Message pushing method and system based on user corresponding behavior relation network

Abstract

The invention provides a message pushing method and system based on a user corresponding behavior relation network, computer equipment, a computer readable storage medium and a computer program product, and relates to the technical field of data processing, and the method comprises the steps: obtaining original behavior data of an original user; updating a behavior relationship network of the original user according to the original behavior data; in response to a triggering operation of a target event, identifying a target user in the original users; obtaining a predicted trigger factor through the behavior relationship network corresponding to the target user; updating the target message and / or the message pushing strategy based on the predicted trigger factor; wherein the target message is optimized according to the common interception factor; optimizing the message pushing strategy according to the personalized interception factor; and executing the latest message pushing strategy, and pushing the target message to the target user, thereby improving the efficiency and pertinence of message pushing.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a message push method, system, computer equipment, computer-readable storage medium and computer program product based on a user corresponding behavior relationship network. Background Art

[0002] In the existing message push system, in order to improve user experience and marketing effects, major companies usually adopt multiple message push methods. However, under the existing technical system, these push methods often operate independently and lack effective integration and coordination, resulting in poor push effects.

[0003] Therefore, a message push method, system, computer device, computer readable storage medium and computer program product based on a user corresponding behavior relationship network. Summary of the invention

[0004] The present specification provides a message push method, system, computer device, computer-readable storage medium and computer program product based on a user-corresponding behavior relationship network, which achieves accurate capture of user behavior patterns by dynamically acquiring and updating the original user's behavior relationship network; when a target event is triggered, the target user can be quickly identified, and the trigger factor can be predicted using its behavior relationship network to improve the pertinence and effectiveness of message push, thereby maximizing the effect of message push.

[0005] The present application provides a message push method based on a user corresponding behavior relationship network, comprising:

[0006] Obtain the original behavior data of the original user;

[0007] Updating the behavior relationship network of the original user according to the original behavior data;

[0008] In response to a triggering operation of a target event, identifying a target user among the original users;

[0009] Obtaining a predicted trigger factor through the behavior relationship network corresponding to the target user; the predicted trigger factor includes: at least one of a common interception factor and an individual interception factor;

[0010] Based on the predicted triggering factor, the target message and / or the message push strategy is updated; wherein the target message is optimized according to the common interception factor; and the message push strategy is optimized according to the individual interception factor;

[0011] Execute the latest message push strategy to push the target message to the target user.

[0012] Optionally, the original behavior data includes business behavior data;

[0013] Optionally, obtaining the original behavior data of the original user includes:

[0014] According to preset retrieval conditions, business operation information is retrieved in batches from the offline table to construct the business behavior data of the original user.

[0015] Optionally, the original behavior data includes push behavior data;

[0016] Optionally, obtaining the original behavior data of the original user includes:

[0017] Get the message push strategy for each original user;

[0018] Pushing the target message to the original user based on the message push strategy;

[0019] Obtain the current reach effect of each of the original users regarding the message push strategy, and construct the push behavior data.

[0020] Optionally, obtaining the predicted trigger factor through the behavior relationship network corresponding to the target user includes:

[0021] The behavior relationship network corresponding to the target user is used as the target behavior relationship network;

[0022] Monitor each target behavior relationship network in real time to determine the degree of association between the region and the target event;

[0023] It is determined whether there is a common interception factor based on the degree of association.

[0024] Optionally, the individual interception factor includes: a channel factor and a time factor;

[0025] The step of obtaining the predicted triggering factor through the behavior relationship network corresponding to the target user further includes:

[0026] According to the channel label in the behavior relationship network corresponding to the target user and the current push channel, determining whether the trigger factor includes the channel factor;

[0027] According to the creation date and the current push time in the behavior relationship network corresponding to the target user, it is determined whether the trigger factor includes the time factor.

[0028] Optionally, also include:

[0029] An identity association is established for the original user; and a message push strategy is uniformly adjusted for the original user with identity association.

[0030] The present application provides a message push system based on a user corresponding behavior relationship network, including:

[0031] An acquisition module is used to obtain the original behavior data of the original user;

[0032] A network updating module, used for updating the behavior relationship network of the original user according to the original behavior data;

[0033] An identification module, configured to identify a target user among the original users in response to a triggering operation of a target event;

[0034] A prediction module, used to obtain a predicted trigger factor through a behavior relationship network corresponding to the target user; the predicted trigger factor includes: at least one of a common interception factor and an individual interception factor;

[0035] A push update module, used to update the target message and / or message push strategy based on the predicted trigger factor;

[0036] The execution module is used to execute the latest message push strategy and push the target message to the target user.

[0037] Optionally, the push update module includes:

[0038] A first updating submodule, configured to optimize the target message according to the common interception factor;

[0039] A second updating submodule, used for optimizing the message push strategy according to the individual interception factor;

[0040] Optionally, the original behavior data includes business behavior data;

[0041] Optionally, the acquisition module includes:

[0042] The first acquisition submodule is used to retrieve business operation information in batches from the offline table according to preset retrieval conditions to construct the business behavior data of the original user.

[0043] Optionally, the original behavior data includes push behavior data;

[0044] Optionally, the acquisition module further includes: a second acquisition submodule;

[0045] The second acquisition submodule includes:

[0046] A first acquisition unit, used to acquire a message push policy for each original user;

[0047] A first pushing unit, configured to push the target message to the original user based on the message pushing strategy;

[0048] The first construction unit is used to obtain the current reach effect of each of the original users regarding the message push strategy and construct the push behavior data.

[0049] Optionally, the prediction module includes: a first identification submodule;

[0050] The first identification submodule includes:

[0051] a marking unit, configured to use the behavior relationship network corresponding to the target user as a target behavior relationship network;

[0052] A monitoring unit, used for monitoring each target behavior relationship network in real time to determine the degree of association between a region and the target event;

[0053] A judging unit is used to judge whether there is a common interception factor based on the correlation degree.

[0054] Optionally, the individual interception factor includes: a channel factor and a time factor;

[0055] Optionally, the prediction module further includes:

[0056] A second identification submodule is used to determine whether the trigger factor includes the channel factor according to the channel label in the behavior relationship network corresponding to the target user and the current push channel;

[0057] The third identification submodule is used to determine whether the trigger factor includes the time factor according to the creation date and the current push time in the behavior relationship network corresponding to the target user.

[0058] Optionally, also include:

[0059] The adjustment module is used to establish identity association with the original user; and to uniformly adjust the message push strategy for the original user with identity association.

[0060] This specification also provides a computer device, wherein the computer device includes:

[0061] processor; and,

[0062] A memory storing computer executable instructions, which when executed cause the processor to perform any of the above methods.

[0063] This specification also provides a computer-readable storage medium, wherein the computer-readable storage medium stores one or more programs / instructions, and when the one or more programs / instructions are executed by a processor, any of the above methods is implemented.

[0064] This specification also provides a computer program product, wherein the computer program product includes: a computer program / instructions, and when the computer program / instructions are executed by a processor, any of the above methods is implemented.

[0065] In the present invention, the original behavior data of the original user is obtained; the behavior relationship network of the original user is updated according to the original behavior data, thereby achieving accurate capture of the user behavior pattern;

[0066] In response to the triggering operation of the target event, the target user among the original users is identified; through the behavioral relationship network corresponding to the target user, a predicted trigger factor is obtained; the predicted trigger factor includes: at least one of a common interception factor and an individual interception factor; the pertinence and effectiveness of message push are significantly improved; based on the predicted trigger factor, the target message and / or message push strategy is personalized; wherein, the target message is optimized according to the common interception factor; the message push strategy is optimized according to the individual interception factor; the latest message push strategy is executed to push the target message to the target user. Through the above method, not only the interference of invalid information is reduced, but also the user experience is enhanced, and the message reception rate and conversion rate are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 A schematic diagram of the principle of a message push method based on a user corresponding behavior relationship network provided in an embodiment of this specification;

[0068] Figure 2 A schematic diagram of a process for obtaining business behavior data of a message push method based on a user corresponding behavior relationship network provided in an embodiment of this specification;

[0069] Figure 3 A schematic diagram of a process for acquiring push behavior data of a message push method based on a user corresponding behavior relationship network provided in an embodiment of this specification;

[0070] Figure 4 A schematic diagram of a behavior relationship network of a message push method based on a user corresponding behavior relationship network provided in an embodiment of this specification Figure 1 ;

[0071] Figure 5 A schematic diagram of a behavior relationship network of a message push method based on a user corresponding behavior relationship network provided in an embodiment of this specification Figure 2 ;

[0072] Figure 6 A schematic diagram of the storage process of graph data of a message push method based on a user corresponding behavior relationship network provided in an embodiment of this specification;

[0073] Figure 7 A schematic diagram of step S7 of a message push method based on a user corresponding behavior relationship network provided in an embodiment of this specification;

[0074] Figure 8 A schematic diagram of the structure of a message push system based on a user corresponding behavior relationship network provided in an embodiment of this specification;

[0075] Fig. 9 A schematic diagram of the structure of a computer device provided in an embodiment of this specification;

[0076] Fig.10 A schematic diagram of a computer-readable storage medium provided in accordance with an embodiment of the present specification. DETAILED DESCRIPTION

[0077] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art can think of other obvious variations. The basic principles of the present invention defined in the following description can be applied to other embodiments, variations, improvements, equivalents, and other technical solutions that do not deviate from the spirit and scope of the present invention.

[0078] The exemplary embodiments of the present invention will now be described more fully with reference to the accompanying drawings. Under the premise of being consistent with the technical concept of the present invention, the features, structures, characteristics or other details described in a particular embodiment do not exclude that they can be combined in one or more other embodiments in a suitable manner.

[0079] The term "and / or" or "and / or" includes all combinations of any one or more of the associated listed items.

[0080] Figure 1 A schematic diagram of a message push method based on a user corresponding behavior relationship network provided in an embodiment of this specification includes:

[0081] S1 obtains the original behavior data of the original user;

[0082] S2 updates the behavior relationship network of the original user according to the original behavior data;

[0083] S3, in response to a triggering operation of a target event, identifying a target user among the original users;

[0084] S4 obtains a predicted trigger factor through the behavior relationship network corresponding to the target user; the predicted trigger factor includes: at least one of a common interception factor and an individual interception factor;

[0085] S5: updating the target message and / or message push strategy based on the predicted trigger factor; wherein the target message is optimized according to the common interception factor; and the message push strategy is optimized according to the individual interception factor;

[0086] S6 executes the latest message push strategy to push the target message to the target user.

[0087] In the existing message push system, in order to improve user experience and marketing effect, major companies usually push messages through multiple push channels (such as SMS, WeChat, Douyin, Alipay, iMessage, IVR outbound calls, etc.). In actual operation, there are mainly the following problems:

[0088] (1) Message push on multiple push channels often operates independently. Therefore, the push effect of message push based on a single push channel is difficult to deeply understand the real needs and preferences of users, resulting in low message push efficiency.

[0089] (2) Due to the lack of real-time monitoring of the push effects of each push channel, if there is a problem during the push process, it may not be possible to respond quickly and adjust the strategy. This delay may reduce the optimization effect of the push strategy, thereby affecting the stability and arrival rate of the message.

[0090] (3) For users to whom push fails, a re-sending mechanism is generally used. However, independent re-sending on multiple push channels may waste resources. If the content or method of push is not targeted or attractive, then even if it is re-sent, it may not significantly improve the reach of the push message, and thus it will not be possible to effectively mine valuable users. Moreover, frequent, non-targeted, repetitive pushes may even cause users to be disgusted or blocked, thereby reducing the efficiency of mining valuable users.

[0091] Based on this, in order to accurately and efficiently push target messages, the present invention provides a message push method based on a user corresponding behavior relationship network, which specifically includes:

[0092] S1 obtains the original behavior data of the original user;

[0093] The original user refers to the recipient when the target message is pushed.

[0094] Original behavior data refers to the behavior data related to the original user. Specifically, original behavior data includes: business behavior data and push behavior data.

[0095] S11 obtains business behavior data on each push channel;

[0096] Business behavior data is used to characterize the business operations performed by the original user.

[0097] S111 builds an offline table;

[0098] An offline record file is constructed in advance for each push channel; the offline record file includes a plurality of offline tables; and the offline tables correspond one to one with the types of business operations.

[0099] That is to say, each business operation type of each push channel corresponds to an offline table.

[0100] Push channels include but are not limited to: terminals, business applications (apps), business messaging platforms supported by third-party applications (for example, WeChat official accounts), and business mini-programs embedded in third-party applications (for example, Douyin mini-programs, Alipay mini-programs, and WeChat mini-programs).

[0101] S112 generates business operation information corresponding to the type of the business operation based on the business operation of the original user;

[0102] In a specific application scenario, the types of business operations include but are not limited to: login, credit, and payment. Business operation information includes but is not limited to: user login information, user credit information, and user payment information.

[0103] Specifically, when the type of business operation is login, the corresponding business operation information is user login information; the user login information is used to record the login behavior of the original user, and the user login information includes but is not limited to: login time.

[0104] When the type of business operation is credit granting, the corresponding business operation information is user credit granting information; user credit granting information is used to record the credit granting behavior of the original user, and the user credit granting information includes but is not limited to: credit granting time.

[0105] When the type of business operation is spending, the corresponding business operation information is user spending information; user spending information is used to record the original user's consumption behavior using the obtained credit line, and the user spending information includes but is not limited to: the usage time of the credit line.

[0106] S113 matches the business operation information with a corresponding offline table based on the push channel of the business operation and the type of the business operation;

[0107] Find the offline record file corresponding to the push channel where the business operation is located; based on the type of business operation, find the corresponding offline table from the corresponding offline record file;

[0108] S114 records the business operation information in the corresponding offline table.

[0109] S115 retrieves the business operation information in batches from the offline table according to the preset retrieval conditions, and constructs the business behavior data of the original user.

[0110] like Figure 2 As shown, Kalka is used to retrieve business operation information from the offline table of the offline record file, extract the original user's business behavior data, and send it to gns-manage.

[0111] S12 obtains the push behavior data of the original user on each push channel;

[0112] Push behavior data is used to characterize the relevant results of pushing messages to original users, such as the push status of messages, the reach of messages, etc.

[0113] S121 obtains the message push strategy for each original user;

[0114] Message push strategies include: target push channel and target push time.

[0115] S122 pushes the target message to the original user based on the message push strategy;

[0116] According to the target push time, the target message is pushed to the original user in the target push channel.

[0117] S123 obtains the current reach effect of each of the original users regarding the message push strategy and constructs the push behavior data.

[0118] In one embodiment of the present specification, the reach effect is monitored according to a preset monitoring time to obtain the current reach effect; wherein the preset monitoring time is adjusted according to actual conditions.

[0119] In another embodiment of the present specification, the reach effect is monitored in real time, and the final reach effect within the preset monitoring time is obtained as the current reach effect.

[0120] Combined with the actual push status of the target message and the current reach effect, push behavior data is determined. Push behavior data includes but is not limited to: current push channel, current push time, and current push result.

[0121] In one embodiment of this specification, Figure 3As shown, based on the message middleware, the message is pushed and the push result is obtained to obtain the push behavior data. Preferably, the message middleware is RocketMQ.

[0122] In a specific application scenario, Alipay messages and Douyin messages are encapsulated through gns-other and pushed to the corresponding message queues through RocketMQ's synchronous sending interface; the background systems of Alipay and / or Douyin perform real-time processing and push Alipay messages and / or Douyin messages to the original users; after the processing is completed, the push results and reach results are returned to gns-other through RocketMQ's reply mechanism or another message queue, and the push results and reach results are synchronized.

[0123] In a specific application scenario, gns-app pushes the SMS content to the message queue of the SMS service provider through RocketMQ's synchronous sending interface; the SMS service provider pushes the SMS, and after successful sending, returns the push result to gns-app through RocketMQ's reply mechanism or another message queue to synchronize the push result.

[0124] In a specific application scenario, according to the user login device information, push messages are pushed to the corresponding user devices through RocketMQ's synchronous sending interface. The user terminal (or the corresponding push service) receives and processes these push messages. After processing, the push results (such as whether they are successfully delivered, whether the user clicks, etc.) are synchronously returned to gns-push through RocketMQ's reply mechanism or another message queue.

[0125] In a specific application scenario, gns-web pushes SMS and / or push messages to the corresponding message queue through RocketMQ's asynchronous sending interface; the recipient (such as SMS service provider and push service) processes the message asynchronously and returns the reach result to gns-web through the callback interface after processing; the reach result is received and processed using an asynchronous callback method.

[0126] In a specific application scenario, the message of the WeChat public account or mini program is pushed to the corresponding message queue through the asynchronous sending interface of RocketMQ; after the message is received and processed, the asynchronous callback push and reach result are performed.

[0127] Get push behavior data based on the message middleware and send it to gns-manage.

[0128] S2 updates the behavior relationship network of the original user according to the original behavior data;

[0129] S21 searches for a behavior relationship network corresponding to the original user;

[0130] S211 retrieves the user information of the original user;

[0131] User information includes but is not limited to: the original user's username (user_no) and unique identification code. The unique identification code includes but is not limited to: the encryption result of the original user's mobile phone number (mobile_no_encryptx) and the encrypted summary of the original user's mobile phone number (mobile_no_md5x).

[0132] S212 determines whether there is a behavior relationship network associated with the user information;

[0133] The behavioral relationship network is associated with user information one by one.

[0134] In one embodiment of the present specification, the user information of each original user corresponds to a behavior relationship network, and the behavior relationship network may include a basic node and several characteristic nodes. The basic node is the center of the behavior relationship network and is connected to each characteristic node.

[0135] In one embodiment of the present specification, the behavior relationship network is graph data, and the behavior relationship network is stored in a Nebula graph database.

[0136] S213: if there is a behavior relationship network associated with the user information, directly acquiring the behavior relationship network associated with the user information;

[0137] S214: If there is no behavior relationship network associated with the user information, initialize the behavior relationship network for the original user, create a basic node based on the user information, build a behavior relationship network centered on the basic node, and establish an association between the user information and the behavior relationship network.

[0138] S22 updates the behavior relationship network of the original user according to the original behavior data;

[0139] Adding a new feature node in the behavior relationship network of the original user based on the original behavior data of the original user;

[0140] S221 adds a behavior feature node in the behavior relationship network of the original user based on the business behavior data, and associates information for the behavior feature node;

[0141] S221-1 When the original behavior data is business behavior data, construct a behavior feature node;

[0142] S221 - 2 adds the behavior feature node to the behavior relationship network.

[0143] S221 - 3 matches a channel label for the behavior feature node based on the current push channel.

[0144] S221 - 4 extracts the behavior feature information corresponding to the behavior feature node based on the business behavior data.

[0145] The behavior characteristic information is used to represent the relevant information of the corresponding push channel.

[0146] Behavior characteristic information includes: first basic information. Figure 4 As shown, when the original user operates on the terminal, the first basic information includes: identification id (reg_id), device number (device_no), mobile phone system (mobile_system), and creation date (date_created).

[0147] When the original user operates on the business application (app), the first basic information includes: the name of the business application (login_app), the login channel (login_channel), and the creation date (date_created).

[0148] When the original user operates on the messaging platform, the first basic information includes: official account code (official_accounts_code), official account name (official_accounts_name), creation date (date_created). For example, when the original user operates on a WeChat public account, the first basic information includes: WeChat public account code (wechat_official_accounts_code), WeChat public account name (wechat_official_accounts_name), creation date (date_created).

[0149] When the original user operates on the mini program, the first basic information includes: the third-party application code (applet_code), the third-party application name (applet_name), and the creation date (date_created). For example, when the original user operates on the Douyin mini program, the first basic information includes: the code of the Douyin mini program (douyin_applet_code), the name of the Douyin mini program (douyin_applet_name), and the creation date (date_created). When the original user operates on the Alipay mini program, the first basic information includes: the code of the Alipay mini program (alipay_applet_code), the name of the Alipay mini program (alipay_applet_name), and the creation date (date_created). When the original user operates on the WeChat mini program, the first basic information includes: the code of the WeChat mini program (wechat_applet_code), the name of the WeChat mini program (wechat_applet_name), and the creation date (date_created).

[0150] When the original user operates on the computer program product (product), the first basic information includes: product code (product_code), business name (biz_name), and creation date (date_created).

[0151] The behavioral characteristic information also includes: additional subscription information.

[0152] The additional subscription information includes one or more of a subscription state (subscribe_state) and a creation date of the subscription.

[0153] In one embodiment of the present specification, when the current push channel is a terminal or the current push channel is a business application, the additional subscription information includes: a creation date.

[0154] When the current push channel is a business message platform supported by a third-party application, or the current push channel is a business applet embedded in a third-party application, the additional subscription information includes the subscription state (subscribe_state) and the creation date of the subscription.

[0155] S222 adds a push feature node in the behavior relationship network of the original user based on the push behavior data, and associates information for the push feature node;

[0156] S222-1 When the original behavior data is push behavior data, construct a push feature node;

[0157] S222-2 adds the push feature node to the behavior relationship network.

[0158] Establish a connection between the push feature node and the basic node.

[0159] S222-3, based on the reaching result of the push behavior data, setting a result feedback label for the push feature node;

[0160] The types of result feedback labels include a first type and a second type.

[0161] Determine whether the reach results of the push behavior data meet the preset reach conditions;

[0162] If the reaching result meets the preset reaching condition, the result feedback label of the pushed feature node is the first type; if the reaching result does not meet the preset reaching condition, the result feedback label of the pushed feature node is the second type.

[0163] The preset reaching condition is used to indicate that the message has successfully reached the original user. The preset reaching condition may refer to that the original user has read the target message, or may refer to that the original user has performed relevant operations on the target message, etc. The specific preset reaching condition may be adjusted according to the actual situation.

[0164] In one embodiment of the present specification, the preset reaching condition is that the target message is in the read state. It is easy to understand that when the target message is in the read state, the corresponding result feedback tag is the first type. When the target message is not in the read state, the corresponding result feedback tag is the second type. Among them, the target message is not in the read state, including but not limited to: the target message is not successfully delivered (the target message is intercepted) and the target message is not read.

[0165] The result feedback label can be a text label, a color label, etc., and there is no limitation here.

[0166] In one embodiment of the present specification, Figure 5 The figure shows a behavioral relationship network of an original user, including several nodes (circles) and edge relationships (directed lines). The middle node is a basic node (circle without color filling), which includes the user name (userNo) and the unique identification code (mobileNo); the surrounding of the basic node is the feature node (circle filled with color), and the feature node has an association relationship with the basic node.

[0167] Each feature node has a corresponding channel label, result feedback label, and creation time. Specifically, the creation time is displayed on the edge relationship between the feature node and the basic node; the channel label is the specific content displayed in the circle of the feature node, such as push messages and WeChat public account messages; the result feedback label is the color of the circle of the feature node, where the circle indicated by green fill corresponds to the first type; the circle indicated by red fill corresponds to the second type.

[0168] S222-4 extracting push feature information corresponding to the push feature node from the push behavior data;

[0169] Push feature information is used to represent relevant information when a message is pushed to an original user through a specific push channel.

[0170] In one embodiment of the present specification, the push characteristic information includes: second basic information and additional channel information.

[0171] The second basic information includes: template code (template_code), message type (msg_type), business purpose (purpose), and creation time (date_created).

[0172] Additional channel information includes: business number (biz_no), reach result (result), report (report), name, creation time (date_created), update time (date_updated), message identifier (msg_id), report description (report_desc), channel (channel), purpose (purpose), third-party number (third_no), location data (for example, province), service provider (sp, Service Provider), signature (signature), withdrawal number (draw_no), status (state), withdrawal amount (draw_amt), and application number (appl_no).

[0173] like Figure 4 As shown, when an iMessage message is pushed to an iOS terminal, the additional channel information includes: business number (biz_no), reach result (result), report (report), creation time (date_created), and update time (date_updated).

[0174] When pushing SMS messages to Android terminals and / or iOS terminals, the additional channel information includes: business number (biz_no), third-party number (third_no), IP address (for example, province), reach result (result), report (report), report description (report_desc), service provider (sp, Service Provider), channel (channel), purpose (purpose), signature (signature), creation time (date_created), and update time (date_updated).

[0175] When a message (push message) is pushed through a business application, the additional channel information includes: business number (biz_no), message identifier (msg_id), reach result (result), report (report), report description (report_desc), channel (channel), purpose (purpose), business application name (app_name), creation time (date_created), and update time (date_updated).

[0176] When messages are pushed through a messaging platform (for example, WeChat official account), additional channel information includes: business number (biz_no), reach result (result), report (report), official account name (official_accounts_name), creation time (date_created), and update time (date_updated).

[0177] When a message (for example, Douyin message / Alipay message / WeChat mini program message) is pushed through a business mini program (for example, Douyin mini program, Alipay mini program, WeChat mini program), the additional channel information includes: business number (biz_no), reach result (result), report (report), name of the business mini program (applet_name), creation time (date_created), and update time (date_updated).

[0178] When a message is pushed through a computer program product (product), in one embodiment of the present specification, the additional channel information includes: draw number (draw_no), state (state), draw amount (draw_amt), creation time (date_created). In another embodiment of the present specification, the additional channel information includes: application number (appl_no), state (state), creation time (date_created).

[0179] In one embodiment of this specification, Figure 6 As shown in the figure, graph data is inserted or updated in nebula through gns-manage.

[0180] The present invention uses graph technology to display and analyze user message push behavior relationships, and by constructing a behavior relationship network, it intuitively presents the overall behavior of original users on multiple push channels. By accurately displaying and analyzing the original users' message push behaviors on different push channels, it understands user preferences and needs.

[0181] S3, in response to a triggering operation of a target event, identifying a target user among the original users;

[0182] Target events include: generating a new target object.

[0183] Based on the triggering of the target event, it is necessary to determine the triggering cause in order to adjust the message push strategy and re-push the message.

[0184] The target object includes: a second type of result feedback tag. In one embodiment of the present specification, the result feedback tag corresponding to the push feature node is monitored in real time; when a new second type of result feedback tag appears, a target event is triggered; in response to the triggering operation of the target event, the corresponding target user is determined based on the behavior relationship network where the second type of result feedback tag is located.

[0185] S4 obtains a predicted trigger factor through a behavior relationship network corresponding to the target user;

[0186] Trigger factors include: common interception factors and individual interception factors. Individual interception factors include but are not limited to: channel factors and time factors.

[0187] When the target event is triggered, it indicates that there is at least one interception factor. Therefore, the predicted triggering factors include: at least one of a common interception factor and an individual interception factor.

[0188] The present invention utilizes graph technology to mine target users whose push fails in each push channel, intelligently analyzes the specific behaviors of target users, deeply analyzes the reasons for push failure, and accurately locates trigger factors.

[0189] S41 monitors and identifies common interception factors based on all the target behavior relationship networks;

[0190] S411: taking the behavior relationship network corresponding to the target user as the target behavior relationship network;

[0191] S412 monitors each target behavior relationship network in real time to determine the degree of association between the region and the target event;

[0192] In one embodiment of the present specification, based on the location data of all target users, the interception ratio of each region within a preset time period is calculated as the correlation degree between the region and the target object.

[0193] The scope of the region can be defined according to actual needs and can be divided into provinces, cities, etc., and there is no restriction here.

[0194] Interception ratio = number of target objects in the same region / number of target objects in all regions.

[0195] S413 determines whether there is a common interception factor based on the correlation degree.

[0196] Determining whether the correlation degree is greater than a preset monitoring threshold;

[0197] If the degree of association is less than or equal to the preset association threshold, it is determined that there is no common interception factor;

[0198] If the degree of correlation is greater than the preset monitoring threshold, it is determined that a common interception factor exists.

[0199] That is, there is a large-scale failure in message push in the same area. Based on this, it can be inferred that the corresponding area automatically intercepted the target message.

[0200] S42, judging whether the trigger factor includes the channel factor according to the channel label in the behavior relationship network corresponding to the target user and the current push channel;

[0201] S421, determining active channels according to channel labels in the target behavior relationship network;

[0202] S421-1 traverses the target behavior relationship network of the target user to find the channel labels corresponding to all the behavior feature nodes;

[0203] S421-2 determines the channel activity of the target user in each push channel based on the number of each channel tag;

[0204] The channel activity is proportional to the number of corresponding channel tags. In one embodiment of this specification, the channel activity is the number of corresponding channel tags;

[0205] In another embodiment of the present specification, the channel activity is the ratio of the number of corresponding channel tags.

[0206] S421 - 3 sets an active weight for each push channel in combination with the behavior feature information of the behavior feature node.

[0207] The most recent access date of each push channel is determined in combination with the push channel basic information and / or additional subscription information of the behavior feature node; and an active weight is set for each push channel based on the most recent access date of each push channel.

[0208] S421-4 determining the active channels of the target user according to the channel activity and the active weight;

[0209] Based on the product of the channel activity and the activity weight of the push channel, an activity index of the push channel is obtained; and the active channel is identified in combination with the activity index.

[0210] In one embodiment of the present specification, the push channels are arranged in descending order according to the activity index; and the push channels with preset rankings are regarded as active channels.

[0211] In another embodiment of the present specification, it is determined whether the channel activity of the push channel is greater than a preset activity threshold, and if so, the corresponding push channel is regarded as an active channel.

[0212] S422 identifies a current push channel based on a target object in the target event;

[0213] According to the newly generated second type of result feedback tag, locate the push feature node corresponding to the second type of result feedback tag; search for the channel tag of the corresponding push feature node to determine the current push channel;

[0214] S423, combining the active channel and the current push channel, determining whether the trigger factor includes a channel factor;

[0215] Determine whether the active channel is consistent with the current push channel;

[0216] If the active channel is consistent with the current push channel, it is determined that the trigger factor does not include the channel factor; if the active channel is inconsistent with the current push channel, it is determined that the trigger factor includes the channel factor.

[0217] S43 determines whether the trigger factor includes the time factor according to the creation date and the current push time in the behavior relationship network corresponding to the target user.

[0218] S431 predicts active time according to the creation date in the target behavior relationship network;

[0219] S431-1 traverses the target behavior relationship network of the target user, finds the behavior feature node corresponding to the active channel; finds the creation date corresponding to the behavior feature node;

[0220] S431 - 2 determines an active time period based on the corresponding creation date as the active time.

[0221] S432 searches for a message push policy corresponding to the target user and searches for a current push time;

[0222] S433 determines whether the trigger factor includes a time factor based on the overlap between the current push time and the active time;

[0223] Find the overlapping period between the current push time and the active time;

[0224] Based on the proportion of overlapping periods in the current push time, the degree of overlap is obtained;

[0225] Based on the degree of overlap and the preset overlap threshold, determine whether the trigger factor includes a time factor; when the degree of overlap is greater than the overlap threshold, it is determined that the trigger factor does not include a time factor; if the degree of overlap is less than or equal to the overlap threshold, it is determined that the trigger factor includes a time factor.

[0226] S5: updating the target message and / or message push strategy based on the predicted trigger factor;

[0227] The message push strategy includes: target push method and target push time.

[0228] S51 optimizes the target message according to the common interception factor;

[0229] When a common interception factor exists, the content of the target message is optimized; and the optimized target message is obtained.

[0230] The present invention monitors the push status of relevant dimensions such as various regions in real time to facilitate timely discovery of the problem of target message being intercepted; by dynamically adjusting the target message, the timely arrival of the message is ensured, thereby greatly improving the stability and reliability of the channel and improving user satisfaction.

[0231] S52 optimizes the message push strategy according to the personalized interception factor;

[0232] S521 adjusts the target push mode of the message push strategy according to the active channel;

[0233] When a channel factor exists, the push method in the message push strategy is replaced with an active channel;

[0234] S522: modifying the target push mode of the message push strategy according to the active time;

[0235] When a time factor exists, based on the active time, the target push time in the message push policy is modified to the active time.

[0236] Based on the individual interception factor, the present invention can identify potential users with potential value, such as users who only use a single push method and have poor results but are actively logged in. For potential users, the message push strategy is adjusted based on actual behavior data to activate these potential users, improve the push effect, and improve the overall reach and marketing effect of push messages.

[0237] The present invention recommends the most suitable push channel and sending time for each user by mining the best matching relationship between target users and different push channels, so as to adjust the push strategy, ensure that the message can reach the user at the best time, effectively reduce the situation of push failure, and improve the user reach rate and conversion rate.

[0238] In order to improve the reach of the message, it also includes:

[0239] S523 temporarily adjusts the message push strategy;

[0240] Specifically, when a channel factor exists, the most recent access time of the target user in the active channel is searched; if the most recent access time is within a preset access threshold, a replacement push is immediately performed based on the active channel.

[0241] In one embodiment of the present specification, when the current push channel is the terminal and the active channel is the business application, if the target user cannot receive the target message push due to abnormal mobile phone signal, based on the target user's normal network signal and the target user has just logged into the business application, push is used to perform a substitute message push to achieve the purpose of reaching the user.

[0242] S6 executes the latest message push strategy to push the target message to the target user.

[0243] In order to further optimize the information reach rate of original users, it also includes:

[0244] S7: establishing identity association for the original user; and uniformly adjusting the message push strategy for the original user with identity association.

[0245] S71 searches for users with the same terminal device based on the behavior feature nodes in the behavior relationship network;

[0246] Collect the original behavior data of the original user and extract the terminal related information, which includes: terminal device information, user login information, and behavior log. Perform preprocessing operations such as cleaning, deduplication, and standardization on the terminal related information. Based on the preprocessed terminal related information, use a graph database or graph computing framework to build a user-device relationship graph; in which each node represents a user or device, each edge represents the association or interaction between the user and the device, and the edge relationship represents a transaction record or identification record. Use a graph algorithm to search for original users with the same terminal device in the user-device relationship graph.

[0247] In one embodiment of this specification, Figure 7 As shown, its nodes include: original users (such as u1, u2), devices (such as t1, t2) and business applications; its edge relationships are used to represent business records (trade_record) or identification records (msg_record).

[0248] S72 clusters original users with the same terminal device to build a user group with the same terminal;

[0249] S73 unifies each message push strategy in the same terminal user group.

[0250] That is, when pushing messages, according to a unified message push strategy, messages are pushed to all original users in the same terminal user group at the same time, so as to achieve the goal of pushing messages to different original users on the same terminal at the same time.

[0251] In one embodiment of the present specification, the above graph data statistics can be triggered based on a scheduled task, and the statistical results can be saved in the Nebula graph database and the mysql statistical table. Specifically, the user-device relationship graph is stored in the Nebula graph database; the same terminal user group is saved in the mysql statistical table.

[0252] Based on the fact that the same terminal may correspond to multiple users, the present invention mines the user group of the same terminal through graph technology, and realizes simultaneous push to the original users of the same terminal to avoid repeated operations, thereby further improving the push efficiency of users of the same terminal and the overall reach rate of push messages.

[0253] In order to improve the efficiency of information processing, the behavior relationship network in the Nebula graph database is traversed to find the creation date of each feature node; when the creation date does not meet the preset saving conditions, the feature node and the corresponding feature information are deleted. The preset saving conditions include: the creation date is within 30 days of the current time interval.

[0254] In one embodiment of the present specification, the conversion effect of each important short message can be displayed based on the graph data for funnel monitoring, so as to provide data reference for subsequent effect improvement.

[0255] Through intelligent analysis and optimization of push strategies, the present invention actually reduces invalid or repeated message delivery, which indirectly reduces the load on the server, improves the throughput and response speed of the server, saves valuable resources, optimizes resource utilization and improves system efficiency.

[0256] Figure 8 A message push system based on a user corresponding behavior relationship network provided in an embodiment of this specification includes:

[0257] An acquisition module 810 is used to acquire original behavior data of an original user;

[0258] A network updating module 820, configured to update the original user's behavior relationship network according to the original behavior data;

[0259] An identification module 830, configured to identify a target user among the original users in response to a triggering operation of a target event;

[0260] The prediction module 840 is used to obtain a predicted trigger factor through the behavior relationship network corresponding to the target user; the predicted trigger factor includes: at least one of a common interception factor and an individual interception factor;

[0261] A push update module 850, configured to update the target message and / or message push strategy based on the predicted trigger factor;

[0262] The execution module 860 is used to execute the latest message push strategy and push the target message to the target user.

[0263] Optionally, the push update module 850 includes:

[0264] A first updating submodule, configured to optimize the target message according to the common interception factor;

[0265] A second updating submodule, used for optimizing the message push strategy according to the individual interception factor;

[0266] Optionally, the original behavior data includes business behavior data;

[0267] Optionally, the acquisition module 810 includes:

[0268] The first acquisition submodule is used to retrieve business operation information in batches from the offline table according to preset retrieval conditions to construct the business behavior data of the original user.

[0269] Optionally, the original behavior data includes push behavior data;

[0270] Optionally, the acquisition module 810 further includes: a second acquisition submodule;

[0271] The second acquisition submodule includes:

[0272] A first acquisition unit, used to acquire a message push policy for each original user;

[0273] A first pushing unit, configured to push the target message to the original user based on the message pushing strategy;

[0274] The first construction unit is used to obtain the current reach effect of each of the original users regarding the message push strategy and construct the push behavior data.

[0275] Optionally, the prediction module 840 includes: a first identification submodule;

[0276] The first identification submodule includes:

[0277] a marking unit, configured to use the behavior relationship network corresponding to the target user as a target behavior relationship network;

[0278] A monitoring unit, used for monitoring each target behavior relationship network in real time to determine the degree of association between a region and the target event;

[0279] A judging unit is used to judge whether there is a common interception factor based on the correlation degree.

[0280] Optionally, the individual interception factor includes: a channel factor and a time factor;

[0281] Optionally, the prediction module 840 further includes:

[0282] A second identification submodule is used to determine whether the trigger factor includes the channel factor according to the channel label in the behavior relationship network corresponding to the target user and the current push channel;

[0283] The third identification submodule is used to determine whether the trigger factor includes the time factor according to the creation date and the current push time in the behavior relationship network corresponding to the target user.

[0284] Optionally, also include:

[0285] The adjustment module is used to establish identity association with the original user; and to uniformly adjust the message push strategy for the original user with identity association.

[0286] The functions of the system of the embodiment of the present invention have been described in the above method embodiment, so for details not fully described in this embodiment, please refer to the relevant description in the above embodiment, and no further description will be given here.

[0287] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0288] The present invention is described with reference to flowcharts and / or block diagrams of methods, systems (devices), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer programs / instructions. These computer programs / instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that instructions executed by the processor of the computer device or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0289] These computer programs / instructions may also be stored in a readable memory of a computer device capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the readable memory of the computer device produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0290] These computer programs / instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer device or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer device or other programmable device provide for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0291] If the technical solution of this application involves personal information, the product using the technical solution of this application has clearly informed the personal information processing rules and obtained the individual's voluntary consent before processing the personal information. If the technical solution of this application involves sensitive personal information, the product using the technical solution of this application has obtained the individual's separate consent before processing the sensitive personal information, and at the same time meets the "explicit consent" requirement. For example, on personal information collection devices such as cameras, clear and prominent signs are set to inform that the personal information collection scope has been entered and personal information will be collected. If the individual voluntarily enters the collection scope, it is deemed that he or she agrees to the collection of his or her personal information; or on the device that processes personal information, the personal information processing rules are notified by obvious signs / information, and the individual's authorization is obtained through pop-up information or by asking the individual to upload his or her personal information; among them, the personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the type of personal information processed.

[0292] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A message push method based on a user corresponding behavior relationship network, characterized in that: include: Obtain the original behavior data of the original user; Updating the behavior relationship network of the original user according to the original behavior data; In response to a triggering operation of a target event, identifying a target user among the original users; Obtaining a predicted trigger factor through a behavior relationship network corresponding to the target user; The predicted triggering factor includes: at least one of a common interception factor and an individual interception factor; Based on the predicted triggering factor, the target message and / or the message push strategy is updated; wherein the target message is optimized according to the common interception factor; and the message push strategy is optimized according to the individual interception factor; Execute the latest message push strategy to push the target message to the target user.

2. A message push method based on a user corresponding behavior relationship network as claimed in claim 1, characterized in that: The original behavior data includes business behavior data; The obtaining of the original behavior data of the original user includes: According to preset retrieval conditions, business operation information is retrieved in batches from the offline table to construct the business behavior data of the original user.

3. A message push method based on a user corresponding behavior relationship network as claimed in claim 1, characterized in that: The original behavior data includes push behavior data; The obtaining of the original behavior data of the original user includes: Get the message push strategy for each original user; Pushing the target message to the original user based on the message push strategy; Obtain the current reach effect of each of the original users regarding the message push strategy, and construct the push behavior data.

4. A message push method based on a user corresponding behavior relationship network as claimed in claim 1, characterized in that: The step of obtaining the predicted triggering factor through the behavior relationship network corresponding to the target user includes: The behavior relationship network corresponding to the target user is used as the target behavior relationship network; Monitor each target behavior relationship network in real time to determine the degree of association between the region and the target event; It is determined whether there is a common interception factor based on the degree of association.

5. A message push method based on a user corresponding behavior relationship network as claimed in claim 1, characterized in that: The individual interception factors include: channel factors and time factors; The step of obtaining the predicted triggering factor through the behavior relationship network corresponding to the target user further includes: According to the channel label in the behavior relationship network corresponding to the target user and the current push channel, determining whether the trigger factor includes the channel factor; According to the creation date and the current push time in the behavior relationship network corresponding to the target user, it is determined whether the trigger factor includes the time factor.

6. A message push method based on a user corresponding behavior relationship network as claimed in claim 1, characterized in that: Also includes: Establishing an identity association for the original user; For original users with identity associations, unified adjustments are made to the message push strategy.

7. A message push system based on a user corresponding behavior relationship network, characterized in that: include: An acquisition module is used to obtain the original behavior data of the original user; A network updating module, used for updating the behavior relationship network of the original user according to the original behavior data; An identification module, configured to identify a target user among the original users in response to a triggering operation of a target event; A prediction module, used to obtain a predicted trigger factor through a behavior relationship network corresponding to the target user; The predicted triggering factor includes: at least one of a common interception factor and an individual interception factor; A push update module, used to update the target message and / or message push strategy based on the predicted trigger factor; wherein the target message is optimized according to the common interception factor; and the message push strategy is optimized according to the individual interception factor; The execution module is used to execute the latest message push strategy and push the target message to the target user.

8. A computer device, characterized in that: The computer equipment includes: processor; and, A memory storing computer executable instructions, which when executed cause the processor to perform the method of any one of claims 1-6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores one or more programs / instructions, and when the one or more programs / instructions are executed by a processor, the method according to any one of claims 1 to 6 is implemented.

10. A computer program product, characterized in that The method comprises a computer program / instruction, which, when executed by a processor, implements the method according to any one of claims 1 to 6.

Images