Emergency early warning information targeted release system and method based on artificial intelligence

Abstract

The invention is suitable for the field of information processing, and provides an emergency early warning information targeted release system and method based on artificial intelligence, and the method comprises the steps: obtaining and recognizing the user ontology information and tendency data of an application program, and generating a targeted region and a targeted time period; monitoring and obtaining the line forming target time of the application program user; setting a risk model; based on the risk model, identifying and generating a risk early warning value of the line target time; the method and the system have the beneficial effects that the method and the system can be used for identifying and analyzing the user tendency data of the application program, generating a target area and a target time period, and calculating and judging adverse factors which influence the user intention to go out based on a risk model, so that the user intention to go out can be quickly and accurately pushed, and the user intention to go out can be quickly and accurately pushed. And after the judgment result is generated, the early warning action is precisely pushed in a targeted manner in combination with the targeted region and the targeted time period, so that the user is helped to carry out auxiliary judgment.

Description

Technical Field

[0001] The present invention belongs to the field of information processing, and in particular relates to an artificial intelligence-based emergency warning information targeted release system and method. Background Art

[0002] Emergency events refer to natural disasters, accidents, public health events and social security events that occur suddenly and cause or may cause serious social harm and require emergency response measures to deal with. These events are usually characterized by suddenness, urgency and destructiveness.

[0003] When some emergencies occur, based on the development of artificial intelligence technology, in order to reduce the harm of emergencies, some fields have set up targeted release systems for early warning information, thereby realizing the effective push of early warning information. However, during the use of some applications, the existing targeted release system has limited accuracy, and there are shortcomings such as overflow of the push range and insufficient targeting of the pushed information. In view of the above problems, it is urgent to develop an emergency warning information targeted release system and method based on artificial intelligence. Summary of the invention

[0004] The purpose of the embodiments of the present invention is to provide an artificial intelligence-based emergency warning information targeted release system and method, aiming to solve the problems raised in the above-mentioned background technology.

[0005] The embodiment of the present invention is implemented as follows: on the one hand, a method for targeted release of emergency warning information based on artificial intelligence, the method comprising: Acquire and identify application targeted user information and tendency data to generate a targeted information set; Based on the targeted information set, monitor and obtain the target time for the application's targeted users to complete the trip; Setting up risk models; Based on the risk model, identify and generate risk warning values ​​for the target travel time; Compare risk warning values ​​and generate application-specific push instructions.

[0006] As a further solution of the present invention, the obtaining and identifying application user information and tendency data and generating a targeted information set specifically includes: Collect application users whose stay time on posts with travel tags exceeds the stay time threshold, generate and mark them as target users; Obtain active and passive feedback data from targeted users; Extracting position information from active feedback data and passive feedback data; Perform duplicate checking on the location information and generate duplicate checking time and high-frequency location information; Generate targeted time periods based on the time of duplicate checking.

[0007] As a further solution of the present invention, monitoring and obtaining the target time of the application user's trip based on the targeted information set specifically includes: Based on the targeted time period, monitor the ticket payment information of the targeted users; Generate a target travel time based on ticket payment information; Generate early warning calculation instructions based on the target completion time.

[0008] As a further solution of the present invention, the setting of the risk model specifically includes: W=(v*2)+(w*1.5)+(r*1.2)+(s*1.8); Where W is the risk warning value, v is the visibility warning value, w is the strong wind warning value, r is the heavy rain warning value, and s is the ice and snow warning value.

[0009] As a further solution of the present invention, the identification and generation of the risk warning value of the target travel time based on the risk model specifically includes: Import the visibility forecast value, wind speed forecast value, rainstorm precipitation forecast value and road icing forecast value three days before and after the target travel time respectively; Based on the risk model, a first risk warning value, a second risk warning value and a third risk warning value are generated; The average values ​​of the first risk warning value, the second risk warning value and the third risk warning value are calculated to generate a risk warning reference value and a risk keyword set.

[0010] As a further solution of the present invention, the comparing risk warning values ​​and generating application directional push instructions specifically include: Determine whether the risk warning reference value is greater than the medium risk threshold; If the risk warning reference value is greater than the medium risk threshold, a medium push frequency is generated; Generate a predicted time period based on the target time for the trip; Obtain high-frequency location information and risk keyword sets to generate conditional packages; Based on the predicted time period, retrieve the post information of the qualified package and generate an event set; Push event sets to targeted users based on the intermediate push frequency; If the risk warning reference value is greater than the high risk threshold, an advanced push frequency is generated; Push event sets to targeted users based on advanced push frequency.

[0011] As a further solution of the present invention, on the other hand, a system for targeted release of emergency warning information based on artificial intelligence, the system comprising: A first acquisition and identification module is used to acquire and identify application target user information; A second acquisition and identification module is used to acquire and identify trend data; A first generation module is used to generate a targeting information set; A monitoring and acquisition module, used to monitor and acquire the target time of the application target user's trip based on the target information set; Risk model module, used to set up risk models; An identification and generation module is used to identify and generate risk warning values ​​for the target travel time based on the risk model; Comparison module, used to compare risk warning values; The second generating module is used to generate an application program directional push instruction.

[0012] As a further solution of the present invention, the first generating module specifically includes: A collection unit, used for collecting application users who stay on posts with travel tags for a time period exceeding a threshold of the stay time; A first generating unit, used to generate and mark a target user; An acquisition unit, used to acquire active feedback data and passive feedback data of targeted users; An extraction unit, used to extract position information from active feedback data and passive feedback data; A duplicate checking unit, used for checking duplicates of location information; The second generating unit is used to generate duplicate checking time and high-frequency position information; The third generating unit is used to generate a targeted time period based on the duplication checking time.

[0013] The embodiments of the present invention provide a system and method for targeted release of emergency warning information based on artificial intelligence. The method and system can identify and analyze application user tendency data, generate targeted areas and targeted time periods, and at the same time, based on the risk model, calculate and judge the adverse factors that affect the user's travel intention. When the judgment result is generated, it is combined with the targeted area and targeted time period to accurately and targetedly push warning actions, thereby helping users to make auxiliary judgments. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a main flow chart of a method for targeted release of emergency warning information based on artificial intelligence.

[0015] Figure 2 The present invention is a flowchart for obtaining and identifying application user information and tendency data to generate a targeted information set in a targeted release method of emergency warning information based on artificial intelligence.

[0016] Figure 3The invention discloses a flowchart for monitoring and obtaining the target time of application users based on a targeted information set in a method for targeted release of emergency warning information based on artificial intelligence.

[0017] Figure 4 The invention is a flowchart of a method for targeted release of emergency warning information based on artificial intelligence, which identifies and generates risk warning values ​​at target times based on risk models.

[0018] Figure 5 The invention discloses a flow chart of generating application-directed push instructions by comparing risk warning values ​​in a targeted release method of emergency warning information based on artificial intelligence.

[0019] Figure 6 It is the main structure diagram of an emergency warning information targeted release system based on artificial intelligence.

[0020] Figure 7 It is a structural block diagram of the first generation module in the emergency warning information targeted release system based on artificial intelligence. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0022] The specific implementation of the present invention is described in detail below in conjunction with specific embodiments.

[0023] The invention provides an artificial intelligence-based targeted release method and system for emergency warning information, which solves the technical problems in the background technology.

[0024] like Figure 1 As shown, it is a main flow chart of a method for targeted release of emergency warning information based on artificial intelligence provided by an embodiment of the present invention, and the method for targeted release of emergency warning information based on artificial intelligence includes: Step S100: Acquire and identify application target user information and tendency data, and generate a target information set; Step S200: Based on the target information set, monitor and obtain the target time for the application target user to complete the trip; Step S300: setting a risk model; Step S400: Based on the risk model, identify and generate a risk warning value for the target travel time; Step S500: Compare the risk warning values ​​and generate a directional push instruction for the application.

[0025] When this embodiment is applied, the method and system are based on a pre-developed application. This application can share content such as lifestyle and travel experience, and can also purchase tickets such as train tickets and air tickets. When a user uses this application, he or she must first fill in user information and register as a user. After successful registration, under the premise of ensuring user privacy, a preset algorithm can be used to implement a tendency push to the user. When the user registration is completed and this application is used, the application target user ontology information and tendency data are obtained and identified, and travel tag features and a stay time threshold are preset. Application users who stay on travel tag feature posts for a time exceeding the stay time threshold are collected, and generated and marked as target users. The tendency data is divided into active feedback data and passive feedback data. The active feedback data is the post data information when the target user performs operations such as likes and collections when using this application. The passive feedback data is the post data information that the target user has browsed when using this application. When the active feedback data and the passive feedback data are identified, the active feedback data and the passive feedback data are extracted. Several location information, and check the duplicates of several location information to generate high-frequency location information and duplicate checking time. Based on the duplicate checking time, a tolerance period of one month is set before and after to generate a targeted time period. After the high-frequency location information, duplicate checking time and risk model are generated, the targeted users of the application are continuously monitored. When the targeted users of the application purchase tickets in this application, the ticket purchase information is extracted to generate a target time for travel. The application is preset with a risk model. Various parameters are imported into the risk model to generate a risk warning value, and then the risk warning value is compared to determine whether the risk warning reference value is greater than the medium risk threshold. If the risk warning reference value is greater than the medium risk threshold, an intermediate push frequency is generated. Based on the target time for travel, a predicted time period is generated, high-frequency location information and a risk keyword set are obtained, a conditional package is generated, and based on the predicted time period, post information that meets the conditional package is retrieved to generate an event set. Based on the intermediate push frequency, the event set is pushed to the targeted user. If the risk warning reference value is greater than the high risk threshold, an advanced push frequency is generated. Based on the advanced push frequency, the event set is pushed to the targeted user.

[0026] like Figure 2 As shown, as a preferred embodiment of the present invention, the acquisition and identification of application user information and tendency data and the generation of a targeted information set specifically include: Step S101: collecting application users whose stay time on posts with travel tags exceeds a stay time threshold, generating and marking them as target users; Step S102: Obtaining active feedback data and passive feedback data of targeted users; Step S103: extracting location information from active feedback data and passive feedback data; Step S104: perform duplicate checking on the location information to generate duplicate checking time and high-frequency location information; Step S105: Generate a targeted time period based on the duplicate checking time.

[0027] When this embodiment is applied, travel tag features and a stay time threshold are preset, and application users who stay on travel tag feature posts for a time exceeding the stay time threshold are collected, generated and marked as targeted users, and then active feedback data and passive feedback data of the targeted users are obtained. The active feedback data is the post data information when the targeted users perform operations such as likes and collections when using this application, and the passive feedback data is the post data information that the targeted users have browsed when using this application. The location information in the active feedback data and the passive feedback data is extracted, the location information is checked for duplicates, and the duplicate checking time and high-frequency location information are generated. Based on the duplicate checking time, a tolerance time of one month is set before and after each time, and then a targeted time period is generated. The targeted users can be continuously monitored within the targeted time period.

[0028] like Figure 3 As shown, as a preferred embodiment of the present invention, the monitoring and obtaining the target time of the application user's trip based on the target information set specifically includes: Step S201: monitoring the ticket payment information of the targeted user based on the targeted time period; Step S202: Generate a travel target time based on the ticket payment information; Step S203: Generate an early warning calculation instruction based on the target travel time.

[0029] When this embodiment is applied, after the target time period and the target user are generated, the ticket payment information of the target user is continuously monitored within the target time period, and a target travel time is generated based on the ticket payment information. The target travel time is the travel time of the target user, and an early warning calculation instruction is generated based on the target travel time, that is, the risk early warning value is calculated through the risk model according to the predicted data of the day at that time.

[0030] As a preferred embodiment of the present invention, the setting of the risk model specifically includes: W=(v*2)+(w*1.5)+(r*1.2)+(s*1.8); Where W is the risk warning value, v is the visibility warning value, w is the strong wind warning value, r is the heavy rain warning value, and s is the ice and snow warning value; It should be understood that W is the risk warning value, v represents the visibility warning value, the weight coefficient of the visibility warning value is 2, and different warning levels are set according to the visibility. When the visibility is less than 100 meters, v=10, when the visibility is 100-500 meters, v=5, and when the visibility is above 500 meters, v=0; w represents the strong wind warning value, and the weight coefficient of the strong wind warning value is 1.5. Different warning levels are set according to the wind speed. When the wind speed exceeds 10 meters / second, w=10, when the wind speed is 5-10 meters / second, w=5, and when the wind speed is less than 5 meters / second, w=0; r represents the heavy rain warning value, and the weight coefficient of the heavy rain warning value is 1.2. Different warning levels are set according to the intensity of the heavy rain or the precipitation. When the precipitation exceeds 50 mm / hour, r=10, when the precipitation is 25-50 mm / hour, r=5, and when the precipitation is less than 25 mm / hour, r=0; s represents the ice and snow warning value, and the weight coefficient of the ice and snow warning value is 1.8. Different warning levels are set according to the degree of ice and snow coverage and the icing conditions of the road surface. When the road surface is severely frozen, s=10, when the road surface is partially frozen, s=5, and when there is no ice on the road surface, s=0.

[0031] like Figure 4 As shown, as a preferred embodiment of the present invention, the risk model-based identification and generation of the risk warning value of the target travel time specifically includes: Step S401: Importing respectively the visibility forecast value, wind speed forecast value, rainstorm precipitation forecast value and road icing forecast value of three days before and after the target travel time; Step S402: generating a first risk warning value, a second risk warning value and a third risk warning value based on the risk model; Step S403: Calculate the average of the first risk warning value, the second risk warning value and the third risk warning value to generate a risk warning reference value and a risk keyword set.

[0032] When this embodiment is applied, based on the data of the weather forecast system, the visibility forecast value, wind speed forecast value, rainstorm precipitation forecast value and road icing forecast value three days before and after the target travel time are respectively imported, and based on the risk model, the first risk warning value, the second risk warning value and the third risk warning value are generated, and the average values ​​of the first risk warning value, the second risk warning value and the third risk warning value are calculated to generate a risk warning reference value and a risk keyword set, and the risk keyword set includes but is not limited to strong winds, heavy rains, icing, landslides, mudslides, etc.

[0033] like Figure 5 As shown, as another preferred embodiment of the present invention, the comparing risk warning values ​​and generating the application directional push instruction specifically includes: Step S501: determine whether the risk warning reference value is greater than the medium risk threshold; Step S502: If the risk warning reference value is greater than the medium risk threshold, generate a medium-level push frequency; Step S503: Generate a predicted time period based on the scheduled departure time; Step S504: Obtain high-frequency location information and a risk keyword set, and generate a condition package; Step S505: Based on the predicted time period, retrieve post information that meets the condition package, and generate an event set; Step S506: Push the event set to the targeted users based on the medium-level push frequency; Step S507: If the risk warning reference value is greater than the high risk threshold, generate a high-level push frequency; Step S508: Push the event set to the targeted users based on the high-level push frequency.

[0034] When this embodiment is applied, when the risk warning value W is less than or equal to a certain value, it indicates that the current meteorological conditions are in a low-risk state and no special measures need to be taken; when the warning value W is greater than the low risk threshold but less than or equal to a certain value, it indicates that the current meteorological conditions are in a medium-risk state and some preventive measures or attention to the dynamics may be required; when the warning value W is greater than the medium risk threshold, it indicates that the current meteorological conditions are in a high-risk state and necessary measures need to be taken immediately to address possible risks. The risk warning value threshold can be set as follows: low risk threshold: W ≤ 10, medium risk threshold: 10 < W ≤ 30, high risk threshold: W > 30; determine whether the risk warning reference value is greater than the medium risk threshold. If the risk warning reference value is greater than the medium risk threshold, generate a medium-level push frequency, generate a predicted time period based on the scheduled departure time. The predicted time period is the time range of one week before and after the date of the scheduled departure time, and then obtain high-frequency location information and a risk keyword set, generate a condition package, retrieve post information that meets the condition package based on the predicted time period, generate an event set, and push the event set to the targeted users based on the medium-level push frequency. If the risk warning reference value is greater than the high risk threshold, generate a high-level push frequency, and push the event set to the targeted users based on the high-level push frequency.

[0035] As Figure 6 shown, as another preferred embodiment of the present invention, on the other hand, an artificial intelligence-based targeted release system for emergency warning information, the system includes: A first acquisition and recognition module 100, configured to acquire and recognize application program targeted user information; A second acquisition and recognition module 200, configured to acquire and recognize tendency data; A first generation module 300, configured to generate a targeted information set; A monitoring and acquisition module 400, configured to monitor and acquire the scheduled departure time of the application program targeted users based on the targeted information set; The risk model module 500 is used to set the risk model; An identification and generation module 600 is used to identify and generate a risk warning value of the target travel time based on the risk model; A comparison module 700 is used to compare risk warning values; The second generating module 800 is used to generate an application program directional push instruction.

[0036] When this embodiment is applied, the first acquisition and identification module 100 acquires and identifies the application target user information, the second acquisition and identification module 200 acquires and identifies the tendency data, the first generation module 300 generates a target information set, based on the target information set, the monitoring and acquisition module 400 monitors and acquires the target time for the application target user to complete the trip, the risk model module 500 sets the risk model, based on the risk model, the identification and generation module 600 identifies and generates a risk warning value for the target time for completing the trip, the comparison module 700 compares the risk warning values, and the second generation module 800 generates a targeted push instruction for the application.

[0037] like Figure 7 As shown, as another preferred embodiment of the present invention, the first generating module 300 specifically includes: A collection unit 301 is used to collect application users who stay on travel tag feature posts for a period exceeding a stay time threshold; A first generating unit 302, configured to generate and mark a target user; An acquisition unit 303 is used to acquire active feedback data and passive feedback data of a targeted user; An extraction unit 304, configured to extract location information from active feedback data and passive feedback data; A duplicate checking unit 305 is used to check the location information for duplicates; The second generating unit 306 is used to generate duplicate checking time and high-frequency position information; The third generating unit 307 is used to generate a target time period based on the duplication checking time.

[0038] When this embodiment is applied, the collection unit 301 collects application users whose stay time on travel tag feature posts exceeds the stay time threshold, the first generation unit 302 generates and marks them as targeted users, the acquisition unit 303 obtains active feedback data and passive feedback data of the targeted users, the extraction unit 304 extracts location information in the active feedback data and passive feedback data, the duplicate checking unit 305 performs duplicate checking on the location information, the second generation unit 306 generates duplicate checking time and high-frequency location information, and based on the duplicate checking time, the third generation unit 307 generates a targeted time period.

[0039] The above embodiment of the present invention provides a method for targeted release of emergency warning information based on artificial intelligence, and provides a system for targeted release of emergency warning information based on artificial intelligence. The method and system are based on a pre-developed application. The application can share content such as lifestyle and travel experience, and can also purchase tickets such as train tickets and air tickets. When a user uses the application, he or she must first fill in the user information and register the user. After the registration is successful, under the premise of ensuring the user's privacy, the user's tendency to be pushed can be realized through a preset algorithm. When the user registration is completed and the application is used, the application's targeted user ontology information and tendency information are obtained and identified. According to the data, travel tag features and stay time thresholds are preset, and application users who stay on posts with travel tag features for a time exceeding the stay time threshold are collected, and they are generated and marked as target users. The tendency data is divided into active feedback data and passive feedback data. The active feedback data is the post data information when the target user performs operations such as likes and collections when using the application, and the passive feedback data is the post data information that the target user has browsed when using the application. When identifying the active feedback data and the passive feedback data, several location information contained in the active feedback data and the passive feedback data are extracted, and the several location information are checked for duplication to generate high-frequency Location information and duplicate checking time. Based on the duplicate checking time, a tolerance period of one month is set before and after to generate a targeted time period. After generating high-frequency location information, duplicate checking time, and risk model, the targeted users of the application are continuously monitored. When the targeted users of the application purchase tickets within the application, the ticket purchase information is extracted to generate a target travel time. The application has a preset risk model. By importing various parameters into the risk model, a risk warning value can be generated. The risk warning value is then compared to determine whether the risk warning reference value is greater than the medium risk threshold. If the risk warning reference value is greater than the medium risk threshold, a medium push frequency is generated. Based on the target travel time, a predicted time period is generated. , obtain high-frequency location information and risk keyword sets, generate conditional packages, retrieve post information that meets the conditional packages based on the predicted time period, generate event sets, and push event sets to targeted users based on the intermediate push frequency. If the risk warning reference value is greater than the high-risk threshold, generate an advanced push frequency, and push event sets to targeted users based on the advanced push frequency; this method and system can identify and analyze application user tendency data, generate targeted areas and targeted time periods, and at the same time, based on the risk model, calculate and judge the adverse factors that affect the user's intended travel. When the judgment result is generated, it is combined with the targeted area and targeted time period to accurately and targetedly push warning actions to help users make auxiliary judgments.

[0040] In order to load the above-mentioned method and system and enable it to run smoothly, the system, in addition to the various modules mentioned above, may also include more or fewer components than described above, or a combination of certain components, or different components, for example, it may include input and output devices, network access devices, buses, processors and memories, etc.

[0041] The processor may be a central processing unit, or other general-purpose processors, digital signal processors, application-specific integrated circuits, off-the-shelf programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the system, and various interfaces and lines are used to connect various parts.

[0042] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0043] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

[0044] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for targeted release of emergency warning information based on artificial intelligence, characterized in that: The method comprises: Acquire and identify application targeted user information and tendency data to generate a targeted information set; Based on the targeted information set, monitor and obtain the target time for the application's targeted users to complete the trip; Setting up risk models; Based on the risk model, identify and generate risk warning values ​​for the target travel time; Compare risk warning values ​​and generate application-specific push instructions.

2. The method for targeted release of emergency warning information based on artificial intelligence according to claim 1 is characterized in that: The obtaining and identifying application user information and tendency data and generating a targeted information set specifically includes: Collect application users whose stay time on posts with travel tags exceeds the stay time threshold, generate and mark them as target users; Obtain active and passive feedback data from targeted users; Extracting position information from active feedback data and passive feedback data; Perform duplicate checking on the location information and generate duplicate checking time and high-frequency location information; Generate targeted time periods based on the duplication checking time.

3. The method for targeted release of emergency warning information based on artificial intelligence according to claim 2 is characterized in that: The monitoring and obtaining of the target time of the application user's trip based on the targeted information set specifically includes: Based on the targeted time period, monitor the ticket payment information of the targeted users; Generate a target travel time based on ticket payment information; Generate early warning calculation instructions based on the target completion time.

4. The method for targeted release of emergency warning information based on artificial intelligence according to claim 1 is characterized in that: The risk model setting specifically includes: W=(v*2)+(w*1.5)+(r*1.2)+(s*1.8); Where W is the risk warning value, v is the visibility warning value, w is the strong wind warning value, r is the heavy rain warning value, and s is the ice and snow warning value.

5. The method for targeted release of emergency warning information based on artificial intelligence according to claim 3 is characterized in that: The method of identifying and generating a risk warning value for the target travel time based on the risk model specifically includes: Import the visibility forecast value, wind speed forecast value, rainstorm precipitation forecast value and road icing forecast value three days before and after the target travel time respectively; Based on the risk model, a first risk warning value, a second risk warning value and a third risk warning value are generated; The average values ​​of the first risk warning value, the second risk warning value and the third risk warning value are calculated to generate a risk warning reference value and a risk keyword set.

6. The method for targeted release of emergency warning information based on artificial intelligence according to claim 5 is characterized in that: The step of comparing the risk warning values ​​and generating the application program directional push instruction specifically includes: Determine whether the risk warning reference value is greater than the medium risk threshold; If the risk warning reference value is greater than the medium risk threshold, a medium push frequency is generated; Generate a predicted time period based on the target time for the trip; Obtain high-frequency location information and risk keyword sets to generate conditional packages; Based on the predicted time period, retrieve the post information of the qualified package and generate an event set; Push event sets to targeted users based on the intermediate push frequency; If the risk warning reference value is greater than the high risk threshold, an advanced push frequency is generated; Push event sets to targeted users based on advanced push frequency.

7. An artificial intelligence-based emergency warning information targeted release system, characterized in that: The system comprises: A first acquisition and identification module is used to acquire and identify application target user information; A second acquisition and identification module is used to acquire and identify trend data; A first generation module is used to generate a targeting information set; A monitoring and acquisition module, used to monitor and acquire the target time of the application target user's trip based on the target information set; Risk model module, used to set up risk models; An identification and generation module is used to identify and generate risk warning values ​​for the target travel time based on the risk model; Comparison module, used to compare risk warning values; The second generating module is used to generate an application program directional push instruction.

8. The artificial intelligence-based emergency warning information targeted release system according to claim 7 is characterized in that: The first generation module specifically includes: A collection unit, used for collecting application users who stay on posts with travel tags for a time period exceeding a threshold of the stay time; A first generating unit, used to generate and mark a target user; An acquisition unit, used to acquire active feedback data and passive feedback data of targeted users; An extraction unit, used to extract position information from active feedback data and passive feedback data; A duplicate checking unit, used for checking duplicates of location information; The second generating unit is used to generate duplicate checking time and high-frequency position information; The third generating unit is used to generate a targeted time period based on the duplication checking time.

Images