Method, device, server and medium for generating information push mode

Abstract

The embodiments of the present disclosure disclose a method, device, server, and medium for generating an information push method. A specific implementation of the method includes: in response to detecting an abnormality in target data, selecting at least one factor from factors associated with the abnormal target data as a key factor, wherein the target data is used to characterize the activity level of the user group, and the factor corresponds to the category to which the users in the user group belong; obtaining information associated with the users in the user group corresponding to the key factor as feature data; selecting user-side information from the user-side information set corresponding to the key factor to form a candidate user-side information set; and generating an information push method corresponding to the user-side information in the candidate user-side information set based on the feature data. This implementation enriches the information push method, improves the pertinence of information push, and improves efficiency.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the field of computer technology, and more particularly to a method, device, server, and medium for generating an information push mode. Background Art

[0002] With the continuous development of Internet technology, the amount of information has grown exponentially, and information push technology has been increasingly widely used.

[0003] Existing methods for pushing information on platforms often require manual determination of details such as target audience, push channels, push timing, and push content. However, information push methods based on manual judgment are often time-consuming and inefficient. Summary of the Invention

[0004] The embodiments of the present disclosure provide a method, device, server, and medium for generating an information push mode.

[0005] In a first aspect, an embodiment of the present disclosure provides a method for generating an information push method, the method comprising: in response to detecting an abnormality in target data, selecting at least one factor from factors associated with the abnormal target data as a key factor, wherein the target data is used to characterize the activity level of a user group, and the factor corresponds to the category to which the users in the user group belong; obtaining information associated with the users in the user group corresponding to the key factor as feature data; selecting user-side information from the user-side information set corresponding to the key factor to form a candidate user-side information set; and generating an information push method corresponding to the user-side information in the candidate user-side information set based on the feature data.

[0006] In some embodiments, the above-mentioned information push method is used to indicate that information is pushed at a specific push time; and the method also includes: in response to determining that the specific push time has been reached, obtaining user behavior information corresponding to the user-end information corresponding to the information push method; based on the matching of the behavior information with the preset rules, determining whether to correct the specific push time.

[0007] In some embodiments, the above-mentioned information push method is used to indicate that information is pushed through a specific push channel; and the above-mentioned information push method corresponding to the user-end information in the candidate user-end information set based on the characteristic data includes: generating an optional channel information set based on a preset channel information set and the current cost constraint; determining the response probability of the user-end information in the candidate user-end information set and the channel information in the optional channel information set corresponding to the characteristic data; for the user-end information in the candidate user-end information set, determining the channel corresponding to the channel information with the highest response probability as the specific push channel corresponding to the user-end information.

[0008] In some embodiments, the above-mentioned channel information set includes a telephone sales channel; and the method also includes: in response to determining that the information push method is used to indicate that information is pushed through the telephone sales channel, selecting matching marketing personnel information from a preset marketing personnel information list, wherein the marketing personnel information list includes the marketing personnel's identification and corresponding feature tags; sending an information push task corresponding to the information push method to the terminal corresponding to the matching marketing personnel information.

[0009] In some embodiments, in response to detecting an abnormality in the target data, at least one factor is selected as a key factor from the factors associated with the abnormal target data, including: obtaining the target data within a preset time period; in response to determining that the fluctuation of the target data exceeds a preset fluctuation threshold, determining the influence corresponding to each factor associated with the target data; and selecting a target number of factors whose influence meets preset conditions as key factors.

[0010] In some embodiments, the above-mentioned user-associated information includes user behavior data and historical information push data within the current preset time period; and the above-mentioned information push method corresponding to the user-side information in the candidate user-side information set generated based on the feature data includes: based on the feature data, using a pre-trained channel prediction model to generate push channel information; based on the feature data, using a pre-trained time prediction model to generate push time information; based on the feature data, using a pre-trained content prediction model to generate push content information; based on the feature data, generating an information push method based on the push channel information, push time information and push content information.

[0011] In some embodiments, the above-mentioned selection of user-side information from the user-side information set corresponding to the key factors to form a candidate user-side information set includes: selecting a matching information push target from a preset correspondence table based on the key factors; inputting the feature data into a pre-trained response probability generation model to generate the response probability of the user corresponding to the feature data to the matching information push target; and selecting user-side information whose response probability is greater than a preset response threshold from the user-side information set corresponding to the key factors to form a candidate user-side information set.

[0012] In some embodiments, the method also includes: establishing a folder corresponding to the matching information push target; generating an information push task based on the generated information push method; storing the generated information push task in the corresponding folder; and in response to determining that there is a folder that matches the key factors, executing the information push task in the matching folder.

[0013] In some embodiments, the above-mentioned execution of the information push task in the matching folder includes: dividing the user-end information in the information push task in the matching folder into a push group and a control group; pushing information to the user end corresponding to the push group in the generated information push method; generating information push evaluation information for the push group and the control group; in response to determining that the information push evaluation information indicates that the push effect does not meet the requirements, executing other information push tasks under the information push target.

[0014] In the second aspect, an embodiment of the present disclosure provides a device for generating an information push method, the device comprising: a first selection unit, configured to, in response to detecting an abnormality in target data, select at least one factor from factors associated with the abnormal target data as a key factor, wherein the target data is used to characterize the activity level of a user group, and the factor corresponds to the category to which the users in the user group belong; a first acquisition unit, configured to acquire information associated with users in the user group corresponding to the key factor as feature data; a second selection unit, configured to select user-side information from the user-side information set corresponding to the key factor to form a candidate user-side information set; a first generation unit, configured to generate an information push method corresponding to the user-side information in the candidate user-side information set based on the feature data.

[0015] In some embodiments, the above-mentioned information push method is used to indicate that information is pushed at a specific push time; the device also includes: a second acquisition unit, configured to obtain user behavior information corresponding to the user-end information corresponding to the information push method in response to determining that the specific push time has been reached; a determination unit, configured to determine whether to correct the specific push time based on the matching of the behavior information with the preset rules.

[0016] In some embodiments, the above-mentioned information push method is used to indicate that information is pushed through a specific push channel; the above-mentioned first generation unit is further configured to: generate an optional channel information set based on a preset channel information set and a current cost constraint; determine the response probability corresponding to the user-end information in the candidate user-end information set and the channel information in the optional channel information set based on the feature data; for the user-end information in the candidate user-end information set, determine the channel corresponding to the channel information with the highest response probability as the specific push channel corresponding to the user-end information.

[0017] In some embodiments, the above-mentioned channel information set includes a telephone sales channel; the device also includes: a third selection unit, configured to select matching marketing personnel information from a preset marketing personnel information list in response to determining an information push method for indicating that information is pushed through a telephone sales channel, wherein the marketing personnel information list includes the marketing personnel's identification and corresponding feature tags; a sending unit, configured to send an information push task corresponding to the information push method to a terminal corresponding to the matching marketing personnel information.

[0018] In some embodiments, the above-mentioned first selection unit is further configured to: obtain target data within a preset time period; in response to determining that the fluctuation of the target data exceeds a preset fluctuation threshold, determine the influence corresponding to each factor associated with the target data; and select a target number of factors whose influence meets the preset conditions as key factors.

[0019] In some embodiments, the user-associated information includes user behavior data and historical information push data within a current preset time period; the first generation unit is further configured to: generate push channel information based on feature data using a pre-trained channel prediction model; generate push time information based on feature data using a pre-trained time prediction model; generate push content information based on feature data using a pre-trained content prediction model; generate information push method based on push channel information, push time information and push content information.

[0020] In some embodiments, the above-mentioned second selection unit is further configured to: select a matching information push target from a preset correspondence table based on key factors; input the feature data into a pre-trained response probability generation model to generate the response probability of the user corresponding to the feature data to the matching information push target; select user-side information with a response probability greater than a preset response threshold from the user-side information set corresponding to the key factors to form a candidate user-side information set.

[0021] In some embodiments, the device also includes: an establishment unit, configured to establish a folder corresponding to the matching information push target; a second generation unit, configured to generate an information push task based on the generated information push method; a storage unit, configured to store the generated information push task to the corresponding folder; and an execution unit, configured to execute the information push task in the matching folder in response to determining that there is a folder that matches the key factor.

[0022] In some embodiments, the above-mentioned execution unit is further configured to: divide the user-end information in the information push task in the matching folder into a push group and a control group; push information to the user end corresponding to the push group in the generated information push method; generate information push evaluation information for the push group and the control group; in response to determining that the information push evaluation information indicates that the push effect does not meet the requirements, execute other information push tasks under the information push target.

[0023] In a third aspect, an embodiment of the present disclosure provides a server comprising: one or more processors; a storage device on which one or more programs are stored; when the one or more programs are executed by one or more processors, the one or more processors implement the method described in any implementation manner in the first aspect.

[0024] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable medium having a computer program stored thereon, which, when executed by a processor, implements the method described in any implementation manner in the first aspect.

[0025] The methods, devices, servers, and media for generating information push methods provided by the embodiments of the present disclosure use target data anomalies as a condition for generating the information push method, and select a suitable target population based on the factors causing the target data anomalies to generate the information push method, thereby enriching the information push method, improving the pertinence of the information push, and enhancing the efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Other features, objects and advantages of the present disclosure will become more apparent from a reading of the detailed description of non-limiting embodiments made with reference to the following drawings:

[0027] Figure 1 is an exemplary system architecture diagram in which an embodiment of the present disclosure may be applied;

[0028] Figure 2 is a flow chart of an embodiment of a method for generating an information push mode according to the present disclosure;

[0029] Figure 3 is a schematic diagram of an application scenario of a method for generating an information push mode according to an embodiment of the present disclosure;

[0030] Figure 4 is a flowchart of another embodiment of a method for generating an information push mode according to the present disclosure;

[0031] Figure 5 is a structural diagram of an embodiment of an apparatus for generating an information push mode according to the present disclosure;

[0032] Figure 6It is a schematic structural diagram of an electronic device suitable for implementing the embodiments of the present disclosure. DETAILED DESCRIPTION

[0033] The present disclosure will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are intended only to illustrate the relevant invention and are not intended to limit the invention. It should also be noted that, for ease of description, only portions relevant to the relevant invention are shown in the accompanying drawings.

[0034] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in the present disclosure may be combined with each other. The present disclosure will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0035] Figure 1 An exemplary architecture 100 is shown to which the method for generating an information push mode or the apparatus for generating an information push mode of the present disclosure can be applied.

[0036] like Figure 1 As shown, system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. Network 104 is a medium for providing communication links between terminal devices 101, 102, 103 and server 105. Network 104 may include various connection types, such as wired or wireless communication links or fiber optic cables.

[0037] The terminal devices 101, 102, and 103 interact with the server 105 via the network 104 to receive or send messages, etc. Various communication client applications can be installed on the terminal devices 101, 102, and 103, such as web browser applications, information applications, shopping applications, search applications, instant messaging tools, email clients, etc.

[0038] The terminal devices 101, 102, and 103 can be hardware or software. When the terminal devices 101, 102, and 103 are hardware, they can be various electronic devices with display screens and supporting human-computer interaction, including but not limited to smartphones, tablet computers, laptop computers, and desktop computers. When the terminal devices 101, 102, and 103 are software, they can be installed in the electronic devices listed above. They can be implemented as multiple software or software modules (for example, software or software modules for providing distributed services), or they can be implemented as a single software or software module. No specific limitation is made here.

[0039] The server 105 may be a server that provides various services, such as a backend server that supports information applications on the terminal devices 101, 102, and 103. The backend server may generate an information push method based on the characteristic data corresponding to the terminal devices 101, 102, and 103 (e.g., sending a text message notification to the terminal device 101 at 12:00), and may also push information based on the generated information push method.

[0040] It should be noted that the server can be either hardware or software. When the server is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or as a single server. When the server is software, it can be implemented as multiple software or software modules (e.g., software or software modules for providing distributed services), or as a single software or software module. No specific limitations are given here.

[0041] It should be noted that the method for generating the information push mode provided in the embodiments of the present disclosure is generally executed by the server 105 , and accordingly, the device for generating the information push mode is generally set in the server 105 .

[0042] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is merely illustrative. Any number of terminal devices, networks and servers may be provided as required.

[0043] Continue to refer Figure 2 , shows a process 200 of an embodiment of a method for generating an information push mode according to the present disclosure. The method for generating an information push mode includes the following steps:

[0044] Step 201 : In response to detecting that target data is abnormal, at least one factor is selected as a key factor from factors associated with the abnormal target data.

[0045] In this embodiment, the execution subject of the method for generating the information push mode (such as Figure 1The server 105 shown) can detect whether the target data is abnormal in various ways. Among them, the above-mentioned target data can be used to characterize the activity level of the user group. As an example, the above-mentioned target data may include but is not limited to at least one of the following: DAU (daily active user, daily active user number), MAU (monthly active users, monthly active user number), PCU (peak concurrent users, peak concurrent users), DNU (Daily New Users, daily new users), ACU (Average concurrent users, average concurrent users), UV (unique visitor, independent visitor number), PV (PageView, pageviews). In response to detecting that the above-mentioned target data is abnormal, the above-mentioned execution entity can select at least one factor as a key factor from the factors associated with the above-mentioned abnormal target data in various ways. Among them, the above-mentioned factors can correspond to the categories to which the users in the above-mentioned user group belong. Specifically, the above-mentioned execution entity can obtain a preset relationship table in advance, and the above-mentioned preset relationship table can be used to characterize the correspondence between the target data and the associated factors. As an example, the above-mentioned target data can be, for example, the number of independent visitors in the current time period. The factors associated with the above target data may include the number of daily active users, the number of monthly active users and the number of daily new users. The categories of users corresponding to the above factors may include daily active users, monthly active users and new users. As another example, the above target data may be, for example, the number of daily active users. The factors associated with the above target data may include the number of male users, the number of female users, the number of young users, the number of middle-aged users, etc. The categories of users corresponding to the above factors may include men (for example, users who select the gender "male" when registering as users), women, young people (for example, users who select the age group of "18-25 years old" when registering as users), and middle-aged people (for example, users who select the age group of "35-45 years old" when registering as users). Thus, the above-mentioned execution entity can use the above-mentioned preset relationship table to parse abnormal target data to select key factors.

[0046] In some optional implementations of this embodiment, the execution entity may further select at least one factor as a key factor from factors associated with the abnormal target data in response to detecting the abnormal target data according to the following steps:

[0047] The first step is to obtain the target data within the preset time period.

[0048] In these implementations, the execution entity may obtain target data within a preset time period from a local or communication-connected electronic device (e.g., an information push data monitoring platform). As an example, the preset time period may be approximately one hour, approximately 15 minutes, etc.

[0049] In the second step, in response to determining that the fluctuation of the target data exceeds a preset fluctuation threshold, the influence degree corresponding to each factor associated with the target data is determined.

[0050] In these implementations, in response to determining that the fluctuation of the target data obtained in the first step exceeds a preset fluctuation threshold, the execution entity may determine the influence of each factor associated with the target data. As an example, the execution entity may first obtain each factor associated with the target data. Thereafter, the execution entity may determine the influence of each factor associated with the target data on the target data using a Gini Index or information gain, for example.

[0051] The third step is to select a target number of factors whose influence meets the preset conditions as key factors.

[0052] In these implementations, the execution entity may select a target number of factors that meet preset conditions from the influences determined in the second step as key factors. As an example, the execution entity may select the three factors with the greatest influence as key factors. As another example, the execution entity may select factors whose influence exceeds a preset influence threshold as key factors.

[0053] Based on the above optional implementation methods, this solution can automatically determine the factors associated with abnormal target data, thereby providing a strong reference basis for determining the target user group for information push.

[0054] In some optional implementations of this embodiment, the above-mentioned information push method can be used to indicate that information is pushed at a specific push time.

[0055] In some optional implementations of this embodiment, the above-mentioned information push method can be used to indicate that information is pushed through a specific push channel.

[0056] Step 202: Acquire information associated with users in the user group corresponding to the key factor as feature data.

[0057] In this embodiment, the execution entity may obtain information associated with users in the user group corresponding to the key factors as feature data through various means. For example, the user-related information may include personal attribute information such as user tags, birthdays, and hobbies. As another example, the user-related information may also include the user's frequent login time, search content, and comment content.

[0058] In some optional implementations of this embodiment, the user-related information may include user behavior data and historical information push data within a preset time period. The user behavior data may include, for example, the user's frequent login time in the past week, the user's most viewed topics in the past week, the number of comments and likes by the user in the past week, etc. The historical information push data may include historical information push effect data and historical information push response data. The historical information push effect data may include, for example, data on successfully reaching the user through phone calls, text messages, in-app notification messages, etc. The historical information push response data may include, for example, data on the user clicking on links contained in text messages, clicking to view notification messages sent in apps, etc.

[0059] Step 203 : Select user terminal information from the user terminal information set corresponding to the key factor to form a candidate user terminal information set.

[0060] In this embodiment, the execution entity may select user-side information from the corresponding user-side information set in various ways to form a candidate user-side information set. For example, the execution entity may randomly select user-side information from the corresponding user-side information set to form a candidate user-side information set. For another example, the execution entity may select matching user-side information from the user-side information set corresponding to the key factors selected in the first step based on the feature data obtained in the second step to form a candidate user-side information set.

[0061] Step 204: Generate an information push mode corresponding to the user terminal information in the candidate user terminal information set based on the feature data.

[0062] In this embodiment, based on the characteristic data acquired in step 202, the execution entity may generate, through various methods, an information push method corresponding to the user-end information in the candidate user-end information set formed in step 203. The information push method may include, but is not limited to, at least one of the following: an information push channel, an information push time, and an information push content. The information push channel may include, but is not limited to, at least one of the following: a phone notification, a text message notification, and an in-app message notification.

[0063] In some optional implementations of this embodiment, based on the information push mode indicating that information is pushed through a specific push channel, the execution entity may further generate, based on the characteristic data, an information push mode corresponding to the user-end information in the candidate user-end information set according to the following steps:

[0064] In the first step, an optional channel information set is generated based on the preset channel information set and the current cost constraint.

[0065] In these implementations, the execution entity may first obtain a preset channel information set and a current cost constraint. As an example, the preset channel information set may include a telephone notification channel, a text message notification channel, and an in-app message notification channel. Generally speaking, the costs corresponding to the above three channels decrease in sequence. The execution entity may determine whether the current information push cost meets the cost constraint, thereby eliminating channels that do not meet the cost constraint (such as telephone notification channels) and generating an optional channel information set.

[0066] In the second step, based on the feature data, the response probability corresponding to the user terminal information in the candidate user terminal information set and the channel information in the optional channel information set is determined.

[0067] In these implementations, based on the characteristic data, the execution entity can determine the response probabilities corresponding to the user-end information in the candidate user-end information set and the channel information in the optional channel information set in various ways. As an example, the candidate user-end information set may include user-end information A, user-end information B, and user-end information C. The execution entity can input the characteristic data corresponding to user-end information A into a pre-trained channel prediction model, thereby obtaining the SMS notification channel response probability and the in-application message notification channel response probability corresponding to user-end information A. The channel prediction model can be, for example, a model trained by machine learning methods such as XGBOOST. Similarly, the SMS notification channel response probability and the in-application message notification channel response probability corresponding to user-end information B and user-end information C, respectively, can be obtained.

[0068] In the third step, for the user terminal information in the candidate user terminal information set, the channel corresponding to the channel information with the highest response probability is determined as the specific push channel corresponding to the user terminal information.

[0069] In these implementations, as an example, if the response probability of the SMS notification channel corresponding to user A's information is greater than the response probability of the in-application message notification channel, the above-mentioned execution entity can determine the SMS notification channel as the specific push channel corresponding to user A's information.

[0070] It should be noted that the specific push channels corresponding to the user terminal information in the above candidate user terminal information set may be the same or different, which is not limited here.

[0071] Based on the above optional implementation methods, this solution can select information push channels based on cost constraints and the matching degree of user feature data, and perform double filtering on alternative channels through cost constraints and the matching degree with user-side information, thereby improving the information push effect while meeting the cost constraints.

[0072] Optionally, the channel information set may include telephone sales channels. The execution subject may further perform the following steps:

[0073] In the fourth step, in response to determining that the information push mode is used to indicate pushing information through a telephone sales channel, matching marketing personnel information is selected from a preset marketing personnel information list.

[0074] In these implementations, in response to determining that the information push method indicates that the information is pushed via a telephone sales channel, the execution entity may select matching marketing personnel information from a preset marketing personnel information list. The marketing personnel information list includes a marketing personnel identifier and corresponding characteristic tags. For example, the marketing personnel identifier may be a work ID or name. The characteristic tags may be, for example, areas of expertise or rating stars.

[0075] The fifth step is to send an information push task corresponding to the information push method to the terminal corresponding to the matching marketing personnel information.

[0076] In these implementations, the execution entity may send an information push task corresponding to the information push method to the terminal corresponding to the matching marketing personnel information selected in step 4. For example, the execution entity may send "Telemarketing to User A" as an information push task to the terminal used by marketing personnel with employee number "003."

[0077] Because marketing personnel have different levels of professional skills and expertise, selecting the right marketing personnel has a significant impact on the effectiveness of telemarketing channels. Based on the above optional implementation methods, this solution can further match the right marketing personnel for the telemarketing channel, thereby further improving the effectiveness of information push.

[0078] In some optional implementations of this embodiment, based on the user-related information including user behavior data and historical information push data within a preset time period, and according to the characteristic data, the execution entity may generate an information push method corresponding to the user terminal information in the candidate user terminal information set according to the following steps:

[0079] The first step is to generate push channel information based on feature data using a pre-trained channel prediction model.

[0080] In these implementations, the channel prediction model may be consistent with the above description and will not be repeated here.

[0081] The second step is to generate push time information based on the feature data using the pre-trained time prediction model.

[0082] In these implementations, as an example, the time prediction model may be a model trained using a machine learning method such as XGBOOST. As another example, the time prediction model may be a time period with the highest number of responses determined by statistically analyzing historical response time periods.

[0083] The third step is to generate push content information based on the feature data using the pre-trained content prediction model.

[0084] In these implementations, the feature data may include the user's recent (e.g., within a week) behavior categories (e.g., comments, likes, reposts, etc.), behavior duration, frequency, etc. The content prediction model may be a model trained using machine learning methods such as XGBOOST to predict user content preferences.

[0085] The fourth step is to generate the information push method based on the push channel information, push time information and push content information.

[0086] In these implementations, the information push method is used to indicate that the content indicated by the push content information generated in the second step is pushed through the push channel indicated by the push channel information generated in the first step at the time indicated by the push time information generated in the second step.

[0087] Based on the above optional implementation methods, this solution can use machine learning methods to generate information push methods that push specific content through specific channels and at specific times based on user feature data, enriching the information push solution and helping to improve the effectiveness of information push from multiple angles.

[0088] In some optional implementations of this embodiment, based on the information push mode for indicating that information is pushed at a specific push time, the above-mentioned execution subject may further perform the following steps:

[0089] The first step is to obtain user behavior information corresponding to user terminal information corresponding to the information push mode in response to determining that a specific push time has been reached.

[0090] In these implementations, the user behavior information is generally used to characterize more recent real-time behaviors, such as the past 15 minutes or the past 1 minute.

[0091] The second step is to determine whether to modify the specific push time based on the matching of the behavior information with the preset rules.

[0092] In these implementations, based on a match between the behavior information and pre-set rules, the execution entity can determine whether to adjust the specific push time. The pre-set rules may, for example, include pre-set adjustment behaviors. If the behavior indicated by the behavior information acquired in the first step matches the pre-set adjustment behavior, then the specific push time is determined to be adjusted. For example, if feature data indicates that user X frequently logs in and browses information at 7:00 PM, an information push notification is generated for 7:02 PM. If, at 7:02 PM, the execution entity detects that user X is not logged in, the push time may be adjusted, for example, by delaying it by 5 minutes.

[0093] In existing technologies, asynchronous information push systems often push information based on preset push times, making it difficult to respond to real-time events. Based on the above optional implementation method, this solution can further verify the information push method based on the user's real-time behavior when the preset information push time arrives, determining whether the push method needs to be adjusted. This can maximize the ability to capture changes in information push scenarios and adopt more appropriate methods to improve information push effectiveness.

[0094] Continue to see Figure 3 , Figure 3 This is a schematic diagram of an application scenario of a method for generating an information push mode according to an embodiment of the present disclosure. Figure 3 In the application scenario, user A, user B, user C, and user D can use terminals 3011, 3012, 3013, and 3014 respectively to interact with the backend server 302. Server 303 can be a server for analyzing the effect of information push by the backend server 302. Server 303 can be used to monitor various data, such as the number of daily active users. In response to detecting that the number of daily active users has dropped by 30% (as shown in Figure 304), server 303 can select the number of young users as a key factor from the factors associated with the above-mentioned number of daily active users (as shown in Figure 305). Then, server 303 can obtain information associated with users A, B, and D in the user group corresponding to the above-mentioned key factors (such as the young user group) from server 302 as feature data (as shown in Figure 306). Afterwards, server 303 can select user-side information from the user-side information set corresponding to the above-mentioned key factors (for example, an information set containing the terminals used by users A, B, and D) to form a candidate user-side information set (as shown in Figure 307). Finally, based on the above-mentioned characteristic data, the server 303 can generate an information push method for pushing information to the terminal 3011 and an information push method for pushing information to the terminal 3014 (as shown in FIG. 308 ).

[0095] Currently, one of the existing technologies usually requires manual determination of push strategies such as target groups, push channels, push time, and push content, which is time-consuming, inefficient, and difficult to adjust accordingly to the effect of information push. However, the method provided by the above-mentioned embodiments of the present disclosure uses target data anomalies as the generation condition for the information push method, and selects the appropriate target group to generate the information push method based on the factors causing the target data anomalies. This enriches the information push method, improves the pertinence of information push, and enhances efficiency.

[0096] Further references Figure 4 , which shows a process 400 of another embodiment of a method for generating an information push mode. The process 400 of the method for generating an information push mode includes the following steps:

[0097] Step 401 : In response to detecting that target data is abnormal, at least one factor is selected as a key factor from factors associated with the abnormal target data.

[0098] Step 402: Acquire information associated with users in the user group corresponding to the key factor as feature data.

[0099] Step 403: Select a matching information push target from a preset correspondence table based on the key factors.

[0100] In this embodiment, the execution subject of the method for generating the information push mode (eg Figure 1 The server 105 shown in the figure can select a matching information push target from a preset correspondence table in various ways. The correspondence table can be used to characterize the correspondence between factors and information push targets. The information push target is usually used to indicate that the target data is developing in a desired direction. As an example, the target data can be the number of daily active users. When the target data decreases, the information push target can be to slow down or solve the reasons for the decrease in the target data; when the target data increases, the information push target can be to maintain or help the reasons for the increase in the target data.

[0101] Step 404: input the feature data into a pre-trained response probability generation model to generate a user response probability corresponding to the feature data to a matching information push target.

[0102] In this embodiment, the response probability generation model may be a model trained by a machine learning method such as XGBOOST, for example.

[0103] Step 405 : Select user terminal information with a response probability greater than a preset response threshold from the user terminal information set corresponding to the key factors to form a candidate user terminal information set.

[0104] In this embodiment, optionally, a preset number (e.g., the first 100, the first 50) of user terminal information may be selected from the user terminal information whose response probability is greater than a preset response threshold to form a candidate user terminal information set, thereby controlling the capacity of the candidate user terminal information set.

[0105] Step 406: Generate an information push mode corresponding to the user terminal information in the candidate user terminal information set based on the feature data.

[0106] The above steps 401, 402 and 406 are respectively consistent with steps 201, 202 and 204 and their optional implementations in the aforementioned embodiment. The above descriptions of steps 201, 202, 204 and their optional implementations are also applicable to steps 401, 402 and 406, and will not be repeated here.

[0107] In some optional implementations of this embodiment, the above-mentioned execution subject may further perform the following steps:

[0108] The first step is to create a folder corresponding to the matching information push target.

[0109] In these implementations, the execution entity may also create folders corresponding to the matched information push targets. As an example, each matched information push target may correspond to a folder.

[0110] The second step is to generate an information push task based on the generated information push method.

[0111] In these implementations, the execution subject may generate an information push task in various ways according to the information push method generated in step 406. The information push task may be used to instruct information push in the above information push method.

[0112] The third step is to store the generated information push task in the corresponding folder.

[0113] In these implementations, the execution entity may store the information push task generated in the second step in a corresponding folder. As an example, the execution entity may store the information push tasks corresponding to the same information push target in a folder corresponding to the information push target.

[0114] In the fourth step, in response to determining that there is a folder matching the key factor, the information push task in the matching folder is executed.

[0115] In these implementations, after step 401 , in response to determining that there is a folder matching the key factor, the execution entity may execute the information push task in the matching folder.

[0116] Based on the above optional implementation methods, this solution can automatically generate information push tasks and execute corresponding tasks according to different information push targets to support target data. Compared with the existing manual creation of push tasks, it saves labor costs and improves efficiency.

[0117] Optionally, in response to determining that there is a folder matching the key factor, the execution subject may further execute the information push task in the matching folder according to the following steps:

[0118] S1. Divide the user-side information in the information push task in the matching folder into a push group and a control group.

[0119] In these implementations, as an example, information push task 1 in the folder of information push target 1 can be used to instruct to push information to 1000 user terminals. The execution entity can divide the 1000 user terminals into a push group (e.g., 800) and a control group (e.g., 200).

[0120] S2. Push information to the user terminal corresponding to the push group in the generated information push method.

[0121] In these implementations, the execution entity may push information to the user terminals corresponding to the push groups divided in step S1 in the generated information push manner.

[0122] S3. Generate information push evaluation information for the push group and the control group.

[0123] In these implementations, the execution entity may generate information push evaluation information for the push group and the control group after the number of information pushes reaches a certain level. The push evaluation information may include, but is not limited to, at least one of the following: push success rate, push response rate, and push efficiency.

[0124] S4. In response to determining that the information push evaluation information indicates that the push effect does not meet the requirements, execute other information push tasks under the information push target.

[0125] In these implementations, in response to determining that the information push evaluation information indicates that the push effect does not meet the requirements, the execution subject may execute other information push tasks under the information push target. For example, the execution subject may execute information push task 2 or information push task 3 in the folder belonging to information push target 1.

[0126] Based on the above optional implementation methods, this solution can select the most suitable information push method based on the effect feedback of information push, so as to improve the effect of information push.

[0127] from Figure 4 As can be seen, process 400 of the method for generating an information push method in this embodiment embodies the steps of selecting matching information push targets based on key factors, and selecting, from the user-end information set corresponding to the key factors, user-end information with a response probability greater than a preset response threshold to form a candidate user-end information set. Thus, the solution described in this embodiment can select target users based on their responses to information push targets, thereby enriching the information push method and improving the targeted nature of information push.

[0128] Further references Figure 5 As an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of a device for generating an information push method. Figure 2 or Figure 4 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0129] like Figure 5 As shown, the apparatus 500 for generating an information push mode provided by this embodiment includes a first selection unit 501, a first acquisition unit 502, a second selection unit 503, and a first generation unit 504. The first selection unit 501 is configured to, in response to detecting an abnormality in target data, select at least one factor from factors associated with the abnormal target data as a key factor, wherein the target data is used to characterize the activity level of a user group, and the factor corresponds to the category to which the users in the user group belong; the first acquisition unit 502 is configured to acquire information associated with the users in the user group corresponding to the key factor as feature data; the second selection unit 503 is configured to select user-end information from the user-end information set corresponding to the key factor to form a candidate user-end information set; and the first generation unit 504 is configured to generate, based on the feature data, an information push mode corresponding to the user-end information in the candidate user-end information set.

[0130] In this embodiment, in the apparatus 500 for generating an information push mode, the specific processing of the first selection unit 501, the first acquisition unit 502, the second selection unit 503 and the first generation unit 504 and the technical effects thereof can be referred to in the respective Figure 2 The relevant descriptions of step 201, step 202, step 203 and step 204 in the corresponding embodiment are not repeated here.

[0131] In some optional implementations of this embodiment, the above-mentioned information push method can be used to indicate that information is pushed at a specific push time. The above-mentioned device 500 for generating an information push method may further include: a second acquisition unit (not shown in the figure), configured to, in response to determining that the specific push time has been reached, acquire user behavior information corresponding to the user terminal information corresponding to the information push method; and a determination unit (not shown in the figure), configured to determine whether to modify the specific push time based on a match between the behavior information and a preset rule.

[0132] In some optional implementations of this embodiment, the above-mentioned information push method can be used to indicate that information is pushed via a specific push channel. The above-mentioned first generation unit 504 can be further configured to: generate a selectable channel information set based on a preset channel information set and a current cost constraint; determine, based on the feature data, the response probability corresponding to the user-end information in the candidate user-end information set and the channel information in the selectable channel information set; and, for the user-end information in the candidate user-end information set, determine the channel corresponding to the channel information with the highest response probability as the specific push channel corresponding to the user-end information.

[0133] In some optional implementations of this embodiment, the channel information set may include a telephone sales channel. The apparatus 500 for generating an information push method may further include: a third selection unit (not shown in the figure), configured to, in response to determining that the information push method indicates that information is pushed via the telephone sales channel, select matching marketing personnel information from a preset marketing personnel information list, wherein the marketing personnel information list includes marketing personnel identifiers and corresponding feature tags; and a sending unit (not shown in the figure), configured to send an information push task corresponding to the information push method to a terminal corresponding to the matching marketing personnel information.

[0134] In some optional implementations of this embodiment, the above-mentioned first selection unit 501 can be further configured to: obtain target data within a preset time period; in response to determining that the fluctuation of the target data exceeds a preset fluctuation threshold, determine the influence corresponding to each factor associated with the target data; and select a target number of factors whose influence meets the preset conditions as key factors.

[0135] In some optional implementations of this embodiment, the user-related information may include user behavior data and historical information push data within a preset time period. The first generating unit 504 may be further configured to: generate push channel information based on the feature data using a pre-trained channel prediction model; generate push time information based on the feature data using a pre-trained time prediction model; generate push content information based on the feature data using a pre-trained content prediction model; and generate an information push method based on the push channel information, push time information, and push content information.

[0136] In some optional implementations of this embodiment, the above-mentioned second selection unit 503 can be further configured to: select a matching information push target from a preset correspondence table based on key factors; input the feature data into a pre-trained response probability generation model to generate the response probability of the user corresponding to the feature data to the matching information push target; select user-side information with a response probability greater than a preset response threshold from the user-side information set corresponding to the key factors to form a candidate user-side information set.

[0137] In some optional implementations of this embodiment, the above-mentioned device 500 for generating an information push method may further include: an establishment unit (not shown in the figure), configured to establish a folder corresponding to the matching information push target; a second generation unit (not shown in the figure), configured to generate an information push task according to the generated information push method; a storage unit (not shown in the figure), configured to store the generated information push task to the corresponding folder; and an execution unit (not shown in the figure), configured to execute the information push task in the matching folder in response to determining that there is a folder that matches the key factor.

[0138] In some optional implementations of this embodiment, the above-mentioned execution unit can be further configured to: divide the user-end information in the information push task in the matching folder into a push group and a control group; push information to the user end corresponding to the push group in the generated information push method; generate information push evaluation information for the push group and the control group; in response to determining that the information push evaluation information indicates that the push effect does not meet the requirements, execute other information push tasks under the information push target.

[0139] The device provided by the above-mentioned embodiments of the present disclosure uses target data anomalies as a condition for generating an information push method, and selects a suitable target population to generate an information push method based on factors causing target data anomalies through the first selection unit 501 and the second selection unit 503, thereby enriching the information push method, improving the pertinence of information push, and enhancing efficiency.

[0140] Reference below Figure 6 , which shows an electronic device (eg Figure 1 The terminal devices in the embodiments of the present application may include, but are not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (such as in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 6The server shown is merely an example and should not limit the functions and scope of use of the embodiments of the present application.

[0141] like Figure 6 As shown, the electronic device 600 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603. Various programs and data required for the operation of the electronic device 600 are also stored in the RAM 603. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0142] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 6 The electronic device 600 is shown with various devices, but it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed instead. Figure 6 Each block shown in the figure may represent one device, or may represent multiple devices as needed.

[0143] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network via the communication device 609, or installed from the storage device 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of the embodiment of the present application are performed.

[0144] It should be noted that the computer-readable medium described in the embodiments of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the embodiments of the present disclosure, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device, or device. In the embodiments of the present disclosure, the computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or convey a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code embodied on the computer-readable medium may be conveyed using any suitable medium, including but not limited to wires, optical cables, RF (Radio Frequency), or any suitable combination thereof.

[0145] The computer-readable medium may be included in the electronic device, or may exist independently and not be incorporated into the server. The computer-readable medium carries one or more programs. When executed by the server, the server: in response to detecting an abnormality in target data, selects at least one factor from factors associated with the abnormal target data as a key factor, wherein the target data is used to characterize the activity level of the user group, and the factors correspond to the categories to which the users in the user group belong; obtains information associated with the users in the user group corresponding to the key factor as feature data; selects user-side information from the user-side information set corresponding to the key factor to form a candidate user-side information set; and generates, based on the feature data, an information push method corresponding to the user-side information in the candidate user-side information set.

[0146] Computer program code for performing the operations of embodiments of the present disclosure may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C", Python, or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0147] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to the various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the module, program segment, or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0148] The units involved in the embodiments described in the present disclosure may be implemented by software or by hardware. The described units may also be provided in a processor, for example, may be described as: a processor comprising a first selection unit, a first acquisition unit, a second selection unit, and a first generation unit. The names of these units do not, in some cases, constitute a limitation on the unit itself. For example, the first selection unit may also be described as "a unit that, in response to detecting an anomaly in target data, selects at least one factor as a key factor from factors associated with the abnormal target data, wherein the target data is used to characterize the activity level of the user group, and the factor corresponds to the category to which the users in the user group belong."

[0149] The above description is merely a preferred embodiment of the present disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also encompass other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the above-mentioned inventive concept. For example, a technical solution formed by mutually replacing the above-mentioned features with (but not limited to) technical features with similar functions disclosed in the embodiments of the present disclosure.

Claims

1. A method for generating an information push method, comprising: In response to detecting an abnormality in the target data, selecting at least one factor from factors associated with the abnormal target data as a key factor, wherein the target data is used to characterize the activity level of the user group, and the factor corresponds to the category to which users in the user group belong; Acquire information associated with users in the user group corresponding to the key factor as feature data, wherein the user-associated information includes user behavior data and historical information push data within a preset time period; Selecting user terminal information from the user terminal information set corresponding to the key factor to form a candidate user terminal information set; An information push mode corresponding to the user terminal information in the candidate user terminal information set is generated according to the feature data.

2. The method according to claim 1, wherein The information push mode is used to indicate that information is pushed at a specific push time; as well as The method further comprises: In response to determining that the specific push time has been reached, obtaining user behavior information corresponding to the user terminal information corresponding to the information push method; According to the matching between the behavior information and the preset rules, it is determined whether to modify the specific push time.

3. The method according to claim 1, wherein The information push mode is used to indicate that information is pushed through a specific push channel; as well as The generating, based on the feature data, an information push method corresponding to the user terminal information in the candidate user terminal information set includes: Generate an optional channel information set based on a preset channel information set and current cost constraints; determining, based on the characteristic data, response probabilities corresponding to user terminal information in the candidate user terminal information set and channel information in the optional channel information set; For the user terminal information in the candidate user terminal information set, the channel corresponding to the channel information with the highest response probability is determined as the specific push channel corresponding to the user terminal information.

4. The method according to claim 3, wherein: The channel information set includes telephone sales channels; and The method further comprises: In response to determining that the information push mode indicates pushing information via a telephone sales channel, selecting matching marketing personnel information from a preset marketing personnel information list, wherein the marketing personnel information list includes marketing personnel identifiers and corresponding feature tags; An information push task corresponding to the information push method is sent to a terminal corresponding to the matched marketing personnel information.

5. The method according to claim 1, wherein In response to detecting that the target data is abnormal, selecting at least one factor from factors associated with the abnormal target data as a key factor includes: Obtain target data within a preset time period; In response to determining that the fluctuation of the target data exceeds a preset fluctuation threshold, determining an influence degree corresponding to each factor associated with the target data; A target number of factors whose influences meet preset conditions are selected as the key factors.

6. The method according to claim 1, wherein generating, based on the feature data, an information push method corresponding to the user terminal information in the candidate user terminal information set comprises: Generate push channel information using a pre-trained channel prediction model based on the feature data; Generate push time information using a pre-trained time prediction model based on the feature data; Generate push content information using a pre-trained content prediction model based on the feature data; An information push method is generated according to the push channel information, push time information and push content information.

7. The method according to any one of claims 1 to 6, wherein: The selecting user terminal information from the user terminal information set corresponding to the key factor to form a candidate user terminal information set includes: According to the key factors, a matching information push target is selected from a preset correspondence table; Inputting the feature data into a pre-trained response probability generation model to generate a response probability of the user corresponding to the feature data to the matched information push target; The user terminal information having a response probability greater than a preset response threshold is selected from the user terminal information set corresponding to the key factor to form the candidate user terminal information set.

8. The method according to claim 7, wherein: The method further comprises: Creating a folder corresponding to the matching information push target; Generate an information push task according to the generated information push method; Store the generated information push task in the corresponding folder; In response to determining that there is a folder matching the key factor, an information push task in the matching folder is executed.

9. The method according to claim 8, wherein The executing the task of pushing information in the matching folder includes: Dividing the user-side information in the information push task in the matching folder into a push group and a control group; Pushing information to the user terminals corresponding to the push group in the generated information push mode; Generate information push evaluation information for the push group and the control group; In response to determining that the information push evaluation information indicates that the push effect does not meet the requirements, other information push tasks under the information push target are executed.

10. A device for generating an information push method, comprising: a first selection unit configured to, in response to detecting an abnormality in target data, select at least one factor from factors associated with the abnormal target data as a key factor, wherein the target data is used to characterize the activity level of a user group, and the factor corresponds to a category to which users in the user group belong; A first acquisition unit is configured to acquire information associated with users in the user group corresponding to the key factor as feature data, wherein the user-associated information includes user behavior data and historical information push data within a preset time period; A second selection unit is configured to select user terminal information from the user terminal information set corresponding to the key factor to form a candidate user terminal information set; The first generating unit is configured to generate, based on the feature data, an information push mode corresponding to the user terminal information in the candidate user terminal information set.

11. A server comprising: one or more processors; a storage device having one or more programs stored thereon; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 9.

12. A computer-readable medium having a computer program stored thereon, wherein: When the program is executed by a processor, the method according to any one of claims 1 to 9 is implemented.

Images