An information push method, apparatus, device, and medium

Abstract

This invention discloses an information push method, apparatus, device, and medium. The method includes: acquiring behavioral data input by a user to be processed; responding to a first operation performed by the user to be processed, and calculating the association probability between the first operation and a second operation based on the behavioral data; determining recommended information based on the association probability and descriptive information of the second operation, and sending it to the user to be processed, so that the user to be processed triggers the execution of the second operation. Embodiments of this invention can improve the accuracy and efficiency of information push and reduce the workload in the information push process.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to an information push method, apparatus, device, and medium. Background Technology

[0002] With the development of information technology, information push has become an important part of network information technology.

[0003] In the process of information push, the target audience and content of the information are usually determined manually. Because this manual method requires experience from staff, the information obtained is often inaccurate and the workload is substantial. Summary of the Invention

[0004] This invention provides an information push method, apparatus, device, and medium to improve the accuracy and efficiency of information push and reduce the workload in the information push process.

[0005] According to one aspect of the present invention, an information push method is provided, the method comprising:

[0006] Acquire behavioral data from user input to be processed;

[0007] In response to a first operation performed by the user to be processed, and based on the operation data, calculate the association probability between the first operation and the second operation;

[0008] Based on the association probability and the description information of the second operation behavior, recommendation information is determined and sent to the user to be processed, so that the user to be processed triggers the execution of the second operation behavior.

[0009] According to another aspect of the present invention, an information push device is provided, the device comprising:

[0010] The behavior data acquisition module is used to acquire the behavior data of the user input to be processed;

[0011] The association probability calculation module is used to respond to the first operation behavior performed by the user to be processed, and to calculate the association probability between the first operation behavior and the second operation behavior based on the behavior data.

[0012] The information sending module is used to determine recommendation information based on the association probability and the description information of the second operation behavior and send it to the user to be processed, so that the user to be processed can trigger the execution of the second operation behavior.

[0013] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0014] At least one processor; and

[0015] A memory communicatively connected to the at least one processor; wherein,

[0016] The memory stores a computer program that can be executed by the at least one processor, which is then executed by the at least one processor to enable the at least one processor to perform the information push method according to any embodiment of the present invention.

[0017] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions, the computer instructions being configured to cause a processor to execute and implement the information push method described in any embodiment of the present invention.

[0018] According to another aspect of the present invention, a computer program product is provided, the computer program product comprising a computer program that, when executed by a processor, implements the information push method described in any embodiment of the present invention.

[0019] The technical solution of this invention acquires behavioral data input by a user to be processed. When the user performs a first operation, the association probability between the first and second operations is calculated based on the behavioral data. Recommended information is determined and sent to the user based on the association probability and the description information of the second operation, triggering the user to execute a second operation. This achieves automatic calculation of the association probability between the first and second operations, reducing the workload of manual calculation and improving the efficiency of association probability calculation, thereby improving the efficiency of sending recommended information. Furthermore, by determining and sending recommended information based on the association probability and the description information of the second operation, the system can send corresponding recommended information based on the description information of the second operation, improving the accuracy of recommended information delivery.

[0020] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart of an information push method provided according to Embodiment 1 of the present invention;

[0023] Figure 2 This is a flowchart of an information push method provided according to Embodiment 2 of the present invention;

[0024] Figure 3 This is a schematic diagram of the structure of an information push device according to Embodiment 3 of the present invention;

[0025] Figure 4 This is a schematic diagram of the structure of an electronic device that implements the information push method of this invention. Detailed Implementation

[0026] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0027] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the term "comprising" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0028] Example 1

[0029] Figure 1 This is a flowchart illustrating an information push method according to Embodiment 1 of the present invention. This embodiment is applicable to situations where information is pushed to users. The method can be executed by an information push device, which can be implemented in hardware and / or software and can be configured in an electronic device. Figure 1 As shown, the method includes:

[0030] S110. Obtain the behavioral data of the user input to be processed.

[0031] "Pending users" refers to users who require information push notifications. Specifically, pending users can be first-time users or users who have previously used the product. The product can be at least one of an application or a website. Behavioral data refers to information associated with a user during product use. Pending user input behavioral data refers to data input by the user, meaning data provided to the electronic device with the user's authorization or consent. The acquisition, storage, use, and processing of behavioral data comply with relevant national laws and regulations. For example, pending user input behavioral data may include at least one of the following: the user's gender, age, name of the product used, number of times the product was used, and data describing operational behavior. Operational behavior refers to the actions performed by the user during product use. For example, operational behavior may include at least one of launching the product, logging into the product, placing an order on the product's page, searching on the product's page, and browsing on the product's page. Data describing operational behavior may include the time of execution, the number of times the operation was performed, and the result of the operation.

[0032] Specifically, the user inputs behavioral data into the product during the product's use, and the product receives the user input behavioral data.

[0033] S120. In response to the first operation performed by the user to be processed, calculate the association probability between the first operation and the second operation based on the behavior data.

[0034] The first action refers to the user's action before the recommendation information is sent. The first action can be determined by the user's input behavior data. The second action refers to the user's action after the recommendation information is sent. The first and second actions can be the same or different. The recommendation information refers to the information displayed to the user, which can be at least one of text, voice, image, and video information. The content of the recommendation information can be a description of a product that the user is interested in or instructions on how to use the product. The association probability between the first and second actions refers to the probability that the user will perform the second action after performing the first action.

[0035] Specifically, based on the behavioral data input by the user to be processed, the first operation to be performed by the user to be processed is determined. In response to the first operation to be performed by the user to be processed, the correlation probability between the first operation and the second operation is calculated based on the acquired behavioral data.

[0036] S130. Based on the association probability and the description information of the second operation behavior, determine the recommendation information and send it to the user to be processed, so that the user to be processed triggers the execution of the second operation behavior.

[0037] The description information is used to instruct the user to perform an action. It may include at least one of the following: the name of the action, the product corresponding to the action, and the steps to perform the action. The description information for the second action is used to instruct the user to perform the second action.

[0038] Specifically, based on the magnitude of the association probability, a user has a certain probability of triggering the second action. The descriptive information of the second action is combined to determine recommended information, which is then sent to the user awaiting processing. Upon receiving the recommended information, the user can trigger the second action. The recommended information can be sent to the user via at least one of the following methods: SMS, voice, or application push notifications. For example, an association probability of 5% indicates that after performing the first action, the user has a 5% chance of performing the second action. The descriptive information of the second action includes: the action name (browsing on the product's page), the corresponding product (application A), and the action execution steps (clicking link XXX). The recommended information determined based on the descriptive information of the second action is "Please click link XXX to browse promotional information on the page of application A." The recommended information is sent to the user awaiting processing via SMS. After receiving the recommended information, the user clicks the link "XXX" in the recommended information to enter the page for application A, thus triggering the second action.

[0039] The technical solution of this invention acquires behavioral data input by a user to be processed. When the user performs a first operation, the association probability between the first and second operations is calculated based on the behavioral data. Recommended information is determined and sent to the user based on the association probability and the description information of the second operation, triggering the user to execute a second operation. This achieves automatic calculation of the association probability between the first and second operations, reducing the workload of manual calculation and improving the efficiency of association probability calculation, thereby improving the efficiency of sending recommended information. Furthermore, by determining and sending recommended information based on the association probability and the description information of the second operation, the system can send corresponding recommended information based on the description information of the second operation, improving the accuracy of recommended information delivery.

[0040] Based on the above embodiments, the step of determining recommendation information and sending it to the user to be processed according to the association probability and the description information of the second operation behavior, so that the user to be processed triggers the execution of the second operation behavior, includes: when the association probability meets the recommendation information sending condition, determining recommendation information to send to the user to be processed according to the description information of the second operation behavior, so that the user to be processed triggers the execution of the second operation behavior.

[0041] The conditions for sending recommendation information refer to the conditions under which recommendation information is allowed to be sent. For example, the condition for sending recommendation information is that the correlation probability is greater than 5%. That is, it is allowed to send recommendation information to users whose correlation probability is greater than 5%; it is prohibited to send recommendation information to users whose correlation probability is less than or equal to 5%.

[0042] Specifically, based on the conditions for sending recommendation letters, it is determined whether the association probability meets the conditions for sending recommendation information. When the association probability meets the conditions for sending recommendation information, the recommendation information is determined according to the description information of the second operation behavior, and the recommendation information is sent to the user to be processed, so that the user to be processed triggers the execution of the second operation behavior. When the association probability does not meet the conditions for sending recommendation information, the sending of recommendation information is prohibited.

[0043] By determining whether the association probability meets the conditions for sending recommendation information, recommendation information is determined and sent to the users to be processed when the association probability meets the conditions. When the association probability does not meet the conditions, recommendation information is prohibited from being sent. This eliminates the need to send recommendation information to each user to be processed, saving the cost of sending recommendation information and reducing the waste of resources used in the process.

[0044] Based on the above embodiments, the behavioral data of the user input to be processed includes user attribute information and user operation data.

[0045] User attribute information refers to information used to distinguish different users. For example, user attribute information may include attributes such as the user's gender and age, provided with the user's authorization or consent. User operation data refers to data generated by a user performing actions, that is, data describing those actions. User operation data may be at least one of the following: the name of the product used, the number of times the product was used, and data describing the action. For example, user operation data could be the number of times the user used the product (3 times) or the user placing an order on the product's page. Both user attribute information and user operation data are data actively entered by the user with their consent.

[0046] By analyzing the behavioral data input by users, different users can be distinguished, and user operation data of each user can be obtained, providing a data analysis basis for determining recommendation information.

[0047] Example 2

[0048] Figure 2 This is a flowchart of an information push method provided in Embodiment 2 of the present invention. Based on the above embodiments, this embodiment further specifies the step of responding to a first operation performed by the user to be processed and calculating the association probability between the first operation and the second operation based on the behavior data. Specifically, it involves responding to a first operation performed by the user to be processed, and calculating the association probability between the first operation and the second operation based on the behavior data using a machine learning model. Figure 2 As shown, the method includes:

[0049] S210. Obtain the behavioral data of the user input to be processed.

[0050] S220. In response to the first operation performed by the user to be processed, the association probability between the first operation and the second operation is calculated using a machine learning model based on the behavior data.

[0051] A machine learning model is an expression of an algorithm that searches for patterns or makes predictions by processing massive amounts of data. In other words, by training a machine learning model on a large amount of data, it acquires the necessary computational capabilities after training. Specifically, machine learning models can use gradient boosting algorithms such as XGBoost or LightGBM, or artificial neural network algorithms, etc., without limitation in this application. In this application, the machine learning model mainly refers to a pre-trained machine learning model used to calculate the association probability between a first operation and a second operation. Specifically, behavioral data can be input into the machine learning model, which will automatically calculate and output the association probability between the first and second operations.

[0052] Specifically, in response to the first operation performed by the user to be processed, the behavioral data input by the user to be processed is input into the machine learning model to obtain the correlation probability between the first operation and the second operation output by the machine learning model.

[0053] S230. Based on the association probability and the description information of the second operation behavior, determine the recommendation information and send it to the user to be processed, so that the user to be processed triggers the execution of the second operation behavior.

[0054] The technical solution of this invention calculates the correlation probability between the first operation and the second operation through a machine learning model, which can reduce manual workload and improve the calculation efficiency of correlation probability.

[0055] Based on the above embodiments, the method further includes: acquiring behavioral data input by multiple target users as training data for the machine learning model; behavioral data input by one target user as sample data in the training data; responding to a first operation performed by the target user, and calculating the sample association probability of the first operation and the second operation for the target user based on the sample data; statistically analyzing each sample association probability to determine the recommendation information sending conditions; determining recommendation information based on the recommendation information sending conditions and the description data of the second operation, and sending it to users who meet the recommendation information sending conditions, so that users who meet the recommendation information sending conditions trigger the execution of the second operation; counting the number of users who meet the recommendation information sending conditions who trigger the execution of the second operation to obtain sample statistical results; and determining that the machine learning model training is complete when the sample statistical results meet the preset quantity judgment condition.

[0056] Specifically, target users refer to users who can send recommendation information, or, more broadly, a subset of users to be processed. There are at least two target users. Sample data refers to the data required for a single calculation by the machine learning model. In this application, one target user corresponds to one behavioral data point, and one behavioral data point corresponds to one sample data point. Sample association probability refers to the association probability between the first and second operational actions calculated after inputting the sample data corresponding to the target user into the machine learning model; it can also be understood as the probability that the target user will execute the second operational action after performing the first operational action. Recommendation information sending conditions refer to conditions determined after statistically analyzing the association probabilities of each sample, representing the lower limit of the probability that the target user will trigger the execution of the second operational action after receiving the recommendation information, thereby improving the rationality of resource allocation for sending recommendation information. There are at least one recommendation information sending condition. For example, all sample association probabilities can be sorted in descending order, and the sample association probability corresponding to the target user at the nth position in the sort can be taken as the lower limit, with the recommendation information sending condition being that the sample association probability is greater than the lower limit. Sample statistical results refer to the number of users corresponding to the sample association probabilities that trigger the execution of the second operational action. The preset quantity judgment condition refers to the condition used to determine whether the machine learning model has completed training, based on the sample statistical results. For example, the preset quantity judgment condition could be that the sample statistical result is greater than 8000. The preset quantity judgment condition can be set according to the actual situation.

[0057] Specifically, after receiving a recommendation message, a user who meets the conditions for receiving the recommendation message may or may not trigger a second action. When a user who meets the conditions for receiving the recommendation message triggers the second action, they input new behavioral data to form new behavioral data. The number of users who meet the conditions for receiving the recommendation message and whose behavioral data after receiving the recommendation message includes the second action is counted to obtain the sample statistical results.

[0058] Based on the above embodiments, the target users are filtered according to the behavioral data input by each target user to obtain control users; the number of control users is the same as the number of users corresponding to the target sample association probability; the recommendation information is sent to each control user to trigger the execution of the second operation behavior; the number of control users who trigger the execution of the second operation behavior is counted to obtain the control statistical result; if the sample statistical result is greater than the control statistical result, it is determined that the machine learning model training is complete.

[0059] Specifically, control users refer to users retained after filtering target users using a control method. This control method differs from the method used to obtain users who meet the conditions for sending recommendation information. Control users form a control group with the users who meet the conditions for sending recommendation information. This can be understood as follows: for the target users, two different methods are used to filter and obtain two groups of equal numbers of users. Users who meet the conditions for sending recommendation information form the first user group, and control users form the second user group. For both user groups, the same recommendation information is sent in the same way. The number of users in the first user group who triggered the second action after receiving the recommendation information is compared with the number of users in the second user group who triggered the second action after receiving the same recommendation information. The comparison result is used to determine whether the machine learning model is superior to the control method. If the machine learning model is superior to the control method, the machine learning model is considered to have completed training; if the machine learning model is not superior to the control method, the machine learning model is considered to have not completed training. For example, if the number of users in the second user group who triggered the second action after receiving the same recommendation information is greater than or equal to the number of users in the first user group who triggered the second action after receiving the recommendation information, the machine learning model is determined to be inferior to the control method; if the number of users in the second user group who triggered the second action after receiving the same recommendation information is less than the number of users in the first user group who triggered the second action after receiving the recommendation information, the machine learning model is determined to be superior to the control method. The control users and the users who meet the recommendation information sending conditions may or may not overlap. Based on the behavioral data input by each target user, target users are filtered according to the control method. This can be based on user operation data within the target user's input behavioral data. For example, target users can be filtered based on the time of product login in the user operation data, retaining target users whose product login time was in May. The control statistical result refers to the number of control users who triggered the second action after receiving the recommendation information.

[0060] Specifically, based on the behavioral data input by each target user, target users are filtered according to preset screening conditions using a comparison method. Target users meeting the preset screening conditions are retained and designated as control users. The number of control users is the same as the number of users meeting the conditions for sending recommendation information. Recommendation information intended for users meeting the conditions is then sent to each control user. After receiving the recommendation information, each control user can choose to trigger a second action or not. When a control user triggers the second action, new behavioral data is generated. The number of new behavioral data inputs is counted, and the number of control users whose behavioral data after receiving the recommendation information includes the second action is calculated, yielding the comparison statistics. If the sample statistics result is greater than the comparison statistics result, the machine learning model is determined to be superior to the comparison method, indicating that the machine learning model training is complete. This demonstrates that the machine learning model can increase the proportion of users triggering the second action when receiving recommendation information, thereby improving the efficiency and rationality of recommendation information delivery. If the sample statistics result is greater than, less than, or equal to, the comparison statistics result, the machine learning model is determined to be inferior to the comparison method, indicating that the machine learning model has failed to increase the proportion of users triggering the second action when receiving recommendation information, and further training of the machine learning model is required.

[0061] By selecting control users from the target users and sending the same recommendation information to them, the control users are prompted to trigger a second action. The number of control users who trigger the second action is counted to obtain the control statistical results. If the sample statistical results are greater than the control statistical results, the machine learning model is considered to have completed training. By forming a control group with users whose association probability with the target sample, the comparison between the sample statistical results and the control statistical results is used to determine whether the machine learning model has completed training. This can improve the accuracy of the association probability calculated by the machine learning model, thereby improving the accuracy and rationality of the recommendation information.

[0062] Based on the above embodiments, the step of determining the effectiveness of the machine learning model when the sample statistical results meet the preset quantity judgment conditions further includes: calculating the proportion of the sample statistical results in the number of users who meet the recommendation information sending conditions; and determining that the machine learning model training is complete when the proportion meets the preset proportion judgment conditions.

[0063] The proportion of the sample statistics among the number of users meeting the recommendation information sending conditions refers to the result obtained by dividing the sample statistics by the number of users meeting the recommendation information sending conditions. A higher proportion indicates that more users among those meeting the recommendation information sending conditions trigger the second action, suggesting higher accuracy in sending the recommendation information. The preset proportion judgment condition refers to the condition used to determine the completion of the machine learning model training. The preset proportion judgment condition can be determined based on actual circumstances. For example, the preset proportion judgment condition can be that the percentage deviation between the proportion and the average association probability of users meeting the recommendation information sending conditions is less than 5%, meaning the proportion needs to satisfy: proportion * (1-5%) < proportion < proportion * (1+5%).

[0064] Specifically, the sample statistics are divided by the number of users who meet the conditions for receiving recommendation information to obtain the proportion of the sample statistics among the total number of users who meet the conditions for receiving recommendation information. Based on this proportion and preset proportion judgment conditions, if the proportion meets the preset proportion judgment conditions, it is determined that the association probability calculation of the machine learning model is accurate and the machine learning model training is complete. If the proportion does not meet the preset proportion judgment conditions, it is determined that the association probability calculation of the machine learning model is not accurate enough and the machine learning model needs to be further trained. For example, the average association probability of users who meet the conditions for sending recommendation information is 4%. The preset weight judgment condition is that the percentage deviation between the weight and the average association probability of users who meet the conditions for sending recommendation information is less than 5%. If the percentage deviation between the weight and the average association probability of users who meet the conditions for sending recommendation information is less than 5%, that is, the weight is greater than 4%*(1-5%) = 3.8% and less than 4%*(1+5%) = 4.2%, it indicates that the probability of the user triggering the second operation after sending recommendation information is basically consistent with the association probability calculated by the machine learning model, which can ensure the accuracy of the association probability calculation and determine that the association probability calculation of the machine learning model is accurate. If the percentage deviation between the weight and the average association probability of users who meet the conditions for sending recommendation information is greater than or equal to 5%, that is, the weight is less than or equal to 3.8% or greater than or equal to 4.2%, it indicates that the probability of the user triggering the second operation after sending recommendation information deviates significantly from the association probability calculated by the machine learning model, and it is determined that the association probability calculation of the machine learning model is not accurate enough, and the machine learning model needs to be further trained.

[0065] By calculating the proportion of the sample statistical results among the number of users who meet the conditions for sending recommendation information, and determining that the machine learning model training is complete when the proportion meets the preset proportion judgment conditions, this method can reduce the amount of computation and improve the efficiency of determining the completion of machine learning model training compared to the method of determining that the machine learning model training is complete when the sample statistical results meet the preset quantity judgment conditions.

[0066] Optionally, by comparing the sample statistical results and the control statistical results, if the sample statistical results are greater than the control statistical results, it is determined that the machine learning model is superior to the control method; by calculating the proportion of the sample statistical results in the number of users who meet the conditions for sending recommendation information, if the proportion meets the preset proportion judgment conditions, it is determined that the machine learning model's association probability calculation is accurate; when the machine learning model is superior to the control method and the association probability calculation is accurate, it is determined that the machine learning model training is complete.

[0067] By combining two criteria—whether the machine learning model is superior to the control method and whether the association probability calculation is accurate—the effectiveness of the machine learning model can be determined, thereby improving the accuracy of the determination results.

[0068] Example 3

[0069] Figure 3 This is a schematic diagram of an information push device provided in Embodiment 3 of the present invention. Figure 3 As shown, the device includes: a behavior data acquisition module 301, an association probability calculation module 302, and an information sending module 303.

[0070] Among them, the behavior data acquisition module 301 is used to acquire the behavior data input by the user to be processed;

[0071] The association probability calculation module 302 is used to respond to the first operation behavior performed by the user to be processed, and to calculate the association probability between the first operation behavior and the second operation behavior based on the behavior data;

[0072] The information sending module 303 is used to determine recommendation information based on the association probability and the description information of the second operation behavior and send it to the user to be processed, so that the user to be processed can trigger the execution of the second operation behavior.

[0073] The technical solution of this invention acquires behavioral data input by a user to be processed. When the user performs a first operation, the association probability between the first and second operations is calculated based on the behavioral data. Recommended information is determined and sent to the user based on the association probability and the description information of the second operation, triggering the user to execute a second operation. This achieves automatic calculation of the association probability between the first and second operations, reducing the workload of manual calculation and improving the efficiency of association probability calculation, thereby improving the efficiency of sending recommended information. Furthermore, by determining and sending recommended information based on the association probability and the description information of the second operation, the system can send corresponding recommended information based on the description information of the second operation, improving the accuracy of recommended information delivery.

[0074] Optionally, the correlation probability calculation module 302 is specifically used for:

[0075] In response to a first operation performed by the user to be processed, the association probability between the first operation and the second operation is calculated using a machine learning model based on the behavior data.

[0076] Optionally, the device may also include:

[0077] The training data acquisition module is used to acquire behavioral data input by multiple target users as training data for the machine learning model; behavioral data input by one target user is a sample data in the training data.

[0078] The sample association probability calculation module is used to respond to the first operation performed by the target user, and calculate the sample association probability of the first operation and the second operation for the target user based on the sample data.

[0079] The sending condition determination module is used to statistically analyze the association probability of each sample and determine the sending conditions for the recommendation information;

[0080] The sample statistics module is used to count the number of users who meet the conditions for sending recommendation information and trigger the execution of the second operation behavior, and to obtain sample statistics results;

[0081] The quantity judgment module is used to determine that the machine learning model training is complete when the sample statistical results meet the preset quantity judgment conditions.

[0082] Optional quantity determination module, including:

[0083] The reference user determination unit is used to filter the target users based on the behavioral data input by each target user to obtain reference users; the number of reference users is the same as the number of users corresponding to the association probability of the target sample;

[0084] The comparison recommendation information sending unit is used to send the recommendation information to each comparison user, so that the comparison user triggers the execution of the second operation behavior;

[0085] The comparison statistics unit is used to count the number of comparison users who triggered the execution of the second operation behavior, and to obtain the comparison statistics results;

[0086] The comparison judgment unit is used to determine that the machine learning model training is complete when the sample statistical result is greater than the comparison statistical result.

[0087] Optional quantity determination module, including:

[0088] The proportion calculation module is used to calculate the proportion of the sample statistical results in the number of users who meet the conditions for sending recommendation information;

[0089] The proportion judgment module is used to determine that the machine learning model training is complete when the proportion meets the preset proportion judgment condition.

[0090] Optionally, the information sending module 303 is specifically used for:

[0091] When the association probability meets the conditions for sending recommendation information, recommendation information is sent to the user to be processed based on the description information of the second operation behavior, so that the user to be processed triggers the execution of the second operation behavior.

[0092] Optionally, the behavioral data of the user input to be processed includes user attribute information and user operation data.

[0093] The information push device provided in the embodiments of the present invention can execute the information push method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method execution.

[0094] Example 4

[0095] Figure 4 A schematic diagram of an electronic device 10 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0096] like Figure 4 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0097] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0098] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as information push methods.

[0099] In some embodiments, the information push method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the information push method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the information push method by any other suitable means (e.g., by means of firmware).

[0100] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0101] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0102] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0103] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0104] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0105] A computing system can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system. It addresses the shortcomings of traditional physical hosts and VPS (Virtual Private Server) services, such as high management difficulty and weak business scalability.

[0106] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0107] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. An information push method characterized by comprising: include: Acquire behavioral data from user input to be processed; In response to a first operation performed by the user to be processed, and based on the operation data, calculate the correlation probability between the first operation and the second operation. Based on the association probability and the description information of the second operation behavior, recommendation information is determined and sent to the user to be processed, so that the user to be processed triggers the execution of the second operation behavior; Wherein, the association probability between the first operation and the second operation refers to the probability that the user to be processed will perform the second operation after performing the first operation. The method further includes: Collect behavioral data from multiple target users as training data for the machine learning model; behavioral data from one target user is used as a sample data in the training data. In response to a first operation performed by the target user, and based on the sample data, calculate the sample association probability of the first operation and the second operation for the target user; Statistical analysis is performed on the association probabilities of each sample to determine the conditions for sending recommendation information; Based on the recommendation information sending conditions and the description data of the second operation behavior, recommendation information is determined and sent to users who meet the recommendation information sending conditions, so that users who meet the recommendation information sending conditions trigger the execution of the second operation behavior; The number of users who meet the conditions for sending recommendation information and trigger the second operation is counted to obtain sample statistics results; When the sample statistical results meet the preset quantity judgment conditions, the machine learning model is determined to have completed training; Wherein, determining that the machine learning model training is complete when the sample statistical results meet the preset quantity judgment condition includes: Based on the behavioral data input by each target user, the target users are filtered to obtain control users; the number of control users is the same as the number of users corresponding to the association probability of the target sample. The recommendation information is sent to each of the control users, so that the control users trigger the execution of the second operation. The number of control users who triggered the second operation was counted, and the control statistics results were obtained. If the sample statistical result is greater than the control statistical result, the machine learning model is determined to have completed training. The step of determining recommendation information and sending it to the user to be processed based on the association probability and the description information of the second operation behavior, so that the user to be processed triggers the execution of the second operation behavior, includes: When the association probability meets the recommendation information sending condition, the recommendation information is sent to the user to be processed according to the description information of the second operation behavior, so that the user to be processed triggers the execution of the second operation behavior; The method further includes: prohibiting the sending of recommendation information when the association probability does not meet the conditions for sending recommendation information.

2. The method of claim 1, wherein, The step of responding to a first operation performed by the user to be processed, and calculating the association probability between the first operation and the second operation based on the operation data, includes: In response to a first operation performed by the user to be processed, the association probability between the first operation and the second operation is calculated using a machine learning model based on the behavior data.

3. The method of claim 1, wherein, The step of determining that the machine learning model training is complete when the sample statistical results meet the preset quantity judgment condition includes: Calculate the proportion of the sample statistics in the number of users who meet the conditions for receiving recommendation information; When the proportion meets the preset proportion judgment condition, the training of the machine learning model is determined to be complete.

4. The method according to claim 1, characterized in that, The behavioral data input by the user to be processed includes user attribute information and user operation data.

5. An information push device, characterized in that, include: The behavior data acquisition module is used to acquire the behavior data of the user input to be processed; The association probability calculation module is used to respond to the first operation behavior performed by the user to be processed, and to calculate the association probability of the first operation behavior and the second operation behavior based on the behavior data. The information sending module is used to determine recommendation information based on the association probability and the description information of the second operation behavior and send it to the user to be processed, so that the user to be processed triggers the execution of the second operation behavior; Wherein, the association probability between the first operation and the second operation refers to the probability that the user to be processed will perform the second operation after performing the first operation. The device further includes: The training data acquisition module is used to acquire behavioral data input from multiple target users as training data for the machine learning model; behavioral data input from one target user is a sample data in the training data. The sample association probability calculation module is used to respond to the first operation performed by the target user, and calculate the sample association probability of the first operation and the second operation for the target user based on the sample data. The sending condition determination module is used to statistically analyze the association probability of each sample and determine the sending conditions for the recommendation information; The sample statistics module is used to count the number of users who meet the conditions for sending recommendation information and trigger the second operation behavior, and obtain the sample statistics results; The quantity judgment module is used to determine that the machine learning model training is complete when the sample statistical results meet the preset quantity judgment conditions; The quantity determination module includes: The reference user determination unit is used to filter the target users based on the behavioral data input by each target user to obtain reference users; the number of reference users is the same as the number of users corresponding to the association probability of the target sample. The comparison recommendation information sending unit is used to send the recommendation information to each comparison user, so that the comparison user triggers the execution of the second operation behavior; The comparison statistics unit is used to count the number of comparison users who triggered the execution of the second operation behavior, and to obtain the comparison statistics results; The comparison judgment unit is used to determine that the machine learning model training is complete when the sample statistical result is greater than the comparison statistical result; Specifically, the information sending module is used for: When the association probability meets the recommendation information sending condition, the recommendation information is sent to the user to be processed according to the description information of the second operation behavior, so that the user to be processed triggers the execution of the second operation behavior; The device further includes a blocking module, which blocks the sending of recommendation information when the association probability does not meet the conditions for sending recommendation information.

6. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the information push method according to any one of claims 1-4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that are used to cause a processor to execute the information push method according to any one of claims 1-4.

Images