Message pushing method and device, computer device and storage medium
By acquiring the profile tags of candidate push targets, the priority of push messages, and historical data, push strategy data is generated to control profile tags, frequency, and real-time blacklists. This solves the problem of inaccurate push in traditional message push technology and improves resource utilization efficiency and user satisfaction.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- ZHAOLIAN CONSUMER FINANCE CO LTD
- Filing Date
- 2022-11-25
- Publication Date
- 2026-08-04
AI Technical Summary
Traditional push notification technology cannot achieve accurate delivery, resulting in wasted resources and user resentment, and it cannot meet the personalized needs of different users.
By acquiring the profile tags of candidate push targets, the priority of push messages, preset frequency, and historical push data, overall push strategy data is generated to control profile tags, frequency, and real-time blacklist management, thereby achieving precise push.
It enables precise message delivery, reduces resource waste, improves user experience, and enhances delivery effectiveness.
Smart Images
Figure CN115883503B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of Internet technology, and in particular to a message push method, apparatus, computer equipment, storage medium, and computer program product. Background Technology
[0002] With the development of internet technology, pushing products to users via the internet has become a popular technology. It can discover or explore the needs of potential users, allowing them to learn about the products being pushed and ultimately purchase them.
[0003] Traditional push notification technology typically delivers messages indiscriminately to users via advertisements, SMS messages, or app splash screens. However, this technology cannot achieve precise targeting. For example, indiscriminate push notifications can lead to users who don't want to receive them feeling overwhelmed, wasting resources and causing user aversion. Conversely, sending only a few indiscriminate push notifications to users who do want them may reduce the effectiveness of the push notifications. Summary of the Invention
[0004] Therefore, it is necessary to provide a message push method, device, computer equipment, computer-readable storage medium, and computer program product that can achieve accurate message push in response to the above-mentioned technical problems.
[0005] Firstly, this application provides a message push method. The method includes:
[0006] Get message push task;
[0007] Extract the candidate push object profile tags, the priority of the required push message, the preset push frequency, historical push data, and the message to be pushed corresponding to the message push task;
[0008] A blacklist of recipients to reject messages is determined based on the historical push data.
[0009] Based on the profile tags, the candidate push objects are managed by profile tags; based on the priority of the required push messages and the preset push frequency, frequency management is performed; and based on the push object blacklist, real-time management is performed to generate overall push strategy data.
[0010] The message to be pushed is pushed based on the overall push strategy data.
[0011] In one embodiment, before extracting the candidate push object profile tags, the priority of the required push message, the preset push frequency, historical push data, and the message to be pushed corresponding to the message push task, the method further includes:
[0012] Obtain the type of the message push task;
[0013] Based on the type of the message push task, the message push object set is initially filtered to obtain the initially filtered push object set.
[0014] Candidate push objects are selected from the set of push objects after initial screening.
[0015] In one embodiment, the step of dividing candidate push objects from the initially screened set of push objects includes:
[0016] The set of push objects after the initial screening is further subdivided proportionally to obtain candidate push objects.
[0017] In one embodiment, the method further includes:
[0018] The set of push objects after the initial screening is proportionally subdivided to obtain candidate push objects and a control group of candidate push objects;
[0019] The message to be pushed is pushed to the candidate push object control group, and the first message push feedback data fed back by the candidate push object control group is collected;
[0020] When the message to be pushed is completed based on the overall push strategy data, the second message push feedback data is collected and fed back.
[0021] By comparing the first message push feedback data and the second message push feedback data, the overall push strategy data evaluation data is obtained.
[0022] In one embodiment, the steps of managing candidate push targets based on the profile tags, managing frequency based on the priority of the required push messages and the preset push frequency, and managing in real time based on the push target blacklist, to generate overall push strategy data, include:
[0023] Based on the profile tags, the candidate push objects are managed by profile tags to obtain push strategy data after profile tag management;
[0024] After obtaining the push strategy data after the portrait tag control, frequency control is performed based on the priority of the required push message and the preset push frequency to obtain the push strategy data after frequency control.
[0025] After obtaining the push strategy data after frequency control, real-time control is performed based on the blacklist of push targets to obtain push strategy data after real-time control.
[0026] Based on the push strategy data after the profile tag management, the push strategy data after the frequency management, and the push strategy data after the real-time management, overall push strategy data is generated.
[0027] In one embodiment, the method further includes:
[0028] Obtain the list of eligible individuals;
[0029] When the candidate push object is in the list of exemptable objects, the candidate push object in the list of exemptable objects is managed in real time based on the push object blacklist, and first push strategy data is generated.
[0030] When the candidate push object is not in the list of exemptable objects, the candidate push object is controlled by the profile tag based on the profile tag, the frequency is controlled based on the priority of the required push message and the preset push frequency, and the push object blacklist is controlled in real time to generate second push strategy data;
[0031] The first push strategy data and the second push strategy data are collected to generate overall push strategy data.
[0032] Secondly, this application also provides a message push device. The device includes:
[0033] The task acquisition module is used to acquire message push tasks;
[0034] The data extraction module is used to extract the candidate push object profile tags, the priority of the required push message, the preset push frequency, historical push data, and the message to be pushed corresponding to the message push task.
[0035] The blacklist determination module is used to determine a blacklist of recipients for rejected messages based on the historical push data.
[0036] The three-level control module is used to control the candidate push objects based on the profile tags, control the frequency based on the priority of the required push messages and the preset push frequency, and control the push objects in real time based on the blacklist of push objects, and generate overall push strategy data.
[0037] The message push module is used to push the message to be pushed based on the overall push strategy data.
[0038] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the following steps:
[0039] Get message push task;
[0040] Extract the candidate push object profile tags, the priority of the required push message, the preset push frequency, historical push data, and the message to be pushed corresponding to the message push task;
[0041] A blacklist of recipients to reject messages is determined based on the historical push data.
[0042] Based on the profile tags, the candidate push objects are managed by profile tags; based on the priority of the required push messages and the preset push frequency, frequency management is performed; and based on the push object blacklist, real-time management is performed to generate overall push strategy data.
[0043] The message to be pushed is pushed based on the overall push strategy data.
[0044] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, performs the following steps:
[0045] Get message push task;
[0046] Extract the candidate push object profile tags, the priority of the required push message, the preset push frequency, historical push data, and the message to be pushed corresponding to the message push task;
[0047] A blacklist of recipients to reject messages is determined based on the historical push data.
[0048] Based on the profile tags, the candidate push objects are managed by profile tags; based on the priority of the required push messages and the preset push frequency, frequency management is performed; and based on the push object blacklist, real-time management is performed to generate overall push strategy data.
[0049] The message to be pushed is pushed based on the overall push strategy data.
[0050] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, performs the following steps:
[0051] Get message push task;
[0052] Extract the candidate push object profile tags, the priority of the required push message, the preset push frequency, historical push data, and the message to be pushed corresponding to the message push task;
[0053] A blacklist of recipients to reject messages is determined based on the historical push data.
[0054] Based on the profile tags, the candidate push objects are managed by profile tags; based on the priority of the required push messages and the preset push frequency, frequency management is performed; and based on the push object blacklist, real-time management is performed to generate overall push strategy data.
[0055] The message to be pushed is pushed based on the overall push strategy data.
[0056] The aforementioned message push method, apparatus, computer equipment, storage medium, and computer program product first acquire a message push task; extract candidate push object profile tags, the priority of the required push message, the preset push frequency, historical push data, and the message to be pushed corresponding to the message push task; determine a blacklist of push objects to reject messages based on the historical push data; perform profile tag control on candidate push objects based on profile tags, frequency control based on the priority of the required push message and the preset push frequency, and real-time control based on the push object blacklist to generate overall push strategy data; and push the message to be pushed based on the overall push strategy data. In this application, the overall push strategy data generated based on profile tag control, frequency control, and real-time control can rely on the overall push strategy data to achieve accurate push of the message to be pushed. Attached Figure Description
[0057] Figure 1 This is an application environment diagram of the message push method in one embodiment;
[0058] Figure 2 This is a flowchart illustrating a message push method in one embodiment;
[0059] Figure 3 This is a flowchart illustrating the steps of a message push method in another embodiment;
[0060] Figure 4 Here is a logic diagram for priority frequency control in another embodiment;
[0061] Figure 5 A flowchart illustrating how a scheduler is used to schedule jobs in a specific application example to execute message push;
[0062] Figure 6 This is a structural block diagram of a message push device in one embodiment;
[0063] Figure 7 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation
[0064] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.
[0065] The message push method provided in this application embodiment can be applied to, for example, Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated on server 104 or placed on a cloud or other network server. First, terminal 102 sends a message push task request to server 104. The message push request contains candidate push object profile tags, the priority of the required push message, the preset push frequency, historical push data, and the message to be pushed. Server 104 receives the message push task request from terminal 102 and extracts the candidate push object profile tags, the priority of the required push message, the preset push frequency, historical push data, and the message to be pushed from the message push task request. Based on the historical push data, a blacklist of push objects to reject messages is determined. Based on the profile tags, candidate push objects are managed using profile tags; based on the priority of the required push message and the preset push frequency, frequency management is performed; and based on the push object blacklist, real-time management is performed, generating overall push strategy data. The message to be pushed is pushed based on the overall push strategy data. Furthermore, server 104 can also feed back the overall push strategy data to terminal 102. The terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, and tablets. The server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers.
[0066] In one embodiment, such as Figure 2 As shown, a message push method is provided, which can be applied to... Figure 1 Taking server 104 as an example, the following steps are included:
[0067] S100, retrieve message push task.
[0068] S200: Extract the candidate push object profile tags, the priority of the required push message, the preset push frequency, historical push data, and the message to be pushed corresponding to the message push task.
[0069] Among them, profile tags refer to factual tags, behavioral tags, credit-related tags, risk-related tags, and AI (Artificial Intelligence) model-related tags for candidate push targets; factual tags may include tags such as gender, birthday, and age; behavioral tags may include tags such as tolerance to push notifications; the priority of the required push message refers to the importance of the required push message. In this application, the priority of the required push message can be manually adjusted; the preset push frequency refers to the pre-designed frequency at which pushes can be made, such as only one message can be pushed to a specific user per day; historical push data refers to the data related to push notifications generated during the push notification process before obtaining the current push notification task, which may include push time, push recipient, and whether the push recipient rejected the message.
[0070] Specifically, the terminal sends a message push task request to the server. The message push request contains the candidate push object profile tag corresponding to the message push task, the priority of the required push message, the preset push frequency, historical push data, and the message to be pushed. The server receives the message push task request from the terminal and extracts the candidate push object profile tag corresponding to the message push task, the priority of the required push message, the preset push frequency, historical push data, and the message to be pushed from the message push task request.
[0071] S300 determines a blacklist of recipients to reject push notifications based on historical push data.
[0072] Specifically, after the server extracts the historical push data, it can determine which push recipients refuse to receive the messages pushed by the server. Refusal to receive push messages can include canceling an account, complaining about previously pushed messages, etc. In other words, a blacklist of push recipients who refuse to receive messages can be determined based on the historical push data.
[0073] S400 performs profile tag management for candidate push targets based on profile tags, frequency management based on the priority of required push messages and preset push frequency, and real-time management based on the push target blacklist, generating overall push strategy data.
[0074] In this application, "profile tag control" refers to the control rules for filtering and selecting candidates based on their profiles; "frequency control" refers to the control rules for limiting the number of marketing calls for the same push message to the same target within a given day, with higher-priority messages taking up more user calls first; "real-time control" refers to the control rules for adding users to the push target blacklist based on real-time events such as account cancellations or complaints, filtering these users out before the message is pushed; and "overall push strategy data" refers to data related to message push, including push targets and push frequency.
[0075] Specifically, after extracting the candidate push object profile tags, the priority of the required push message, the preset push frequency, historical push data, and the messages to be pushed corresponding to the message push task, and determining the blacklist of push objects to reject messages based on the historical push data, the server can manage the candidate push objects based on their profile tags. By limiting the preset push frequency of the same push message to the same push object on the same day, the priority of the message to be pushed can be determined, and the frequency of the pushed messages can be managed based on the priority. High-priority messages can be pushed to candidate push objects first, while low-priority messages will be pushed less frequently. Real-time management is also performed based on the push object blacklist. After managing the candidate push object profile tags, frequency, and real-time management, data related to message push will be generated, namely, overall push strategy data, including the actual push objects and the number of times the messages to be pushed on the same day.
[0076] S500 pushes messages to be pushed based on the overall push strategy data.
[0077] Specifically, after the server obtains the overall push strategy data, it generates a message push file based on the overall push strategy data, and then uses the message push file to push the message to be pushed to the actual push recipient.
[0078] In the aforementioned message push method, the message push task is first obtained; the candidate push object profile tags, the priority of the required push message, the preset push frequency, historical push data, and the message to be pushed are extracted according to the message push task; a blacklist of push objects to reject messages is determined based on the historical push data; profile tag control is performed on the candidate push objects based on the profile tags, frequency control is performed based on the priority of the required push message and the preset push frequency, and real-time control is performed based on the push object blacklist to generate overall push strategy data; and the message to be pushed is pushed based on the overall push strategy data. In this application, the overall push strategy data generated based on profile tag control, frequency control, and real-time control can rely on the overall push strategy data to achieve accurate push of the message to be pushed.
[0079] In one embodiment, before extracting the candidate push object profile tags, the priority of the required push message, the preset push frequency, historical push data, and the message to be pushed corresponding to the push task, the method further includes:
[0080] Obtain the type of message push task; based on the type of message push task, perform an initial screening of the message push object set to obtain the initial screening push object set; divide the initial screening push object set into candidate push objects.
[0081] For example, there are application a, application b, application c, and message push task X. Each application has its own corresponding push target group. The push target groups of all applications are regarded as the message push object set. By obtaining the type of message push task, we can know which application the message push task will be pushed to. By performing an initial screening of the message push object set, we can obtain the push object set of the corresponding application after the initial screening. Furthermore, from the push object set after the initial screening, candidate push objects can be divided proportionally.
[0082] In this embodiment, by obtaining the type of message push task, the message push object set can be initially screened, thereby finding the message push object more accurately.
[0083] In one embodiment, dividing the candidate push objects from the initially screened set of push objects includes:
[0084] The initial set of target users is further subdivided proportionally to obtain candidate target users.
[0085] Specifically, after obtaining the set of push objects after initial screening, the method for dividing the set of push objects into candidate push objects is to subdivide the set of push objects according to a certain proportion.
[0086] Furthermore, the proportion can be subdivided in a way that allows for equal division using an AB-test approach.
[0087] In this embodiment, by proportionally subdividing the set of push targets after initial screening, more accurate message push targets can be obtained.
[0088] In one embodiment, the method further includes: proportionally subdividing the initial screening set of push objects to obtain candidate push objects and a control group of candidate push objects.
[0089] The control group refers to a randomly selected subset of the experimental subjects, in which individuals receive no special treatment.
[0090] Specifically, by proportionally subdividing the set of push targets after the initial screening, not only can candidate push targets be obtained, but another part of the push target set after proportional subdivision can also be used as a control group of candidate push targets.
[0091] Push the message to be pushed to the candidate push object control group, and collect the first message push feedback data from the candidate push object control group.
[0092] In this application, the first message push feedback data refers to the behavioral response of the push object in the candidate push object control group to the push message after receiving it.
[0093] Specifically, since the candidate push object control group is a control group for the candidate push objects and is used to judge the excellence of the overall push strategy data, the overall push strategy data is used as a variable that needs to be controlled. At this time, there is no need to generate the overall push strategy data. The message to be pushed is directly pushed to the candidate push object control group. After the push is completed, the push objects in the push object control group will make corresponding behavioral responses according to the message to be pushed. This behavioral response is collected and used as the first message push feedback data, which is fed back to the server.
[0094] Furthermore, behavioral responses can include clicking to enter message browsing, clicking "not interested in message," closing the message, and complaining about the message.
[0095] When the push of the message to be pushed is completed based on the overall push strategy data, the second message push feedback data is collected.
[0096] In this application, the second message push feedback data refers to the behavioral response of the push object in the candidate push object to the push message after receiving the push message.
[0097] Specifically, after the process of pushing messages based on the overall push strategy data is completed, the actual push objects in the candidate push objects will make corresponding behavioral responses according to the messages to be pushed. At this time, these behavioral responses are collected and used as second message push feedback data, which is then fed back to the server.
[0098] By comparing the feedback data from the first push notification with the feedback data from the second push notification, we obtain the overall push strategy data evaluation data.
[0099] In this application, strategy data evaluation data refers to data used to evaluate the quality of the overall push strategy data.
[0100] Specifically, the first message push feedback data is the feedback data obtained by pushing the message to the candidate push object based on the overall push strategy data, and the second message push feedback data is the feedback data obtained by directly pushing the message to the candidate push object without generating the overall push strategy data. By comparing the obtained first message push feedback data and second message push feedback data, the merits of the overall push strategy data, the only variable in the generation process of these two types of push feedback data, can be evaluated, and the overall push strategy data evaluation data can be obtained.
[0101] In this embodiment, by obtaining a control group of candidate push objects, which does not generate overall push strategy data, and by comparing the push feedback data after pushing with the control group of candidate push objects, the overall push strategy data evaluation data can be obtained, thereby determining the advantages of the overall push strategy data.
[0102] In one embodiment, such as Figure 3 As shown, S400 includes:
[0103] S420: Based on profile tags, perform profile tag management on candidate push objects to obtain push strategy data after profile tag management.
[0104] Specifically, profile tags are used to manage candidate push targets, including factual tags, behavioral tags, credit-related tags, risk-related tags, and AI model-related tags.
[0105] Taking gender in fact labels as an example for profile label control, if a certain message push task is suitable to be pushed to women, then it is necessary to control the profile labels of the candidate push objects based on the gender represented by the profile labels of the candidate push objects. Select the female candidate push objects as the objects to be pushed. That is, the push strategy data obtained after profile label control includes the selection of female candidates.
[0106] S440: After obtaining the push strategy data after profile tag control, frequency control is performed based on the priority of the required push messages and the preset push frequency to obtain the push strategy data after frequency control.
[0107] Specifically, after obtaining the target audience for push notifications after obtaining the profile tags, frequency control can be performed based on the priority of the required push messages and the preset push frequency to obtain push strategy data after frequency control. That is, the push strategy data after frequency control includes determining the target audience for the push message to be pushed based on the preset push frequency and the priority of the required push message, as well as the number of times the push message is pushed to the target audience on the same day.
[0108] For example, there are two messages, message A and message B, to be pushed out. The candidate push recipients for messages A and B overlap, including an overlapping user 1. This overlapping user 1 only receives messages A and B. The preset push frequency for this overlapping user 1 is once a day. Therefore, before pushing messages to overlapping user 1, the push frequency needs to be controlled. It needs to determine which message, A or B, has a higher priority for overlapping user 1. If message A has a higher priority than message B, then the preset push frequency for overlapping user 1 is allocated to message A. Send to overlapping user 1; if overlapping user 1 only receives messages A and B, but the preset push frequency for overlapping user 1 is three times a day, determine which message A or B has a higher priority for overlapping user 1. Based on the priority, the number of times messages A and B are pushed can be determined. However, the number of times the message with higher priority is pushed is greater than the number of times the message with lower priority is pushed. That is, when the preset push frequency is three times a day, when the number of times message A is pushed accounts for two-thirds of the preset push frequency, the number of times message B is pushed accounts for the remaining one-third. At this time, both messages A and B need to be pushed out. It is not possible to push message A without pushing message B.
[0109] Furthermore, if the priority levels are the same, the priority among users at the same level is determined by the predicted time of push notification to overlapping user 1; additionally, if... Figure 4 The implementation logic diagram for priority and frequency control is shown. The implementation logic for priority and frequency control can be as follows: Figure 4 The right image shows the result output table, and the left image shows the candidate push object table. Push objects that have been successfully pushed are written to the result output table. The candidate push objects and the result output table are then compared.
[0110] 1. Filter out the candidate push object IDs that are not in the results output table, i.e., usr-id;
[0111] 2. Group the selected candidate push object IDs. Sort the messages to be pushed to the candidate push objects in the same group according to priority. Take the message with higher priority and push it to the candidate push object. Filter out the push objects to which the message to be pushed is to be pushed.
[0112] 3. Rewrite the target audience of the selected messages to be pushed into the result output table;
[0113] 4. Generate push strategy data after frequency control based on the output table.
[0114] S460, after obtaining the push strategy data after frequency control, performs real-time control based on the push target blacklist, and obtains push strategy data after real-time control.
[0115] Specifically, once the frequency-controlled push targets are obtained, if the frequency-controlled push targets are added to the push target blacklist on the day the push is to be sent, i.e., the candidate push targets are users who have closed their accounts or complained about, the frequency-controlled push targets are controlled in real time based on the push target blacklist. The push targets in the push target blacklist are filtered out, and messages are no longer pushed to the push targets in the push target blacklist. The real-time controlled push strategy data is obtained, which includes the real-time filtering of the candidate push targets that cannot be pushed based on the push target blacklist.
[0116] S480 generates overall push strategy data based on push strategy data after profile tag control, push strategy data after frequency control, and push strategy data after real-time control.
[0117] Specifically, push strategy data can be generated based on push strategy data after profile tag control, push strategy data after frequency control, and push strategy data after real-time control. The overall push strategy data includes profile tag control for candidate push targets based on profile tags, frequency control based on the priority of required push messages and preset push frequency, and real-time control based on the push target blacklist.
[0118] In this embodiment, by controlling the profile tags, frequency, and real-time control of candidate push targets, accurate push can be achieved without excessively disturbing the push targets.
[0119] In one embodiment, the method further includes:
[0120] Obtain the list of exemptable objects; when a candidate push object is in the list of exemptable objects, perform real-time control on the candidate push object in the list of exemptable objects based on the push object blacklist, and generate the first push strategy data.
[0121] Among them, exemptable objects refer to objects that can be exempted from the restrictions of the rules. In this application, exemptable objects refer to objects that are exempt from portrait label control and frequency control. Exemptable objects do not need to be subject to portrait label control and frequency control.
[0122] Specifically, the list of exemptable targets is obtained, and compared with the candidate push targets. When a candidate push target is in the list of exemptable targets, no profile tag control or frequency control is applied to the candidate push target in the list of exemptable targets. At this time, if the candidate push target enters the push target blacklist on the same day, that is, the candidate push target is a closed user or a complaining user, the candidate push target is controlled in real time based on the push target blacklist. The candidate push targets in the push target blacklist are filtered, and no more messages are pushed to the candidate push targets in the push target blacklist. The first push strategy data is obtained. The first push strategy data includes filtering out the objects that cannot be pushed to in the candidate push targets in real time based on the push target blacklist.
[0123] When the candidate push target is not on the list of exemptable targets, the system performs profile tag control on the candidate push target, frequency control based on the priority of the required push message and the preset push frequency, and real-time control based on the push target blacklist, generating second push strategy data.
[0124] Specifically, when the candidate push target is not on the list of exemptable targets, that is, when the candidate push target is not restricted by the list of exemptable targets, the second push strategy data can be generated by controlling the candidate push target based on profile tags, controlling the frequency based on the priority of the required push message and the preset push frequency, and controlling the push target in real time based on the push target blacklist. At this time, the second push strategy data includes the overall push strategy data, which includes controlling the candidate push target based on profile tags, controlling the frequency based on the priority of the required push message and the preset push frequency, and controlling the push target in real time based on the push target blacklist.
[0125] Collect the data from the first push strategy and the second push strategy to generate the overall push strategy data.
[0126] Among them, aggregation refers to the aggregation of the first push strategy data and the second push strategy data.
[0127] Specifically, by summarizing and centralizing the first push strategy data and the second push strategy data, overall push strategy data can be generated.
[0128] In this application, by obtaining the list of exemptable users, the candidate push targets in the list can be exempted from profile tag control and frequency control, so as to achieve accurate push to the target target.
[0129] In one embodiment, such as Figure 5 As shown, message pushing can be performed by using a scheduler to schedule jobs.
[0130] Job scheduling refers to reviewing whether the system can meet the resource requirements of user jobs based on the information in the job control block, and selecting certain jobs from the backup queue in the external storage according to a certain algorithm, loading them into memory, creating processes for them, allocating necessary resources, and then inserting the newly created processes into the ready queue for execution.
[0131] Specifically, message push can be divided into five tasks, including initial screening, subdivision, three-level control, document generation, and message push. Each task is executed through scheduling, and the execution time of each task starts immediately from the start of the day's tasks.
[0132] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.
[0133] Based on the same inventive concept, this application also provides a message push device for implementing the message push method described above. The solution provided by this device is similar to the implementation described in the above method; therefore, the specific limitations in one or more message push device embodiments provided below can be found in the limitations of the message push method described above, and will not be repeated here.
[0134] In one embodiment, such as Figure 6 As shown, a message push device is provided, including: a task acquisition module 100, a data extraction module 200, a blacklist determination module 300, a three-level control module 400, and a message push module 500, wherein:
[0135] The task acquisition module 100 is used to acquire message push tasks.
[0136] The data extraction module 200 is used to extract the candidate push object profile tags, the priority of the required push message, the preset push frequency, historical push data, and the message to be pushed corresponding to the message push task.
[0137] The blacklist determination module 300 is used to determine the blacklist of recipients of rejected messages based on historical push data.
[0138] The three-level control module 400 is used to control candidate push targets based on profile tags, control frequency based on the priority of required push messages and preset push frequency, and control push targets in real time based on a blacklist of push targets, generating overall push strategy data.
[0139] The message push module 500 is used to push messages to be pushed based on the overall push strategy data.
[0140] In one embodiment, the system further includes a preliminary screening module, which is used to obtain the type of message push task; perform a preliminary screening of the message push object set based on the type of message push task to obtain a preliminary screening push object set; and divide the preliminary screening push object set into candidate push objects.
[0141] In one embodiment, the initial screening module is further used to proportionally subdivide the set of push objects after initial screening to obtain candidate push objects.
[0142] In one embodiment, the system further includes a comparison push module, which is used to proportionally subdivide the set of push objects after initial screening to obtain candidate push objects and a candidate push object control group; push the message to be pushed to the candidate push object control group and collect the first message push feedback data fed back by the candidate push object control group; when the push of the message to be pushed based on the overall push strategy data is completed, collect the feedback of the second message push feedback data; and compare the first message push feedback data and the second message push feedback data to obtain the overall push strategy data evaluation data.
[0143] In one embodiment, the three-level control module 400 is further configured to perform profile tag control on candidate push objects based on profile tags to obtain push strategy data after profile tag control; after obtaining the push strategy data after profile tag control, frequency control is performed based on the priority of the required push message and the preset push frequency to obtain push strategy data after frequency control; after obtaining the push strategy data after frequency control, real-time control is performed based on the push object blacklist to obtain push strategy data after real-time control; and push overall strategy data is generated based on the push strategy data after profile tag control, the push strategy data after frequency control, and the push strategy data after real-time control.
[0144] In one embodiment, the system further includes a push exemption module, which is used to obtain a list of exemptable objects; when a candidate push object is in the list of exemptable objects, the system updates the overall push strategy data; and pushes the message to be pushed based on the updated overall push strategy data.
[0145] Each module in the aforementioned message push device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.
[0146] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 7 As shown, this computer device includes a processor, memory, input / output interfaces (I / O), and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media. The database stores overall push strategy data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When executed by the processor, the computer program implements a message push method.
[0147] Those skilled in the art will understand that Figure 7 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.
[0148] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.
[0149] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.
[0150] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.
[0151] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data shall comply with the relevant laws, regulations and standards of the relevant countries and regions.
[0152] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.
[0153] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above 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.
[0154] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.
Claims
1. A message pushing method, characterized by, The method includes: Get message push task; Obtain the type of the message push task; based on the type of the message push task, perform an initial screening of the message push object set to obtain the initial screening push object set; further subdivide the initial screening push object set proportionally to obtain candidate push objects and a candidate push object control group, wherein the candidate push object control group is randomly selected; Extract the candidate push object profile tags, the priority of the required push message, the preset push frequency, historical push data, and the message to be pushed corresponding to the message push task; A blacklist of recipients to reject messages is determined based on the historical push data. Based on the profile tags, candidate push targets are managed using profile tags to obtain push strategy data after profile tag management; after obtaining the push strategy data after profile tag management, frequency management is performed based on the priority of the required push messages and the preset push frequency to obtain push strategy data after frequency management; after obtaining the push strategy data after frequency management, real-time management is performed based on the push target blacklist to obtain push strategy data after real-time management; based on the push strategy data after profile tag management, the push strategy data after frequency management, and the push strategy data after real-time management, overall push strategy data is generated. The message to be pushed is pushed based on the overall push strategy data; The message to be pushed is pushed to the candidate push object control group, and the first message push feedback data fed back by the candidate push object control group is collected; when the push of the message to be pushed is completed based on the overall push strategy data, the feedback of the second message push feedback data is collected, and the overall push strategy data evaluation data is obtained by comparing the first message push feedback data and the second message push feedback data, wherein the overall push strategy data evaluation data is used to judge the merits of the overall push strategy data. The frequency-controlled push strategy data includes determining the target audience for the push message and the number of times the push message will be pushed to that audience on the same day, based on a preset push frequency and the priority of the required push message; the frequency control based on the priority of the required push message and the preset push frequency to obtain the frequency-controlled push strategy data includes: There are two messages, A and B, to be pushed out. The candidate push objects for messages A and B overlap. One of the overlapping objects is an overlapping user 1, and overlapping user 1 only receives messages A and B. If the preset push frequency of overlapping user 1 is once a day, then message A will be pushed to overlapping user 1 if message A has a higher priority than message B. If the preset push frequency of overlapping user 1 is three times a day, then the number of times messages A and B are pushed is determined according to their priorities, with the higher-priority message being pushed more often than the lower-priority message.
2. The method of claim 1, wherein, The method further includes: Obtain the list of eligible individuals; When the candidate push object is in the list of exemptable objects, the candidate push object in the list of exemptable objects is managed in real time based on the push object blacklist, and first push strategy data is generated. When the candidate push object is not in the list of exemptable objects, the candidate push object is controlled by the profile tag based on the profile tag, the frequency is controlled based on the priority of the required push message and the preset push frequency, and the push object blacklist is controlled in real time to generate second push strategy data; The first push strategy data and the second push strategy data are collected to generate overall push strategy data.
3. The method of claim 1, wherein, The method further includes: If the priority levels are the same, the priority among users of the same level is determined by the predicted time when the push is sent to overlapping user 1.
4. The method of claim 1, wherein, The profile tags refer to factual tags, behavioral tags, credit-related tags, risk-related tags, and AI model-related tags for the candidate push targets; the behavioral tags include tolerance tags for push notifications; the priority of the push notification refers to the importance of the push notification.
5. The method of claim 1, wherein, The historical push data refers to the data related to message push generated during the message push process before obtaining this message push task. The historical push data includes the push time, the push recipient, and whether the push recipient rejected the message.
6. A message pushing apparatus characterized by comprising: The device includes: The task acquisition module is used to acquire message push tasks; The initial screening module is used to obtain the type of the message push task; based on the type of the message push task, the message push object set is initially screened to obtain the initial screened push object set; The comparison push module is used to proportionally subdivide the set of push objects after the initial screening to obtain candidate push objects and candidate push object control group, wherein the candidate push object control group is randomly selected; The data extraction module is used to extract the candidate push object profile tags, the priority of the required push message, the preset push frequency, historical push data, and the message to be pushed corresponding to the message push task. The blacklist determination module is used to determine a blacklist of recipients for rejected messages based on the historical push data. The three-level control module is used to perform profile tag control on candidate push objects based on the profile tags, and obtain push strategy data after profile tag control; after obtaining the push strategy data after profile tag control, frequency control is performed based on the priority of the required push message and the preset push frequency, and frequency control push strategy data is obtained; after obtaining the push strategy data after frequency control, real-time control is performed based on the push object blacklist, and real-time control push strategy data is obtained; based on the push strategy data after profile tag control, the push strategy data after frequency control, and the push strategy data after real-time control, overall push strategy data is generated. The message push module is used to push the message to be pushed based on the overall push strategy data; The push module is used to push the message to be pushed to the candidate push object control group and collect the first message push feedback data fed back by the candidate push object control group; when the push of the message to be pushed is completed based on the overall push strategy data, the second message push feedback data is collected, and the overall push strategy data evaluation data is obtained by comparing the first message push feedback data and the second message push feedback data, wherein the overall push strategy data evaluation data is used to judge the merits of the overall push strategy data. The frequency-controlled push strategy data includes determining the target audience for the message to be pushed and the number of times the message will be pushed to that target audience that day, based on the preset push frequency and the priority of the message to be pushed. The three-level control module is also used when there are two messages, message A and message B, to be pushed out, and the candidate target audiences for message A and message B overlap. Among the overlapping target audience is overlapping user 1, and overlapping user 1 only receives message A and message B. If the preset push frequency of overlapping user 1 is once a day, then message A will be pushed to overlapping user 1 if message A has a higher priority than message B. If the preset push frequency of overlapping user 1 is three times a day, then the number of times messages A and B will be pushed will be determined based on their priorities, wherein the number of times the message with higher priority is pushed is greater than the number of times the message with lower priority is pushed.
7. The apparatus of claim 6, wherein, The device also includes a push exemption module, used to obtain a list of exemptable objects; when the candidate push object is in the list of exemptable objects, the device performs real-time control on the candidate push object in the list of exemptable objects based on the push object blacklist, and generates first push strategy data. When the candidate push object is not in the list of exemptable objects, the candidate push object is controlled by the profile tag based on the profile tag, the frequency is controlled based on the priority of the required push message and the preset push frequency, and the push object blacklist is controlled in real time to generate second push strategy data; the first push strategy data and the second push strategy data are collected to generate overall push strategy data.
8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.
9. A computer readable storage medium having stored thereon a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.
10. A computer program product comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.