CN111680830B

CN111680830B - Epidemic situation prevention method and device based on aggregation risk early warning

Info

Publication number: CN111680830B
Application number: CN202010450426.2A
Authority: CN
Inventors: 张思洁; 黄铿龙; 黄又平
Original assignee: Guangzhou Henghao Data Technology Co ltd
Current assignee: Guangzhou Henghao Data Technology Co ltd
Priority date: 2020-05-25
Filing date: 2020-05-25
Publication date: 2024-01-26
Anticipated expiration: 2040-05-25
Also published as: CN111680830A

Abstract

The application discloses an epidemic prevention method and device based on aggregation risk early warning, wherein the method comprises the following steps: acquiring communication KPI data of a target cell in a selected area in a first preset history period; and carrying out statistical processing on the communication KPI data of the target cell to obtain characteristic data of the target cell, inputting the characteristic data into a target passenger flow volume prediction model to obtain predicted passenger flow volume of the target cell at a future appointed moment, judging whether the predicted passenger flow volume of the target cell at the appointed moment exceeds a first threshold value, and if so, executing a first alarm operation to remind a target object of epidemic prevention. The method can realize the purpose of judging the aggregation risk in advance and informing related personnel, so that the related personnel can take measures in advance to block the occurrence of the aggregation risk, reduce the epidemic of infectious diseases and reduce harm.

Description

Epidemic situation prevention method and device based on aggregation risk early warning

Technical Field

The application relates to the technical field of artificial intelligence, in particular to an epidemic prevention method and device based on aggregation risk early warning.

Background

With the development of society and traffic technology, the flow of people between different regions or areas is more frequent, and the flow scale of people is also more and more large. The flow of people can drive social development and also cause people to gather, thus posing some risk of gathering. For example, in the presence of infectious or epidemic diseases, the flow and aggregation of people can accelerate the spread speed of the diseases and expand the spread range of the diseases, so that the difficulty of disease treatment is correspondingly increased, and the public health and human health are increasingly challenged.

Therefore, it is needed to provide a scheme capable of timely early warning the aggregation risk so as to remind related personnel to take precautionary measures in time.

Disclosure of Invention

The embodiment of the application provides an epidemic situation prevention method and device based on aggregation risk early warning so as to timely early warn the aggregation risk and remind relevant personnel to take measures for prevention.

In a first aspect, an embodiment of the present application provides an epidemic prevention method based on aggregation risk early warning, where the method includes:

acquiring communication KPI data of a target cell in a selected area in a first preset history period;

carrying out statistical processing on the communication KPI data of the target cell to obtain the characteristic data of the target cell;

Inputting the characteristic data of the target cell into a target passenger flow volume prediction model to obtain the predicted passenger flow volume of the target cell at a future appointed moment, wherein the target passenger flow volume prediction model is trained based on the historical characteristic data of at least one cell and a first preset algorithm, the historical characteristic data of the at least one cell is obtained by carrying out statistical processing on the historical communication KPI data of the at least one cell, and the type of the at least one cell is the same as the type of the target cell;

judging whether the predicted passenger flow quantity of the target cell at the appointed moment exceeds a first threshold value, if so, executing a first alarm operation to remind a target object of epidemic prevention, wherein the first alarm operation is used for informing the target object that the target cell has aggregation risk at the appointed moment.

In a second aspect, an embodiment of the present application further provides an epidemic prevention device based on aggregation risk early warning, where the device includes:

the first data acquisition module is used for acquiring communication KPI data of a target cell in a selected area in a first preset history period;

the first data processing module is used for carrying out statistical processing on the communication KPI data of the target cell to obtain the characteristic data of the target cell;

The passenger flow prediction module is used for inputting the characteristic data of the target cell into a target passenger flow prediction model to obtain the predicted passenger flow of the target cell at a future appointed moment, wherein the target passenger flow prediction model is trained based on the historical characteristic data of at least one cell and a first preset algorithm, the historical characteristic data of the at least one cell is obtained by carrying out statistical processing on the historical communication KPI data of the at least one cell, and the type of the at least one cell is the same as the type of the target cell;

the first alarm module is used for executing a first alarm operation when the predicted passenger flow volume of the target cell at the appointed moment exceeds a first threshold value so as to remind the target object of epidemic prevention, wherein the first alarm operation is used for informing the target object that the target cell has aggregation risk at the appointed moment.

According to the at least one technical scheme adopted by the embodiment of the application, the predicted passenger flow of the target cell at a certain moment in the future can be predicted in advance by utilizing the communication KPI data of the target cell and the target passenger flow prediction model obtained through training, and when the predicted passenger flow of the cell at the moment exceeds the set threshold value, alarm information is sent to the target object to give an alarm, so that the purposes of judging the aggregation risk in advance and informing related personnel can be realized, the related personnel can take measures in advance to block the aggregation risk, the epidemic of infectious diseases is reduced, and the harm is reduced.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute an undue limitation to the application. In the drawings:

fig. 1 is a flow chart of an epidemic prevention method based on aggregate risk early warning according to an embodiment of the present application.

Fig. 2 is a flow chart of an epidemic prevention method based on aggregation risk early warning according to another embodiment of the present application.

Fig. 3 is a flow chart of an epidemic prevention method based on aggregate risk early warning according to another embodiment of the present application.

Fig. 4A is a schematic structural view of the inner, middle and outer layers of the aggregation zone provided in the embodiments of the present application.

Fig. 4B is another schematic structural view of the inner, middle and outer layers of the accumulation zone provided in the embodiments of the present application.

Fig. 5A is a schematic diagram of an alarm effect of an aggregation area (activity center) according to an embodiment of the present application.

Fig. 5B is a cell list associated with the aggregation area shown in fig. 5.

Fig. 6 is a schematic structural diagram of an epidemic prevention device based on aggregation risk early warning according to an embodiment of the present application.

Fig. 7 is a schematic structural diagram of an epidemic prevention device based on aggregation risk early warning according to another embodiment of the present application.

Fig. 8 is a schematic structural diagram of an epidemic prevention device based on aggregation risk early warning according to another embodiment of the present application.

Fig. 9 is a system architecture diagram of an epidemic prevention scheme based on aggregate risk early warning in practical application according to an embodiment of the present application.

Fig. 10 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed Description

For the purposes, technical solutions and advantages of the present application, the technical solutions of the present application will be clearly and completely described below with reference to specific embodiments of the present application and corresponding drawings. It will be apparent that the described embodiments are only some, but not all, of the embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.

In order to timely make a scheme of early warning on the aggregation risk so as to remind related personnel to take precautionary measures in time and block epidemic situation diffusion and transmission, the embodiment of the application provides an epidemic situation precaution method and device based on the aggregation risk early warning. In other words, the method may be performed by software or hardware installed at a terminal device or a server device. The service end includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, and the like. As shown, the method may include the following steps.

The epidemic situation prevention method and device based on aggregation risk early warning provided by the embodiment of the application have the technical conception that: for a selected area (which may contain a plurality of cells (distinguishable by base stations), such as for a city, which may include thousands of cells), at least one of a traffic prediction model, a cluster model, and an activity time prediction model is first trained using at least one of historical communications KPI data, historical user location information data, and industrial parameter data for the cells; and then, based on corresponding data and corresponding models of the cells in a period of time which is closest to the current moment, predicting relevant aggregation information of the cells, and alarming according to a prediction result to remind relevant personnel to take blocking or precautionary measures in time so as to prevent epidemic situation diffusion, for example, when the passenger flow quantity of a certain cell exceeds a set safety threshold value when the passenger flow quantity prediction model predicts that the passenger flow quantity of the certain cell exceeds a set safety threshold value at a certain moment in the future, the cell is highlighted and alarming is triggered, and the relevant personnel are notified through front-end display or through mail and short message modes, so that the relevant personnel can know the situation more quickly.

The communication KPI data of a cell refers to base station performance evaluation index data corresponding to the cell, such as a maximum radio resource control (Radio Resource Control, RRC) connection user number (which can be understood as how many users the base station has), communication quality parameters, and the like; the user position information data in one cell is derived from signaling data and/or GPS data of an operator; the engineering parameter data of a cell is also called engineering parameter data of a base station corresponding to the cell, and the engineering parameter data comprise data such as base station ID, azimuth angle, inclination, longitude and latitude (base station position) and the like of the base station.

The above passenger flow prediction model, the cluster model and the activity time prediction model may be obtained through deep learning or other machine learning training, and will be described in detail in the following embodiments, which are not repeated here.

The epidemic prevention scheme based on the aggregation risk early warning provided by the embodiments of the application is described in detail below with reference to the accompanying drawings.

Example 1

Fig. 1 shows a flow chart of an epidemic prevention method based on aggregate risk early warning according to an embodiment of the present application, as shown in fig. 1, the method may include the following steps:

step 101, start.

Step 102, acquiring communication KPI data of a target cell in a selected area in a first preset history period.

The selected area may be any area where aggregate risk prediction is desired, such as a city, a region in a city, and so on. The selected area may include a plurality of cells (base stations), and the target cell may be all or part of the cells in the selected area.

The first preset history period may be a history period nearest to the current time, the length of the period may be set as required, the period may or may not include the current time, and the period may or may not be continuous, such as a year, a month, a week, a last ten days of a month, a working day of a week, and so on. For example, assuming that the current time is 12 pm on day 20 of month 5 of 2020, the first preset history period is one week and continuous in length, the first preset history period may be a period of 12 pm on day 13 of month 5 of 2020 to 12 pm on day 20 of month 5 of 2020.

In this embodiment, the communication KPI data may be a maximum number of RRC connection users, and the maximum number of RRC connection users in a cell may reflect a traffic volume of the cell.

In practical application, the communication KPI data of the target cell may be stored in a preset database, and then these data are read from the database. The preset database can be a relational database or a non-relational database.

And 103, carrying out statistical processing on the communication KPI data of the target cell to obtain the characteristic data of the target cell.

Specifically, after the communication KPI data of the target cell in the first preset history period is obtained, preprocessing (such as cleaning dirty data, removing abnormal data, filling missing values, etc.) may be first performed, and then feature engineering (statistical analysis) is performed to obtain feature data of the target cell.

When there are a plurality of target cells, it may be assumed that the communication KPI data of the target cell acquired in step 102 is shown in table 1, and the obtained feature data after the preprocessing and the feature engineering are shown in table 2.

TABLE 1

TABLE 2

In table 2, the maximum RRC connection user number of the previous day, the daily average value of the recent maximum RRC connection user number, the standard deviation of the recent maximum RRC connection user number, the time variable (hours), and the like are characteristics of the target cell.

And 104, inputting the characteristic data of the target cell into a target passenger flow volume prediction model to obtain the predicted passenger flow volume of the target cell at the future appointed moment.

The target passenger flow volume prediction model is trained based on historical feature data of at least one cell and a first preset algorithm, wherein the historical feature data of the at least one cell is also obtained by carrying out statistical processing on historical communication KPI data of the at least one cell, and the type of the at least one cell is the same as that of the target cell.

The first preset algorithm may include one of the following algorithms: XGboost, lightweight gradient lifting (Light Gradient Boost & Adaboost, GBM), gated loop units (Gated Recurrent Unit, GRU), and Long-Short-Term-memory networks (LSTM).

Since in a selected area, typically comprising thousands of cells, training a traffic prediction model for each cell would result in too much training effort and therefore need to be avoided. The applicant has found that in a selected area, some cells often have commonalities, such as schools, hospitals, shops, etc., so that the cells can be classified and then a traffic prediction model can be trained for one cell type, which can greatly reduce the model training effort. Specifically, the plurality of cells in a selected area can be classified based on a clustering algorithm and according to the similarity of the traffic variation trend of the cells, for example, 50 types of cells in a city are clustered, and then training 50 traffic prediction models at most is enough. The target passenger flow volume prediction model is a passenger flow volume prediction model corresponding to the cell type of the target cell.

It can be understood that the characteristic data of the cell adopted by the target passenger flow volume prediction model is mutually corresponding to the characteristic data of the target cell input when the target passenger flow volume prediction model is applied, and which kind of characteristics are adopted during training, and which kind of characteristics are also input during application.

The specified time may be determined based on a predicted time step used during training, and assuming that the predicted time step used during training is 1 hour and the current time is 12 pm on day 20 of month 5 in 2020, the specified time is 13 pm on day 20 of month 5 in 2020. The predicted time step used in training can be set according to the service requirement, and if the service requirement is to reasonably plan according to the change trend of the future passenger flow, the predicted time step can be set to be larger, and can generally reach 1-3 months.

In general, the predicted time step has no direct relationship with the length of the history period corresponding to the history data used during training, but if the predicted time step is relatively long (e.g., after half a year), the length of the history period corresponding to the history data used during training should be longer, at least greater than 6 months.

The traffic volume of a cell refers to the number of people passing through the cell per unit time.

Step 105, judging whether the predicted passenger flow volume of the target cell at the appointed moment exceeds a first threshold value, if yes, executing step 106, otherwise executing step 107.

The first threshold may be a preset passenger flow volume safety threshold, and if the predicted passenger flow volume of the target cell at the designated time exceeds the threshold, it is indicated that the passenger flow volume of the cell will be abnormal at the designated time, and an alarm needs to be given to related personnel.

And 106, executing a first alarm operation to remind the target object of epidemic prevention.

The first alarming operation is used for informing the target object that the target cell has aggregation risk at the future appointed moment. The target object may be a relevant responsible person of the target cell. The first alert operation includes, but is not limited to: the method comprises the steps of pushing alarm information to a target object, highlighting the mode of displaying the cell, sending alarm sound and the like on a monitoring system of the target cell, and calling other early warning functions to remind the target object that the target cell has aggregation risk at a certain moment (the designated moment) in the future, so that the aggregation propagation of infectious diseases is likely to be caused, and measures need to be taken in advance for precaution, such as assigning related personnel to evacuate personnel in the target cell, sending protective articles and the like.

In specific implementation, the warning information can be sent to the related responsible person of the target cell by sending a mail, a chart, a short message and the like, and the warning information can include the position (which can be expressed by longitude and latitude) of the target cell, the name of the target cell, the predicted passenger flow volume of the target cell at the appointed moment in the future and the like.

Step 107, end.

According to the epidemic situation prevention method based on the aggregation risk early warning, the predicted passenger flow of the target cell at a certain moment in the future can be predicted in advance by using the communication KPI data of the target cell and the target passenger flow prediction model obtained through training, and when the predicted passenger flow of the cell at the moment exceeds a set threshold value, warning information is sent to a target object to warn, so that the purposes of judging the aggregation risk in advance and informing related personnel can be achieved, the related personnel can take measures in advance to block the occurrence of the aggregation risk, the epidemic of infectious diseases is reduced, and the harm is reduced.

Example 2

Fig. 2 is a schematic flow chart of an epidemic prevention method based on aggregate risk early warning according to another embodiment of the present application, where the selected area includes a plurality of cells, and the target cell is any cell of the plurality of cells. As shown in fig. 2, the method may further include, in addition to steps 101 to 107 in fig. 1, after step 104, the steps of:

And 108, when the appointed time is reached, acquiring the real passenger flow of the cells at the appointed time.

Specifically, when the specified time is reached, the communication KPI data of the plurality of cells at the specified time can be read, so as to determine the actual passenger flows of the plurality of cells at the specified time.

Step 109, determining a plurality of abnormal cells in the plurality of cells based on the predicted passenger flow volume of the plurality of cells at the designated time and the real passenger flow volume of the plurality of cells at the designated time.

As an example, when the difference between the actual passenger flow volume of one cell in the plurality of cells at the designated time and the predicted passenger flow volume of the cell at the designated time exceeds a second threshold, the cell is determined to be an abnormal cell, and the second threshold may be a preset safety threshold of the cell at the designated time.

And 110, clustering the plurality of abnormal cells based on a target clustering model to obtain at least one abnormal cell cluster.

The target clustering model is trained based on preset position data and a preset clustering algorithm, wherein the preset position data comprises position data of a historical abnormal cell or position data of a user in the historical abnormal cell, and the position data can be longitude and latitude data.

The preset clustering algorithm comprises one of Density-based clustering algorithm (Density-Based Spatial Clustering of Applications with Noise, DBSCAN) and mean shift algorithm.

Optionally, when the target clustering model is trained, parameters such as sample size parameters, maximum distance and the like of the clusters can be set, outliers are deleted, and only more aggregated points are considered. In order to find all dense areas of the abnormal cell and regard the dense areas as cluster clusters, for example, a DBSCAN algorithm can be adopted to perform density clustering, so that noise points far from a density core can be robust, the number of cluster clusters does not need to be judged in advance, and cluster clusters with any shape can be found.

And step 111, determining the position of the target aggregation area based on the center of the at least one abnormal cell cluster.

In one mode, the center of the at least one abnormal cell cluster can be directly determined to be the position (which can be expressed by longitude and latitude) of the target aggregation area, so that the target aggregation area with the same number as the cluster is obtained.

In another manner, at least one target cluster meeting a preset condition in the at least one abnormal cell cluster can be determined, the center of the target cluster is determined to be the position of the target aggregation area, wherein the preset condition comprises at least one of that the number of cells is larger than the preset number and that the maximum distance between the cells and the cluster center is smaller than the preset distance, namely, outliers are deleted, only the relatively aggregated points are considered, and thus, the determined aggregation area is more accurate and the error of judging the non-aggregation area as the aggregation area does not occur. In other words, the density of the minimum aggregation area should have a threshold value, for example, 10, and when the traffic of people in at least 10 cells close to each other is abnormal, the aggregation situation is considered to possibly occur, so as to obtain a clustering center (longitude and latitude) and a clustering radius.

It will be appreciated that for a single aggregation activity, the traffic of cells surrounding the activity site will generally change within a period of time (e.g., 1-2 hours) before the activity begins, so if it is desired to predict the aggregation activity site (aggregation zone), cells with abnormal traffic may be first identified and clustered, and then the cluster center is used as the aggregation activity site (aggregation zone). For example, for a sports center, during the first 1 and 2 hours of the beginning of a concert, the traffic of the sports center and the surrounding cells may be abnormal, so that the cells can be found out and then clustered to find the position (target aggregation area) where aggregation activity is about to occur.

And 112, executing a second alarm operation to remind the target object of epidemic prevention.

The second alarm operation is used for informing the target object of the position of the target aggregation area.

Similar to the first alarm operation, the second alarm operation may also include, but is not limited to: pushing alarm information to target objects, highlighting the target aggregation areas and associated cells (cells in a cluster where the target aggregation areas are located) on a monitoring system in a selected area, sending out alarm sounds and other modes, calling other early warning functions and the like. In specific implementation, the warning information can be sent to the related responsible person of the target cell by sending a mail, a chart, a short message and the like, and the warning information can include the position of the target aggregation area, the position and the name of the related cell (the cell in the cluster where the target aggregation area is located) and the like, so as to remind the target object that the target aggregation area has aggregation risk, possibly cause aggregation propagation of infectious diseases, and take measures in advance to prevent, such as assigning related personnel to the target aggregation area for personnel evacuation, issuing protective articles and the like.

According to the epidemic situation prevention method based on aggregation risk early warning, the predicted passenger flow volume and the real passenger flow volume of a plurality of cells in the selected area at a certain moment in the future can be used for determining a plurality of abnormal passenger flow volume cells in the selected area, then the abnormal cells are clustered by using a clustering model, a target aggregation area which is about to occur aggregation activity is found out, and warning is carried out, so that the purpose of prejudging the aggregation area in advance and timely informing related personnel can be achieved, the related personnel can take measures in advance to block the occurrence of aggregation risk, the epidemic of infectious diseases is reduced, and harm is reduced.

Example 3

Fig. 3 is a schematic flow chart of an epidemic prevention method based on aggregate risk early warning according to still another embodiment of the present application, where the selected area includes a plurality of cells, and the target cell is any cell of the plurality of cells. As shown in fig. 3, the method may further include, in addition to steps 101 to 107 and steps 108 to 112, after step 111:

and 113, taking the target aggregation area as a center, and respectively acquiring communication KPI data of the inner layer, the middle layer and the outer layer of the target aggregation area in a second preset history period.

The inner layer of one gathering area comprises a gathering area center, the distances between the inner layer, the middle layer and the outer layer of the gathering area and the gathering area center decrease in sequence, the outer edge of the inner layer of the gathering area is tangent to the inner edge of the middle layer of the gathering area, and the outer edge of the middle layer of the gathering area is tangent to the inner edge of the outer layer of the gathering area.

Specifically, the inner layer, middle layer and outer layer of an aggregation zone may have two cases as shown in fig. 4A and 4B, in which the solid dot in the center represents the aggregation zone in fig. 4A and 4B.

In the first case shown in fig. 4A, a first distance (a), a second distance (B), and a third distance (C) are extended outward in the longitudinal and latitudinal directions with the aggregation area as the center, respectively, to obtain three squares with side lengths of 2A (abbreviated as square a), 2B (abbreviated as square B), and 2C (abbreviated as square C), where a < B < C, then the inner layer of the aggregation area is square a, the non-overlapping portion of square B and square a is the middle layer of the aggregation area, and the non-overlapping portion of square C and positive direction B is the outer layer of the aggregation area.

In the second case shown in fig. 4B, a fourth distance (D), a fifth distance (E), and a sixth distance (F) are extended outward with the aggregation area as the center, respectively, to obtain three circles D, E, and F having radii D, E and F, respectively, where D < E < F, and then the circle D is taken as the inner layer of the aggregation area, the non-overlapping portion of the circle E and the circle D is taken as the middle layer of the aggregation area, and the non-overlapping portion of the circle F and the circle E is taken as the outer layer of the aggregation area.

After determining the inner layer, middle layer and outer layer of the target aggregation zone, communication KPI data of the inner layer, middle layer and outer layer of the target aggregation zone within a second preset history period is acquired, respectively, and then step 114 is performed.

And 114, carrying out statistical processing on communication KPI data of the inner layer, the middle layer and the outer layer of the target gathering area at a plurality of moments in the second preset historical period to obtain characteristic data of the target gathering area at the plurality of moments.

As an example, the acquired KPI data may include an evolved radio access bearer (Evolved Radio Access Bearer, E-RAB) drop rate, an X2 interface handover success rate, an S1 interface handover success rate, an RRC connection success rate, and the statistical processing to be performed may include: the E-RAB dropping rate, the X2 interface switching success rate, the S1 interface switching success rate and the RRC connection success rate are median values in the inner layer, the middle layer and the outer layer of the target aggregation area, respectively, the sum of average RRC connection user numbers in the inner layer, the middle layer and the outer layer of the target aggregation area, the sum of connection user number difference values among the inner layer, the middle layer and the outer layer of the target aggregation area, and the like, and these data are feature data obtained through statistics, and these features can be regarded as explanatory variables of the following active time prediction model.

And 115, inputting the characteristic data of the target aggregation area at the multiple moments into an activity time prediction model to obtain the time when the current moment is away from the target aggregation area and aggregation activity occurs.

The activity time prediction model is trained based on characteristic data of a history aggregation area, labels and a second preset algorithm, the characteristic data of the history aggregation area are obtained by carrying out statistical processing on communication KPI data of an inner layer, a middle layer and an outer layer of the history aggregation area at a plurality of history moments, and the labels of the history aggregation area are determined according to the aggregation activity time of the history aggregation area from the plurality of history moments. It can be understood that the manner of statistically obtaining the feature data of the history aggregation area is consistent with the manner of statistically obtaining the feature data of the target aggregation area, which is not described in detail.

The labels of the history aggregation area can be understood as target variables of the activity time prediction model, and the target variables are constructed according to the time when the activity occurs from a certain historical moment, and the construction of the labels of the history aggregation area is described below with reference to table 3.

TABLE 3 Table 3

Date of day	Time	Label (target variable: hours)	Interpretation variable 1	Interpretation variable 2	…
						…	…	…	…	…	…
2019.03.30	12:00	6	…	…	…
						2019.03.30	13:00	5	…	…	…
2019.03.30	14:00	4	…	…	…
						2019.03.30	15:00	3	…	…	…
2019.03.30	16:00	2	…	…	…
						2019.03.30	17:00	1	…	…	…
2019.03.30	18:00	0	…	…	…
						2019.03.30	19:00	-1	…	…	…
2019.03.30	20:00	-2	…	…	…
						2019.03.30	21:00	-3	…	…	…
2019.03.30	22:00	-4	…	…	…

As shown in table 3, according to the history, the activity record is gathered, the activity occurrence time is found to be 18:00 of the same day, the target variable of the activity time is set to 0, the target variable of the rest time is set according to the distance from the activity occurrence time, for example, at 12 points of 30 days of 3 months of 2019, the target variable is set to 6 after the activity occurs for 6 hours; at 13 points of 3.30.2019, the distance activity occurs for 5 hours, and the target variable is set to be 5; then, at 19 points on 30 days of 3 months in 2019, the activity has occurred for 1 hour, the target variable is set to-1, and so on, and no further description is given.

The second preset algorithm may include one of the following algorithms: XGboost, lightweight gradient lifting (Light Gradient Boost & Adaboost, GBM), gated loop units (Gated Recurrent Unit, GRU), and Long-Short-Term-memory networks (LSTM).

Step 116, executing a third alarm operation to remind the target object to perform epidemic prevention.

The third alarm operation is used for informing the position of the target aggregation area of the target object and the time when the current moment is away from the target aggregation area and aggregation activity occurs.

Similar to the first and second alarm operations, the third alarm operation may also include, but is not limited to: the method comprises the steps of pushing alarm information to target objects, highlighting and displaying the target aggregation areas, the time of occurrence of distance aggregation activities and associated cells (cells in a cluster where the target aggregation areas are located) on a monitoring system of a selected area, sending out alarm sounds and other early warning functions and the like to remind the target objects of aggregation risks when the target aggregation of the target objects is likely to occur, possibly causing aggregation propagation of infectious diseases, and taking measures in advance to prevent, such as assigning related personnel to the target aggregation areas in advance for personnel evacuation, issuing protective articles and the like.

In specific implementation, the warning information can be sent to the related responsible person of the target cell by sending a mail, a chart, a short message and the like, and the warning information can include the position of the target aggregation area, the time when the distance from the target aggregation area to the aggregation activity occurs, the position and the name of the associated cell (the cell in the cluster where the target aggregation area is located), and the like.

As shown in fig. 5A, the following information may be highlighted on the monitored thermodynamic diagram of the selected region (e.g., the city of shenzhen):

the current time: 2020-02-24:00:00;

aggregating active locations (latitude and longitude of active points): [114.08052,22.54499];

predicted traffic flow (predicted number of users) corresponding to the current time: 974;

real passenger flow volume (real user number) corresponding to the current time: 1365;

distance activity time: 1.583.

optionally, as shown in fig. 5B, a list of cells associated with the aggregated activity center may be additionally presented.

Optionally, the prediction result contained in the alarm information may be stored for later reference.

According to the epidemic situation prevention method based on aggregation risk early warning, not only can the predicted passenger flow volume and the actual passenger flow volume of a plurality of cells in a selected area at a certain moment in the future be utilized to determine a plurality of abnormal passenger flow volume cells in the selected area, but also a clustering model is utilized to cluster the abnormal passenger flow volume cells to find out a target aggregation area which is about to generate aggregation activity, and the time of the target aggregation area from the aggregation activity can be prejudged and the warning is carried out, so that the purposes of prejudging the position of the aggregation area, the time of the aggregation activity from the aggregation activity, the passenger flow volume and the like in advance and timely informing related personnel can be achieved, and therefore the related personnel can take measures in advance to block the aggregation risk, reduce the epidemic of infectious diseases and reduce the harm.

In the embodiments of the present application, the thresholds such as the first threshold and the second threshold may be set in advance by epidemic prevention and control personnel or health safety related personnel, or the thresholds of different cells and core areas (including a plurality of cells) may be dynamically changed according to effects such as seasons and tides by using a data analysis platform (as shown in fig. 9).

The epidemic prevention method based on the aggregation risk early warning is described above, and based on the same technical conception, the application also provides an epidemic prevention device based on the aggregation risk early warning, and the epidemic prevention device is briefly described below.

Example 4

Fig. 6 illustrates a schematic structural diagram of an epidemic prevention device based on aggregate risk early warning according to an embodiment of the present application, as shown in fig. 6, in a software implementation, an epidemic prevention device 600 based on aggregate risk early warning may include: a first data acquisition module 601, a first data processing module 602, a passenger flow volume prediction module 603, a judgment module 604 and a first alarm module 605.

A first data acquisition module 601, configured to acquire communication KPI data of a target cell in a selected area within a first preset history period.

And the first data processing module 602 is configured to perform statistical processing on the communication KPI data of the target cell, so as to obtain feature data of the target cell.

The passenger flow prediction module 603 is configured to input feature data of the target cell into a target passenger flow prediction model to obtain a predicted passenger flow of the target cell at a future designated moment, where the target passenger flow prediction model is obtained by training based on historical feature data of at least one cell and a first preset algorithm, the historical feature data of the at least one cell is obtained by performing statistical processing on historical communication KPI data of the at least one cell, and a type of the at least one cell is the same as a type of the target cell.

A judging module 604, configured to judge whether the predicted passenger flow volume of the target cell at the specified time exceeds a first threshold, and if yes, trigger a first alarm module 605 described below, otherwise end.

The first alarm module 605 is configured to execute a first alarm operation when the predicted passenger flow volume of the target cell at the specified time exceeds a first threshold, so as to remind the target object of epidemic prevention.

The first alarm operation is used for informing a target object that the target cell has aggregation risk at the appointed moment.

According to the epidemic situation prevention device based on aggregation risk early warning, the predicted passenger flow of the target cell at a certain moment in the future can be predicted in advance by using the communication KPI data of the target cell and the target passenger flow prediction model obtained through training, and when the predicted passenger flow of the cell at the moment exceeds a set threshold value, alarm information is sent to a target object to give an alarm, so that the purposes of judging the aggregation risk in advance and informing related personnel can be achieved, the related personnel can take measures in advance to block the occurrence of the aggregation risk, the epidemic of infectious diseases is reduced, and the harm is reduced.

Example 5

Fig. 7 is a schematic structural diagram of an epidemic prevention device based on aggregate risk early warning according to another embodiment of the present application, as shown in fig. 7, in a software implementation, an epidemic prevention device 600 based on aggregate risk early warning may include: the first data acquisition module 601, the first data processing module 602, the passenger flow volume prediction module 603, the judgment module 604, and the first alarm module 605 may further include: a second data acquisition module 608, an abnormal cell determination module 609, a clustering module 610, an aggregation zone determination module 611, and a second alert module 612.

And a second data obtaining module 608, configured to obtain real passenger flows of the plurality of cells at the specified time when the specified time is reached.

An abnormal cell determining module 609, configured to determine a plurality of abnormal cells in the plurality of cells based on predicted traffic volumes of the plurality of cells at the specified time and actual traffic volumes of the plurality of cells at the specified time.

And the clustering module 610 is configured to cluster the plurality of abnormal cells based on a target clustering model to obtain at least one abnormal cell cluster.

The aggregation area determining module 611 is configured to determine, based on the center of the at least one abnormal cell cluster, a location where the target aggregation area is located.

The second alarm module 612 is configured to perform a second alarm operation to remind the target object to take epidemic prevention.

Similar to the first alarm operation, the second alarm operation may also include, but is not limited to: pushing alarm information to target objects, highlighting the target aggregation areas and associated cells (cells in a cluster where the target aggregation areas are located) on a monitoring system in a selected area, sending out alarm sounds and other modes, calling other early warning functions and the like. In specific implementation, the warning information may be sent by sending a mail, a chart, a short message, etc. to the relevant responsible person of the target cell, where the warning information may include the location of the target aggregation area, the location and name of the associated cell (the cell in the cluster where the target aggregation area is located), etc.

According to the epidemic situation prevention device based on aggregation risk early warning, the predicted passenger flow volume and the real passenger flow volume of a plurality of cells in the selected area at a certain moment in the future can be used for determining a plurality of abnormal passenger flow volume cells in the selected area, then the abnormal cells are clustered by using a clustering model, a target aggregation area which is about to occur aggregation activity is found out, and warning is carried out, so that the purpose of prejudging the aggregation area in advance and timely informing related personnel can be achieved, the related personnel can take measures in advance to block the occurrence of aggregation risk, the epidemic of infectious diseases is reduced, and the harm is reduced.

Example 5

Fig. 8 is a schematic structural diagram of an epidemic prevention device based on aggregate risk early warning according to another embodiment of the present application, as shown in fig. 8, in a software implementation, an epidemic prevention device 600 based on aggregate risk early warning may include: the first data acquisition module 601, the first data processing module 602, the traffic prediction module 603, the judgment module 604, and the first alarm module 605, the second data acquisition module 608, the abnormal cell determination module 609, the clustering module 610, the aggregation area determination module 611, and the second alarm module 612 may further include: a third data acquisition module 613, a second data processing module 614, an activity time prediction module 615, and a third alert module 616.

The third data obtaining module 613 is configured to obtain, with the target aggregation area as a center, communication KPI data of an inner layer, an intermediate layer, and an outer layer of the target aggregation area within a second preset history period, respectively.

And the second data processing module 614 is configured to perform statistical processing on communication KPI data of the inner layer, the middle layer, and the outer layer of the target aggregation area at multiple times within the second preset history period, so as to obtain feature data of the target aggregation area at the multiple times.

The activity time prediction module 615 is configured to input feature data of the target aggregation area at the multiple times into an activity time prediction model, so as to obtain a time when the current time is away from the target aggregation area and aggregation activity occurs.

And a third alarm module 616 configured to perform a third alarm operation to remind the target object to perform epidemic prevention.

Similar to the first and second alarm operations, the third alarm operation may also include, but is not limited to: the method comprises the steps of pushing alarm information to target objects, highlighting and displaying the target aggregation areas, the time of occurrence of distance aggregation activities and associated cells (cells in a cluster where the target aggregation areas are located) on a monitoring system of a selected area, sending out alarm sounds and other early warning functions.

According to the epidemic situation precaution device based on aggregation risk early warning, not only can the predicted passenger flow volume and the real passenger flow volume of a plurality of cells in a selected area at a certain moment in the future be utilized to determine a plurality of abnormal passenger flow volume cells in the selected area, then a clustering model is utilized to cluster the abnormal passenger flow volume cells to find out a target aggregation area where aggregation activity is about to occur, but also the time when the target aggregation area is away from the aggregation activity is about to occur can be prejudged, and warning is carried out, so that the purposes of prejudging the position of the aggregation area, the time when the distance aggregation activity occurs, the passenger flow volume and the like in advance and timely informing related personnel can be achieved, and therefore the related personnel can take measures in advance to block the occurrence of aggregation risk, reduce the epidemic of infectious diseases and reduce harm.

It should be noted that, the apparatus provided in the embodiments of the present application corresponds to the method provided in the embodiments of the present application, so that description of the embodiments of the apparatus is brief, and reference may be made to the description of the embodiments of the corresponding method for relevant details.

It should also be noted that, the epidemic situation prevention method or device based on the aggregation risk early warning provided by the embodiment of the application can be used for epidemic situation prevention, specifically, through the pre-judgment and early warning of the aggregation risk, related personnel can be assisted to evacuate related regional personnel in advance, resource loss and epidemic situation propagation caused by processing the regional passenger flow exceeding a set threshold are avoided, and the prospective and working efficiency of epidemic situation prevention are improved.

Example 7

Fig. 9 is a schematic diagram of a support system in practical application of an epidemic prevention scheme based on aggregation risk early warning according to an embodiment of the present application. As shown in fig. 9, the system may include: the data source, platform layer and application layer are each described in detail below.

Data sources include, but are not limited to, communications KPI data, user location information data, and project data.

The platform layer includes, but is not limited to, a data analysis platform, an AI service platform, and a process automation processing platform.

The data analysis platform is used for carrying out statistical analysis and the like on data (such as one or more of communication KPI data, user position information data and industrial parameter data) acquired from a data source, and comprises offline calculation and real-time stream calculation.

The AI service platform is used for providing various AI service capacity call, such as training and application of passenger flow prediction model, clustering model and activity time prediction model, etc., and the platform can receive user data input, and returns the result to the service caller after processing by the AI model.

The process automation processing platform is used for freely combining various operation components in a process arrangement mode to form a business process (such as an epidemic situation prevention business process based on aggregation risk early warning) of the thematic application, and the process automation processing data, calling a service and generating a storage result for the front end to present.

An electronic device provided in an embodiment of the present application is described below.

Fig. 10 is a schematic structural diagram of an electronic device according to an embodiment of the present application. Referring to fig. 10, at the hardware level, the electronic device includes a processor, and optionally an internal bus, a network interface, and a memory. The Memory may include a Memory, such as a Random-Access Memory (RAM), and may further include a non-volatile Memory (non-volatile Memory), such as at least 1 disk Memory. Of course, the electronic device may also include hardware required for other services.

The processor, network interface, and memory may be interconnected by an internal bus, which may be an ISA (Industry Standard Architecture ) bus, a PCI (Peripheral Component Interconnect, peripheral component interconnect standard) bus, or EISA (Extended Industry Standard Architecture ) bus, among others. The buses may be classified as address buses, data buses, control buses, etc. For ease of illustration, only one bi-directional arrow is shown in FIG. 10, but not only one bus or type of bus.

And the memory is used for storing programs. In particular, the program may include program code including computer-operating instructions. The memory may include memory and non-volatile storage and provide instructions and data to the processor.

The processor reads the corresponding computer program from the nonvolatile memory to the memory and then operates the computer program to form an epidemic prevention device based on aggregation risk early warning on a logic level. The processor is used for executing the programs stored in the memory and is specifically used for executing the following operations:

The method executed by the epidemic prevention device based on the aggregate risk early warning disclosed in the embodiment shown in fig. 10 of the present application may be applied to a processor or implemented by the processor. The processor may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware in a processor or by instructions in the form of software. The processor may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), etc.; but also digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), field programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The disclosed methods, steps, and logic blocks in the embodiments of the present application may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of a method disclosed in connection with the embodiments of the present application may be embodied directly in hardware, in a decoded processor, or in a combination of hardware and software modules in a decoded processor. The software modules may be located in a random access memory, flash memory, read only memory, programmable read only memory, or electrically erasable programmable memory, registers, etc. as well known in the art. The storage medium is located in a memory, and the processor reads the information in the memory and, in combination with its hardware, performs the steps of the above method.

The embodiments of the present application also provide a computer-readable storage medium storing one or more programs, where the one or more programs include instructions, which when executed by an electronic device including a plurality of application programs, enable the electronic device to perform a method performed by an epidemic prevention device based on aggregate risk early warning in the embodiment shown in fig. 7, and specifically configured to perform:

It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

It should be noted that, in the present application, each embodiment is described in a related manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, and reference is made to the description of the method embodiments in part.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises an element.

The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and changes may be made to the present application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. which are within the spirit and principles of the present application are intended to be included within the scope of the claims of the present application.

Claims

1. An epidemic prevention method based on aggregation risk early warning is characterized by comprising the following steps:

acquiring communication KPI data of a target cell in a selected area within a first preset history period, wherein the selected area comprises a plurality of cells, and the target cell is any cell in the plurality of cells;

inputting the characteristic data of the target cell into a target passenger flow volume prediction model to obtain the predicted passenger flow volume of the target cell at a future appointed moment, wherein the target passenger flow volume prediction model is trained based on the historical characteristic data of at least one cell and a first preset algorithm, the historical characteristic data of at least one cell is obtained by carrying out statistical processing on the historical communication KPI data of the at least one cell, the type of the at least one cell is the same as the type of the target cell, the target passenger flow volume prediction model is a passenger flow volume prediction model corresponding to the type of the cell to which the target cell belongs, and one cell type is correspondingly trained to a passenger flow volume prediction model;

judging whether the predicted passenger flow quantity of the target cell at the appointed moment exceeds a first threshold value, if so, executing a first alarm operation to remind a target object of epidemic prevention, wherein the first alarm operation is used for informing the target object that the target cell has aggregation risk at the appointed moment;

When the appointed time is reached, acquiring the real passenger flow of the cells at the appointed time;

determining a plurality of abnormal cells in the plurality of cells based on the predicted traffic of the plurality of cells at the specified time and the actual traffic of the plurality of cells at the specified time, wherein when the difference between the actual traffic of one of the plurality of cells at the specified time and the predicted traffic of the one of the plurality of cells at the specified time exceeds a second threshold, the one of the plurality of cells is determined to be the abnormal cell;

clustering the plurality of abnormal cells based on a target clustering model to obtain at least one abnormal cell cluster, wherein the target clustering model is obtained by training based on preset position data and a preset clustering algorithm, and the preset position data comprises position data of historical abnormal cells or position data of users in the historical abnormal cells;

determining at least one target cluster meeting a preset condition in the at least one abnormal cell cluster, wherein the preset condition comprises at least one of the number of cells being greater than a preset number and the maximum distance of the cells from a cluster center being smaller than a preset distance; determining the center of the target cluster as the position of the target aggregation area;

Executing a second alarm operation to remind the target object of epidemic prevention, wherein the second alarm operation is used for informing the target object of the position of the target aggregation area;

taking the target aggregation area as the center, respectively acquiring communication KPI data of an inner layer, a middle layer and an outer layer of the target aggregation area in a second preset history period, wherein the inner layer of the aggregation area comprises the aggregation area center, the distances between the inner layer, the middle layer and the outer layer of the aggregation area and the aggregation area center decrease in sequence, the outer edge of the inner layer of the aggregation area is tangent to the inner edge of the middle layer of the aggregation area, and the outer edge of the middle layer of the aggregation area is tangent to the inner edge of the outer layer of the aggregation area;

carrying out statistical processing on communication KPI data of the inner layer, the middle layer and the outer layer of the target gathering area at a plurality of moments in the second preset historical period to obtain characteristic data of the target gathering area at the plurality of moments;

inputting characteristic data of the target aggregation area at a plurality of moments into an activity time prediction model to obtain the time of aggregation activity of the current moment from the target aggregation area, wherein the activity time prediction model is trained based on characteristic data, labels and a second preset algorithm of a history aggregation area, the characteristic data of the history aggregation area are obtained by carrying out statistical processing on communication KPI data of an inner layer, a middle layer and an outer layer of the history aggregation area at a plurality of history moments, and the labels of the history aggregation area are determined according to the time of aggregation activity of the plurality of history moments from the history aggregation area;

And executing a third alarm operation to remind the target object of epidemic prevention, wherein the third alarm operation is used for informing the position of the target aggregation area of the target object and the time of aggregation activity of the current moment from the target aggregation area.

2. The method of claim 1, wherein the step of determining the position of the substrate comprises,

the preset clustering algorithm comprises one of a density-based clustering algorithm DBSCAN and a mean shift algorithm.

3. The method of claim 1, wherein the step of determining the position of the substrate comprises,

the first preset algorithm or the second preset algorithm includes one of the following algorithms: XGboost, lightweight gradient lifting LightGBM, gating cycle unit GRU and long and short time memory network LSTM.

4. An epidemic prevention device based on aggregation risk early warning, characterized in that the device comprises:

a first data acquisition module, configured to acquire communication KPI data of a target cell in a selected area within a first preset history period, where the selected area includes a plurality of cells, and the target cell is any cell of the plurality of cells;

The passenger flow prediction module is used for inputting the characteristic data of the target cell into a target passenger flow prediction model to obtain the predicted passenger flow of the target cell at a future appointed moment, wherein the target passenger flow prediction model is trained based on the historical characteristic data of at least one cell and a first preset algorithm, the historical characteristic data of the at least one cell is obtained by carrying out statistical processing on the historical communication KPI data of the at least one cell, the type of the at least one cell is the same as the type of the target cell, the target passenger flow prediction model is a passenger flow prediction model corresponding to the type of the cell to which the target cell belongs, and one cell type is correspondingly trained to one passenger flow prediction model;

the first alarm module is used for executing a first alarm operation when the predicted passenger flow of the target cell at the appointed moment exceeds a first threshold value so as to remind a target object of epidemic prevention, wherein the first alarm operation is used for informing the target object that the target cell has aggregation risk;

the second data acquisition module is used for acquiring the real passenger flow of the cells at the appointed time when the appointed time is reached;

An abnormal cell determining module, configured to determine a plurality of abnormal cells in the plurality of cells based on predicted traffic volumes of the plurality of cells at the specified time and actual traffic volumes of the plurality of cells at the specified time, where when a difference between the actual traffic volume of one of the plurality of cells at the specified time and the predicted traffic volume of the one of the plurality of cells at the specified time exceeds a second threshold, determine the one of the plurality of cells as an abnormal cell;

the clustering module is used for clustering the plurality of abnormal cells based on a target clustering model to obtain at least one abnormal cell cluster, wherein the target clustering model is obtained by training based on preset position data and a preset clustering algorithm, and the preset position data comprise position data of historical abnormal cells or position data of users in the historical abnormal cells;

the aggregation area determining module is used for determining at least one target cluster meeting a preset condition in the at least one abnormal cell cluster, wherein the preset condition comprises at least one of the number of cells being larger than a preset number and the maximum distance between the cells and the cluster center being smaller than a preset distance; determining the center of the target cluster as the position of the target aggregation area;

The second alarm module is used for executing a second alarm operation to remind the target object of epidemic prevention, wherein the second alarm operation is used for informing the target object of the position of the target aggregation area;

the third data acquisition module is used for respectively acquiring communication KPI data of an inner layer, a middle layer and an outer layer of the target gathering area in a second preset history period by taking the target gathering area as a center, wherein the inner layer of the gathering area comprises a gathering area center, the distances between the inner layer, the middle layer and the outer layer of the gathering area and the gathering area center are sequentially decreased, the outer edge of the inner layer of the gathering area is tangent to the inner edge of the middle layer of the gathering area, and the outer edge of the middle layer of the gathering area is tangent to the inner edge of the outer layer of the gathering area;

the second data processing module is used for carrying out statistical processing on communication KPI data of the inner layer, the middle layer and the outer layer of the target gathering area at a plurality of moments in the second preset historical period to obtain characteristic data of the target gathering area at the plurality of moments;

the activity time prediction module is used for inputting the characteristic data of the target aggregation area at the plurality of moments into an activity time prediction model to obtain the time of aggregation activity of the target aggregation area at the current moment, wherein the activity time prediction model is trained based on the characteristic data of a history aggregation area, a label and a second preset algorithm, the characteristic data of the history aggregation area are obtained by carrying out statistical processing on communication KPI data of an inner layer, a middle layer and an outer layer of the history aggregation area at the plurality of history moments, and the label of the history aggregation area is determined according to the time of aggregation activity of the history aggregation area at the plurality of history moments;

And the third alarm module is used for executing a third alarm operation to remind the target object of epidemic prevention, wherein the third alarm operation is used for informing the position of the target aggregation area of the target object and the time of aggregation activity of the current moment from the target aggregation area.