the impact indicator; The level information determination module is used to fit the characteristic data according to the trained place classification model to obtain the indicator parameters of the infectious disease corresponding to the place to be evaluated, and determine the level information of the place to be evaluated through the indicator parameters of the infectious disease.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the place classification method described in any one of claims 1 to 9 is implemented.

12. An electronic device, characterized in that: include: processor; as well as A memory, configured to store executable instructions of the processor; Wherein, the processor is configured to execute the place classification method described in any one of claims 1-9 by executing the executable instructions.

Images