Information pushing method and device and computer equipment

Abstract

The embodiment of the invention provides an information pushing method and device and computer equipment, the method is applied to the computer equipment, and the method comprises the steps that registration information and interaction data corresponding to a user are acquired, and the interaction data comprise information interaction data between the user and information; and device interaction data between the user and the computer device; determining key features based on the interaction data and a preset hierarchical decision model; determining a user type of the user based on the registration information; and based on the key features and the user type, determining the pushing frequency of pushing information to the user. The user experience and the information delivery efficiency can be balanced.

Description

Technical Field

[0001] This application belongs to the field of big data technology, relates to information push technology, and particularly relates to an information push method, device and computer device. Background Art

[0002] Information push is a form of information release that delivers information to terminal access users. By pushing information to users when they are connected to the network, for example, it can include information subscribed by users, or information predicted to be of interest to users, etc. However, pushing information to users too frequently may affect the user experience, and too infrequent information push will affect the information delivery efficiency. In related technologies, it is usually dependent on capturing short-term user behaviors (such as information click-through rate) to adjust the information delivery frequency. However, the regulation mechanism relying on short-term behavior indicators such as click-through rate is prone to situations of redundant information exposure and resource misallocation, and it is difficult to balance the user experience and information delivery efficiency. Summary of the Invention

[0003] Embodiments of this application provide an information push method, device and computer device, which can solve the technical problem that it is difficult to balance the user experience and information delivery efficiency in related technologies.

[0004] In a first aspect of the embodiments of this application, an information push method is provided, which is applied to a computer device. The method includes: obtaining registration information and interaction data corresponding to a user, where the interaction data includes information interaction data between the user and information, and device interaction data between the user and the computer device; determining key features based on the interaction data and a preset hierarchical decision model; determining the user type of the user based on the registration information; and determining the push frequency of pushing information to the user based on the key features and the user type.

[0005] In some embodiments of this application, the determining key features based on the interaction data and a preset hierarchical decision model includes: obtaining the number of information clicks, information residence duration, payment data, the number of times the user uses the functions of the computer device, and the number of information sharing times from the interaction data; determining the payment probability based on the payment data and a preset payment probability prediction model; determining the usage probability based on the number of usage times; obtaining a level score value according to the first sub-model of the hierarchical decision model, the payment probability, the usage probability and the number of information sharing times; determining the user level in the key features through the level score value; determining the information complaint rate and click-through rate deviation based on the number of information clicks; and determining the information sensitivity in the key features according to the second sub-model of the hierarchical decision model, the information complaint rate, the click-through rate deviation and the information residence duration.

[0006] In some embodiments of the present application, determining the user level in the key features based on the level scoring value includes: if the level scoring value is within a first preset range, determining that the user level is the first level; if the level scoring value is within a second preset range, determining that the user level is the second level, where the upper limit value of the second preset range is less than the lower limit value of the first preset range; if the level scoring value is within a third preset range, determining that the user level is the third level, where the upper limit value of the third preset range is less than the lower limit value of the second preset range.

[0007] In some embodiments of the present application, after determining the push frequency of the information pushed to the user, the method further includes: if the information complaint rate is greater than a preset threshold within a preset statistical period, reducing the push frequency.

[0008] In some embodiments of the present application, determining the push frequency of the information pushed to the user based on the key features and the user type includes: if the registration duration of the user is greater than a preset duration, determining that the user type of the user is the first type; for the users of the first type, determining the push frequency according to the user level in the key features.

[0009] In some embodiments of the present application, determining the push frequency according to the user level in the key features includes: if the user level is the first level, determining the push frequency based on a preset reference push frequency, a preset level weight parameter, and the level scoring value corresponding to the first level; if the user level is the second level, determining the push frequency based on the reference push frequency, the level weight parameter, the level scoring value corresponding to the second level, the information sensitivity in the key features, and a preset sensitivity weight parameter; if the user level is the third level, determining the push frequency based on the reference push frequency, the level weight parameter, and the level scoring value corresponding to the third level.

[0010] In some embodiments of the present application, the method further includes: if the registration duration of the user is less than or equal to the preset duration, determining that the user type of the user is the second type; for the users of the second type, determining the push frequency according to the preset frequency and the number of information items pushed each time in the key features.

[0011] In some embodiments of the present application, the information includes advertisements, and the method further includes determining the type of information to be pushed to the user, including: if the user type is the first type and the user level is the first level, pushing brand advertisements to the user; if the user type is the first type, the user level is the second level, and the information sensitivity is low sensitivity, pushing effect advertisements to the user; if the user type is the second type, pushing the brand advertisements to the user.

[0012] The present application also provides an information push device, which is applied to a computer device. The information push device includes: an acquisition module, configured to acquire the registration information and interaction data corresponding to the user, where the interaction data includes the information interaction data between the user and the information, and the device interaction data between the user and the computer device; a data processing module, configured to determine key features based on the interaction data and a preset hierarchical decision model; an information determination module, configured to determine the user type of the user based on the registration information; and an information push module, configured to determine the push frequency of the information to be pushed to the user based on the key features and the user type.

[0013] An embodiment of the present application also provides a computer device, including: a memory and a processor, where the processor executes computer-readable instructions stored in the memory to implement the information push method described above.

[0014] In the information push method provided by the embodiment of the present application, acquiring the registration information and interaction data corresponding to the user provides a data basis for determining the push frequency of the information to be pushed to the user. Determining key features through the interaction data and a preset hierarchical decision model can determine key features at different levels through the hierarchical decision model. Determining the user type of the user through the registration information. Combining the key features with the user type makes full use of key features at different levels and the user type, improving the accuracy of the push frequency of the information to be pushed to the user. Solving the technical problem of being difficult to balance the user experience and the information delivery efficiency, and improving the information push efficiency to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0016] Figure 1 It is a schematic diagram of the application environment of the information push method provided by the embodiment of the present application.

[0017] Figure 2It is a flowchart of an information push method provided by an embodiment of the present application.

[0018] Figure 3 It is a schematic diagram of user type and level division provided by another embodiment of the present application.

[0019] Figure 4 It is a flowchart for determining key features provided by an embodiment of the present application.

[0020] Figure 5 It is a flowchart for pushing information provided by an embodiment of the present application.

[0021] Figure 6 It is a schematic diagram of an information push device provided by an embodiment of the present application.

[0022] Figure 7 It is a schematic diagram of the structure of a computer device provided by an embodiment of the present application. Detailed implementation manners

[0023] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0024] It should be noted that in the present application, "at least one" means one or more, and "a plurality" means two or more than two. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist simultaneously, and B exists alone, where A and B may be singular or plural. The terms "first", "second", "third", "fourth", etc. (if any) in the specification, claims and drawings of the present application are used to distinguish similar objects, rather than to describe a specific order or sequence.

[0025] In the embodiments of the present application, words such as "exemplary" or "for example" are used to represent examples, illustrations or explanations. Any embodiment or design solution described as "exemplary" or "for example" in the embodiments of the present application should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Rather, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific manner. Without conflict, the following embodiments and the features in the embodiments may be combined with each other.

[0026] Information push is a form of information release that delivers information to terminal access users by pushing information to them when they are connected to the network. For example, it can include advertisements subscribed by users or information predicted to be of interest to users. However, pushing information to users too frequently may affect the user experience, and too infrequent information push will affect the information delivery efficiency. In related technologies, it is usually dependent on capturing short-term user behaviors (such as information click-through rate) to adjust the information push frequency. However, the regulation mechanism relying on short-term behavior indicators such as click-through rate is prone to situations of redundant information exposure and resource misallocation, and it is difficult to balance the user experience and the information delivery efficiency.

[0027] To solve the technical problem of being difficult to balance the user experience and the information delivery efficiency, the embodiments of the present application provide an information push method, device, and computer device, which can determine the push frequency of pushing information to users based on the key features and user types determined by a hierarchical decision model, specifically including the push frequency and the number of pushes, and balance the user experience and the information delivery efficiency. First, a schematic diagram of the application environment of the information push method provided by the embodiments of the present application will be described below.

[0028] Figure 1 is a schematic diagram of the application environment of the information push method provided by the embodiments of the present application. As Figure 1 shown, the computer device 10 is communicatively connected to the client 20. The communication method can be wired network communication or wireless network communication. The wired network can be any one of a local area network, a metropolitan area network, and a wide area network, and the wireless network can be any one of Bluetooth (BT), Wireless Fidelity (Wi-Fi), Near Field Communication (NFC), ZigBee Wireless Networks (ZigBee) technology, Infrared (IR) technology, Ultra Wideband (UWB) technology, Universal Serial Bus (USB) wireless, etc.

[0029] The computer device 10 can be an electronic device such as a computer, a personal computer (PC), a server (including an independent physical server, a server cluster, and a cloud server). The computer device 10 can also be a computer system that provides certain services to the client 20. For example, it provides management services corresponding to a game platform, an e-commerce platform, and other application programs.

[0030] The client 20 includes a display screen, and a human-computer interaction interface 210 is displayed on the display screen. The client 20 can be an electronic device such as a mobile phone, a tablet computer, a personal computer (PC), a wearable device, etc. that can run an application (APP). The client 20 can install various applications, such as game applications, e-commerce applications, etc. The human-computer interaction interface 210 can be used to display push messages and can also be used to receive registration information input by the user, etc.

[0031] As Figure 1 shown, the user can view the pushed messages through the human-computer interaction interface 210. For example, the message can remind the user that the client 20 has received a push message in ways such as "new message reminder" or "N new messages". The present application does not limit the prompting method.

[0032] In some embodiments of the present application, the user performs application registration through the human-computer interaction interface 210 and performs related operations (such as click operations) in the application after registration. The client 20 records the user's registration information and the interaction data between the user and the application. The client 20 sends the registration information and the interaction data to the computer device 10 for storage and management. The computer device 10 generates a push frequency according to the registration information and the interaction data, and thus displays it to the user through the human-computer interaction interface 210 based on the push frequency.

[0033] It can be understood that, in addition to the above information push method provided by the embodiments of the present application implemented by data interaction between the computer device 10 and the client 20, it can also be implemented based on the computer device 10 or the client 20 alone. When the information push method provided by the embodiments of the present application is executed by the client 20 alone, the computing process performed by the above computer device 10 can be executed by the processor of the client 20. That is, in this case, the computer device 10 can be understood as the processor of the client 20. When the information push method provided by the embodiments of the present application is executed by the computer device 10 alone, the computer device 10 can perform human-computer interaction with the user through its own display device, so as to implement information push. The present application embodiments do not limit the interaction device for implementing the information push method provided by the embodiments of the present application. Optionally, an example of the information push method executed by the interaction between the computer device 10 and the client 20 is used for illustration.

[0034] Figure 2 is a flowchart of an information push method provided by an embodiment of the present application. As Figure 2 shown, the information push method provided by the embodiments of the present application is applied in a computer device (such as Figure 1 the computer device 10). According to different requirements, the order of the steps in this flowchart can be changed, and some steps can be omitted. AsFigure 2 As shown, it includes the following steps: Step S201, obtain the registration information and interaction data corresponding to the user.

[0035] In some embodiments of the present application, the computer device can communicate with multiple clients, and each client is provided with a corresponding display device. The display device can be a display screen configured on the client, or it can also be a monitor externally connected to the client. The present application does not limit this. The user can interact with the client through the human-computer interaction interface on the display device.

[0036] In some embodiments of the present application, taking the client as a tablet computer and the display device as a touchable display screen as an example for illustration. When it is detected that the user wakes up the display screen of the client, the computer device controls the display screen to display the human-computer interaction interface, and multiple controls are presented on the human-computer interaction interface. Different controls can correspond to different functions. For example, each control corresponds to an application (Application, APP). The user can change the position of the control on the human-computer interaction interface by dragging, and can also enter the application interface corresponding to the control by clicking on the control. Multiple display areas can be presented in the application interface, and by selecting a display area, the information, pictures, etc. corresponding to the display area can be jumped and presented.

[0037] A registration area can also be displayed on the human-computer interaction interface. The user fills in the corresponding registration content in the registration area, and the client packages the registration content filled in by the user according to the interface protocol (such as JSON / XML). The client transmits the packaged registration content to the computer device according to the communication protocol with the computer device. After receiving the registration content sent by the client, the computer device verifies the registration content. In the case where the registration content passes the verification, the computer device can send a registration success instruction to the client and generate registration information according to the registration content. The client can remind the user that the registration has been completed through the human-computer interaction interface. The registration information can include, but is not limited to, the registration time, user identity information, contact phone number, etc.

[0038] Users can operate all applications on the client through the human-computer interaction interface to implement corresponding functions, and can also interact with the prompt information through the human-computer interaction interface. In one example, the user can click on an application, and the client records the device interaction data between the user and the computer device. The user can also click on the push information popped up in the application, and the push information includes advertisements, and the client can record the information interaction data between the user and the information. When the client is in an online state, the interaction data can be sent to the computer device in real time, and the interaction data can include device interaction data and information interaction data. When the client is in an offline state, the interaction data can be temporarily stored in the client until it is connected to the network, and then the interaction data is uploaded to the computer device.

[0039] In some embodiments of the present application, after the computer device obtains the registration information and interaction data corresponding to the user, it stores and manages the registration information and interaction data corresponding to the user. In order to facilitate subsequent personalized processing of the user, the identity information can be associated with the corresponding registration information and interaction data, so that after the identity information of the user is determined, it is associated with the recorded corresponding interaction data. In addition, the corresponding registration information and interaction data can be quickly queried through the identity information.

[0040] Step S202, determine the key features based on the interaction data and the preset hierarchical decision model.

[0041] In some embodiments of the present application, the computer device can pre-deploy a hierarchical decision model. The hierarchical decision model is a systematic method that decomposes complex decision-making problems into multiple levels and optimizes the decision-making process by analyzing and synthesizing various factors at each level layer by layer. Its core lies in integrating decision-making elements with different levels of abstraction through a hierarchical structure, combining quantitative and qualitative analysis to improve the scientificity and adaptability of decision-making.

[0042] The interaction data can include the number of information clicks, the time of clicking the information, the type of the clicked information, the duration of information stay, payment data, the number of times the user uses the functions of the computer device, the number of information sharing times, and the user active time, etc. Among them, the number of information clicks can include the number of times of clicking the closing information, the number of times of clicking the complaint information, etc. In one example, taking the information as an advertisement for illustration, after the computer device pushes an advertisement through the human-computer interaction interface, if the user clicks on the advertisement and enters the details page, the number of times of clicking the advertisement is recorded once. The advertisement details page includes a closing control and a complaint control. If the user clicks on the closing control, the number of times of clicking the closing advertisement is recorded once. If the user clicks on the complaint control, in order to avoid the user's misoperation, the user needs to click on the complaint control twice, and the operation of completing two clicks on the complaint control is recorded as the number of times of clicking the complaint advertisement once.

[0043] In another example, taking news as the information for illustration, after the computer device pushes news through the human-computer interaction interface, if the user clicks on the news and enters the details page, the number of times of clicking on the news is recorded. The news details page includes a close control and a complaint control. If the user clicks on the close control, the number of times of clicking to close the news is recorded. If the user clicks on the complaint control, indicating that the user may not be interested in this type of news or doubts the authenticity of this news, then to avoid misoperation by the user, the user needs to click on the complaint control twice, and the operation of clicking on the complaint control twice is recorded as the number of times of clicking to complain about the news.

[0044] The time of clicking on the information can include the time when the user clicks to close the information and the time when the user clicks to complain about the information; the type of the clicked information can include the type of the information corresponding to the clicked close information and the type of the information corresponding to the clicked complaint information; the information stay duration can be the time when the user does not perform any operation on the pushed information after the computer device pushes the information; the payment data can include the number of times the user uses the payment function based on the pushed information and the number of times of entering the payment page, etc.; the number of times the user uses the functions of the computer device can include the number of times the user uses all the functions on the computer device, and can also include the number of times the user uses the local functions within any application; the number of times of information sharing can include the number of times the user clicks to share the information; the user active time can include the time when the user uses the computer device, and can also include the time when the user uses any application, etc.

[0045] In some embodiments of the present application, the computer device can receive the interaction data generated by the user's real-time operation, or can periodically receive the interaction data sent by the client. The computer device can periodically call the hierarchical decision model to process the interaction data, or can call the hierarchical decision model to process the interaction data when meeting the preset conditions. Among them, the periodic processing method can be that when reaching the preset time, the computer device automatically calls the hierarchical decision model to process the interaction data according to the triggering mechanism. For the situation of meeting the preset conditions, since the activity of each user is different, it can be determined that the preset conditions are met when the quantity of the interaction data reaches the preset quantity. Or, based on the interaction data, determine the information complaint rate, and determine that the preset conditions are met when the complaint rate reaches the preset threshold.

[0046] In some embodiments of the present application, through the processing of the interaction data by the hierarchical decision model, the user level and information sensitivity in the key features can be obtained. Specifically, the first model in the hierarchical decision model can be used to process the interaction data to obtain the user level in the key features. The second model in the hierarchical decision model can be used to process the interaction data to obtain the information sensitivity in the key features. The processing processes of the first model and the second model can refer to the following Figure 4The embodiments shown.

[0047] In other embodiments of the present application, since the interaction data is data that increases over time, in order to avoid less interaction data affecting the accuracy of key feature determination, the computer device may preset a benchmark push frequency as a key feature, and the benchmark push frequency includes a preset frequency and the number of messages pushed each time. Therefore, the key features may include one or more types of data such as the benchmark push frequency, user level, and information sensitivity.

[0048] Step S203: Determine the user type of the user based on the registration information.

[0049] In some embodiments of the present application, the registration information may include the registration duration. The user type may include a first type and a second type. If the user's registration duration is greater than the preset duration, then the user with a registration duration greater than the preset duration is determined as a user of the first type. If the user's registration duration is less than or equal to the preset duration, then the user with a registration duration less than or equal to the preset duration is determined as a user of the second type. Among them, the preset duration may be set according to actual needs. For example, the preset duration may be seven days.

[0050] In one example, taking the preset duration as 7 days as an example, users with a registration duration greater than 7 days are marked as users of the first type, which can also be called old users. Users with a registration duration less than or equal to 7 days are marked as users of the second type, which can also be called new users.

[0051] Step S204: Determine the push frequency of pushing information to the user based on the key features and the user type.

[0052] In some embodiments of the present application, after determining the user type, for users of different user types, the push frequency of pushing information to the user can be determined based on different key features, and the push frequency includes the push frequency and the number of times. As Figure 3 shown, the user type is divided into a first type and a second type. For the first type, it can be divided into a first level, a second level, and a third level according to the user level of the key features. For the second level, it can be further divided into a first sensitivity level and a second sensitivity level according to the information sensitivity of the key features. Among them, the division basis of the first level, the second level, the third level, the first sensitivity level, and the second sensitivity level can refer to the embodiments shown in the following Figure 4 The embodiments shown.

[0053] In some embodiments of the present application, for users of the first type and belonging to the first level, the push frequency is determined according to the user level in the key features. Specifically, if the level score value is within the first preset range, the user level is determined to be the first level, and users at the first level can be high-value users. Based on the preset baseline push frequency, the preset level weight parameter, and the level score value corresponding to the first level, the push frequency is determined. It is expressed by the formula: , where represents the push frequency, represents the baseline push frequency, represents the level weight parameter, represents the level score value. Substituting the baseline push frequency, the preset level weight parameter, and the level score value corresponding to the first level into this formula, the push frequency can be obtained. In an example, assuming that the baseline push frequency is 10 times per preset period, the level weight parameter is 0.3, and the level score value corresponding to the first level is 0.85, the calculated push frequency is 7 times per preset period to push information. Among them, the preset period can be per hour, per day, or a preset time period, and the present application does not limit the setting of the preset period.

[0054] In another example, for users of the first type and belonging to the first level, the push frequency can also be determined according to the baseline push frequency and the first preset ratio. For example, assuming that the baseline push frequency is 10 and the first preset ratio is 0.8, the push frequency is 8 times per day to push information.

[0055] In some embodiments of the present application, for users of the first type and belonging to the second level. If the level score value is within the second preset range, and the upper limit value of the second preset range is less than the lower limit value of the first preset range, the user level is determined to be the second level, and these users belong to potential users. Then, the push frequency corresponding to different information sensitivities at the second level can be further determined in combination with the information sensitivity, so as to push information to users at the second level more accurately. For users corresponding to different information sensitivities at the second level, based on the baseline push frequency, the level weight parameter, the level score value corresponding to the second level, the information sensitivity in the key features, and the preset sensitivity weight parameter, the push frequency is determined. It is expressed by the formula: , where represents the sensitivity weight parameter, represents the information sensitivity. In an example, assuming that the baseline push frequency is 10, the level weight parameter is 0.3, the sensitivity weight parameter is 0.5, the level score value corresponding to the second level is 0.7, and the information sensitivity is 0.3, the calculated push frequency is 9 times per day to push information.

[0056] In another example, for users of the first type, belonging to the second level and the first sensitivity level, where the information sensitivity less than or equal to the preset sensitivity threshold is determined as the first sensitivity level. The push frequency can be determined according to the baseline push frequency and the second preset ratio. For example, assuming the baseline push frequency is 10 and the second preset ratio is 1.5, the push frequency is to push information 15 times a day. For users of the first type, belonging to the second level and the second sensitivity level, the push frequency can be determined according to the baseline push frequency and the third preset ratio, where the information sensitivity greater than the preset sensitivity threshold is determined as the second sensitivity level. For example, assuming the baseline push frequency is 10 and the third preset ratio is 0.7, the push frequency is to push information 7 times a day.

[0057] In some embodiments of the present application, for users of the first type and belonging to the third level. If the level score value is within the third preset range, and the upper limit value of the third preset range is less than the lower limit value of the second preset range, then the user level is determined to be the third level. The push frequency can be determined based on the baseline push frequency, the level weight parameter, and the level score value corresponding to the third level. Expressed by the formula: . In one example, assuming the baseline push frequency is 10, the level weight parameter is 0.3, and the level score value corresponding to the third level is 0.25, the calculated push frequency is to push information 9 times a day.

[0058] In another example, for users of the first type and belonging to the third level. The push frequency can be determined based on the baseline push frequency and the fourth preset ratio. For example, assuming the baseline push frequency is to push 10 times per preset period, and the fourth preset ratio is 2, the push frequency is to push 20 times per preset period.

[0059] The above are just examples, and the first preset ratio, the second preset ratio, the third preset ratio, and the fourth preset ratio can be key features set according to actual applications.

[0060] For users of the second type, the push frequency can be determined according to the preset frequency and the number of messages pushed each time in the key features, so as to enable the protection mode for users of the second type and reduce the uninstallation rate of new users caused by frequent message pushing.

[0061] In other embodiments of the present application, if the information complaint rate is greater than the preset threshold within the preset statistical period, it means that the user may not want to be disturbed by message pushing during this period, then the push frequency can be reduced. In one example, if the information complaint rate suddenly increases by 50%, the frequency is immediately reduced by 30%.

[0062] Through the above embodiments, the registration information and interaction data corresponding to the user are obtained, providing a data basis for determining the push frequency of information pushed to the user. By determining the key features through the interaction data and the preset hierarchical decision model, the key features at different levels can be determined through the hierarchical decision model. The user type of the user is determined through the registration information. By combining the key features with the user type, the key features at different levels and the user type are fully utilized to improve the accuracy of the push frequency of information pushed to the user. This solves the technical problem of being difficult to balance the user experience and the information delivery efficiency, and improves the information push efficiency to a certain extent. In particular, by dividing users into different levels and further dividing the sensitivity of users in the second level to information, the accuracy of the pushed information is improved, and the situation of fuzzy push is avoided.

[0063] In addition, compared with the simplification of information (such as advertisements) delivery strategies in related technologies, the present application can consider the user's tolerance of advertisements from multiple dimensions, thereby enhancing the user experience of information push and the accuracy of information push. Moreover, in the embodiments of the present application, without affecting the normal use of the user at all, the user value evaluation and advertisement sensitivity detection are automatically completed through the acquired hierarchical decision model, avoiding the abruptness of the user being labeled in the traditional solution. The advertisement frequency adjustment adopts a progressive transition strategy to ensure that the user does not perceive a sudden change in the advertisement volume. For example, the daily increase or decrease does not exceed 2 exposures, and this smooth transition significantly improves the usage fluency. Regarding the new user experience, the system specifically designs a protective strategy. The first advertisement display will be triggered only on the second day after the new user registers. At the same time, the complaints and retention situations of each user in the first seven days will be concerned, and the advertisement frequency will be dynamically adjusted in real time to achieve a balance between revenue and experience.

[0064] Figure 4 It is a flowchart for determining the key features provided by an embodiment of the present application. As Figure 4 shown, the key features can be determined through the interaction data and the hierarchical decision model, including the following steps: Step S401, obtain the information click count, information stay duration, payment data, the usage count of the functions of the computer device by the user, and the information sharing count from the interaction data.

[0065] In some embodiments of the present application, interaction data will be generated during the user's interaction with the client. The interaction data may include, but is not limited to, the information click count, information stay duration, payment data, the usage count of the functions of the computer device by the user, and the information sharing count, etc.

[0066] Among them, the number of information clicks may include the number of times of clicking to close the information, the number of times of clicking to complain about the information, etc.; the information stay duration may be the time when the user does not perform any operation on the pushed information after the computer device pushes the information; the payment data may include the number of times the user uses the payment function based on the pushed information, the number of times of entering the payment page, and the trigger points for triggering the user to pay; the number of times the user uses the functions of the computer device may include the number of times the user uses all the functions on the computer device, and may also include the number of times the user uses the local functions within any application; the number of information sharing times may include the number of times the user clicks to share the information; the user active time may include the time when the user uses the computer device, and may also include the time when the user uses any application, etc.

[0067] In addition, the interaction data may further include the time of clicking the information and the type of the clicked information, etc.

[0068] Step S402: Determine the payment probability based on the payment data and the preset payment probability prediction model.

[0069] In some embodiments of the present application, the payment probability prediction model is used to predict the potential of user payment. The construction of the payment probability prediction model can be based on a variety of methods, and the specific selection needs to be combined with data characteristics, business scenarios, and prediction goals. The payment probability prediction model can be a logistic regression model, decision tree and random forest, gradient boosting machine combined with the Boosting algorithm, deep cross-attention network, behavior sequence model, Transformer model, multi-task learning model, and Bayesian network, etc.

[0070] Taking the Transformer model as an example of the payment prediction probability for illustration. Before inputting the payment data into the Transformer model, data cleaning and format unification are required: removing outliers (such as negative payment amounts) and correcting missing fields (completing the active days through time interpolation), and at the same time standardizing the formats of multi-source data (user attributes, behavior logs, marketing records) (such as unifying the timestamp to YYYY-MM-DD, and normalizing the numerical fields by Z-score). Subsequently, based on the cleaned data, time series features are extracted to construct a user behavior sequence, including historical payment frequency, active days in the past 30 days, task completion rate, etc., and the time series information is explicitly injected through the sine function position encoding.

[0071] In the model core layer, the multi-head self-attention mechanism is adopted to dynamically capture the long-term and short-term dependencies of the behavior sequence: for example, the strong correlation between the current payment behavior and the high active days in the past 7 days is given a higher attention weight, while the low-frequency historical behaviors are weakened. Further, the cross-attention mechanism is used to correlate external variables: taking the user behavior sequence as the Query, and the marketing activity features (such as promotion intensity, activity period) as the Key / Value to achieve cross-modal feature fusion. For example, when a user participates in a "time-limited discount" activity, the model automatically enhances the attention intensity of the key nodes in its payment path.

[0072] Finally, the output of the Transformer encoder is aggregated into a global feature vector through global average pooling, mapped to a single neuron through a fully connected layer, and converted into a payment probability by the Sigmoid function.

[0073] Step S403, determine the usage probability based on the usage times.

[0074] In some embodiments of the present application, the usage probability may be the function usage rate. In one example, the number of times a user uses the camera function within a preset time period is counted, and the number of active times the user opens the camera software within the preset time period is counted. Then, according to the number of times the camera function is used within the preset time period and the number of active times the camera software is opened within the preset time period, the usage probability of using the camera function is calculated. The above is only an example, and it may also be to calculate the usage probability of other functions, such as the payment function, the video function, etc.

[0075] Step S404, obtain the level score value according to the first sub-model of the hierarchical decision model, the payment probability, the usage probability, and the information sharing times.

[0076] In some embodiments of the present application, the first sub-model is expressed by the formula as follows: level score value (UVS) = 0.4 × payment probability + 0.3 × usage probability + 0.3 × information sharing times. After determining the payment probability, the usage probability, and the information sharing times, input the payment probability, the usage probability, and the information sharing times into the first sub-model, and the level score value (UVS) can be calculated.

[0077] Step S405, determine the user level in the key features through the level score value.

[0078] In some embodiments of the present application, if the level score value is within the first preset range, the user level is determined to be the first level. Among them, the first preset range may include UVS≥80. If the level score value is within the second preset range, the user level is determined to be the second level. Among them, the upper limit value of the second preset range is less than the lower limit value of the first preset range, so the second preset range may include 30≤UVS<80. If the level score value is within the third preset range, the user level is determined to be the third level. Among them, the upper limit value of the third preset range is less than the lower limit value of the second preset range, so the third preset range may include UVS<30.

[0079] The thresholds (such as 30 and 80) for dividing different levels above are only examples and can be set according to actual situations. The present application does not limit this.

[0080] Step S406: Based on the number of information clicks, determine the deviation degree between the information complaint rate and the click-through rate.

[0081] In some embodiments of the present application, the number of times the user clicks on information complaints is determined from the number of information clicks. The number of times the user clicks on information complaints and the number of information pushes are counted. According to the number of times the user clicks on information complaints and the number of information pushes, the information complaint rate can be calculated. The click-through rate deviation degree represents the deviation degree between the preset optimal click threshold and the actual number of clicks of the user. The number of times the user clicks to view information is counted, and according to the number of times the user clicks to view information and the number of information pushes, the click-through rate deviation degree is calculated.

[0082] Step S407: According to the second sub-model of the hierarchical decision-making model, the information complaint rate, the click-through rate deviation degree, and the information stay duration, determine the information sensitivity in the key features.

[0083] In some embodiments of the present application, the second sub-model is represented by the following formula: Information Sensitivity (ASI) = 0.6×Information Complaint Rate + 0.3×Click-Through Rate Deviation Degree - 0.1×Information Stay Duration. After determining the information complaint rate, the click-through rate deviation degree, and the information stay duration, the information complaint rate, the click-through rate deviation degree, and the information stay duration are input into the second sub-model, and the information sensitivity (ASI) can be calculated.

[0084] If the information sensitivity (ASI) is less than or equal to the preset sensitivity threshold, it is determined to be the first sensitivity level. For example, ASI≤40. If the information sensitivity (ASI) is greater than the preset sensitivity threshold, it is determined to be the second sensitivity level. For example, ASI>40. The thresholds (such as 40) for dividing different sensitivity levels above are only examples and can be set according to actual situations. The present application does not limit this.

[0085] In other embodiments of the present application, after the hierarchical decision model is deployed on a computer device, the impact of the hierarchical decision model on advertising revenue and user retention rate can be periodically verified, so as to update the hierarchical decision model in real time. The verification can be an A / B test. Specifically, set users with a click-through rate of 10% as users who are not sensitive to the user. Set users with a click-through rate of 20% as sensitive users. Perform A\B tests on these two types of users to observe the changes in click-through rate and complaint rate of the two after increasing the frequency of advertisements. If the click-through rate remains unchanged and the revenue increases, it means that the value is optimal. If the click-through rate decreases and the complaint rate increases, then this value must be called back to update the hierarchical decision model.

[0086] Through the above embodiments, users are divided into different levels, and different strategies are implemented for users of different levels. In addition, since the second level belongs to the level of fuzzy interval, in order to push information to users more accurately, the sensitivity level can be further calculated for users of the second level, so as to push information to users in combination with the sensitivity level. The accuracy of information pushed to users is improved, and frequent interruptions to users can be avoided, thereby balancing user experience and information delivery efficiency.

[0087] Ad push is a business form that delivers advertisements to terminal users on wired or wireless networks. It delivers advertisements to users when they access the network, thereby creating profit growth points for network operators and advertisers. Network operators hope to push advertisements to users through applications or clients to expand the audience that receives advertisements. Too high a frequency of ad push may increase the user's application uninstall rate, while too low a frequency of ad push will affect the efficiency of ad delivery. Related ad push solutions rely on capturing users' short-term behaviors (such as ad click-through rates) to adjust the frequency of ad delivery. However, such solutions may cause highly active users to be lost due to frequent ad push, while low-active users will not be able to increase advertising profit points. Therefore, it is particularly important to push what type of advertisements to what type of users. The following is based on Figure 5 Describes what type of information should be pushed to what type of users. Figure 5 As shown, the information is an advertisement for explanation.

[0088] Figure 5 This is a flow chart of information push provided by an embodiment of the present application. Figure 5 As shown, the following steps are included: Step S501, determining whether the user type is the first type.

[0089] In some embodiments of the present application, the first type of user may be an old user whose registration time is longer than the preset time. If the user type is not the first type, step S502 is executed, and if the user type is the first type, step S503 is executed.

[0090] Step S502, push brand ads to the user.

[0091] In some embodiments of the application, brand ads are a marketing and communication method with the core goal of enhancing brand value and shaping a long-term brand image. Its essence is to establish consumers' awareness and loyalty to the brand through differential positioning and emotional resonance.

[0092] Step S503, determine whether the user level is the first level.

[0093] In some embodiments of the present application, users at the first level can be high-value users. For users at the first level, the payment intention can be intelligently identified. When a user at the first level views the payment function page multiple times, the ad frequency will be automatically reduced by 50%, and a prompt for the ad-free privilege will be displayed. When a user at the first level registers for a team, it indicates that they are more likely to pay for premium services and functions. Reducing ads for such users protects the user experience and can also achieve more high-value conversions. If the user level is the first level, step S502 is executed; if the user level is not the first level, step S504 is executed.

[0094] Step S504, determine whether the user level is the second level.

[0095] In some embodiments of the present application, users at the second level can be medium-value users. If the user level is not the second level, step S505 is executed; if the user level is the second level, step S506 is executed.

[0096] Step S505, push ads of any type to the user.

[0097] In some embodiments of the present application, ads of any type can include performance ads, brand ads, social ads, cultural ads, media ads, etc.

[0098] Step S506, determine whether the user's sensitivity is low sensitivity.

[0099] In some embodiments of the present application, if the user's sensitivity level is the first sensitivity level, it is determined that the user's sensitivity is low sensitivity. If the user's sensitivity is low sensitivity, step S507 is executed. If the user's sensitivity is not low sensitivity, step S505 is executed. Among them, non-low sensitivity can be high sensitivity, and high sensitivity can be the second sensitivity level. If the user's sensitivity is high sensitivity, ads of any type can be pushed to the user.

[0100] Step S507, push performance ads to the user.

[0101] In some embodiments of the present application, performance-based advertising is an advertising form oriented towards quantifiable results, and its core lies in that advertisers only need to pay for measurable user behaviors (such as clicks, registrations, purchases, etc.).

[0102] Through the above embodiments, different advertisements can be pushed to users with different levels and different sensitivity levels, improving the accuracy of pushing and the user experience. In addition, the above recommendations for advertisement types with different levels or different sensitivities are only examples and can be adjusted according to actual situations.

[0103] In other embodiments of the present application, the computer device is also provided with an anti-cheating mechanism to detect abnormal click behaviors through the anti-cheating mechanism, thereby excluding robot traffic that affects the evaluation results and improving the accuracy of obtaining interaction data.

[0104] Figure 6 It is a schematic diagram of the information push device provided by the embodiments of the present application. It is a functional embodiment of the information push method of the present application. The information push device includes an acquisition module 601, a data processing module 602, an information determination module 603, and an information push module 604. Among them: The acquisition module 601 is used to acquire the registration information and interaction data corresponding to the user. The interaction data includes the information interaction data between the user and the information, and the device interaction data between the user and the computer device; The data processing module 602 is used to determine key features based on the interaction data and a preset hierarchical decision model; The information determination module 603 is used to determine the user type of the user based on the registration information; The information push module 604 is used to determine the push frequency of pushing information to the user based on the key features and the user type.

[0105] In some embodiments of the present application, the determining of the key features based on the interaction data and a preset hierarchical decision model includes: obtaining the information click count, information stay duration, payment data, the number of times the user uses the functions of the computer device, and the information sharing count from the interaction data; determining the payment probability based on the payment data and a preset payment probability prediction model; determining the usage probability based on the number of times of use; obtaining a level score value according to the first sub-model of the hierarchical decision model, the payment probability, the usage probability, and the information sharing count; determining the user level in the key features through the level score value; determining the information sensitivity in the key features according to the second sub-model of the hierarchical decision model, the information complaint rate, the click-through rate deviation, and the information stay duration.

[0106] In some embodiments of the present application, determining the user level in the key features through the level scoring value includes: if the level scoring value is within a first preset range, determining that the user level is the first level; if the level scoring value is within a second preset range, determining that the user level is the second level, where the upper limit value of the second preset range is less than the lower limit value of the first preset range; if the level scoring value is within a third preset range, determining that the user level is the third level, where the upper limit value of the third preset range is less than the lower limit value of the second preset range.

[0107] In some embodiments of the present application, after determining the push frequency of the information pushed to the user, the method further includes: if the information complaint rate is greater than a preset threshold within a preset statistical period, reducing the push frequency.

[0108] In some embodiments of the present application, determining the push frequency of the information pushed to the user based on the key features and the user type includes: if the registration duration of the user is greater than a preset duration, determining that the user type of the user is the first type; for the users of the first type, determining the push frequency according to the user level in the key features.

[0109] In some embodiments of the present application, determining the push frequency according to the user level in the key features includes: if the user level is the first level, determining the push frequency based on a preset reference push frequency, a preset level weight parameter, and the level scoring value corresponding to the first level; if the user level is the second level, determining the push frequency based on the reference push frequency, the level weight parameter, the level scoring value corresponding to the second level, the information sensitivity in the key features, and a preset sensitivity weight parameter; if the user level is the third level, determining the push frequency based on the reference push frequency, the level weight parameter, and the level scoring value corresponding to the third level.

[0110] In some embodiments of the present application, it further includes: if the registration duration of the user is less than or equal to the preset duration, determining that the user type of the user is the second type; for the users of the second type, determining the push frequency according to the preset frequency in the key features and the number of information items pushed each time.

[0111] In some embodiments of the present application, the information includes advertisements, and the method further includes determining the type of information to be pushed to the user, including: if the user type is the first type and the user level is the first level, pushing brand advertisements to the user; if the user type is the first type, the user level is the second level, and the information sensitivity is low sensitivity, pushing effect advertisements to the user; if the user type is the second type, pushing the brand advertisements to the user.

[0112] Another embodiment of the present application further provides a computer device. Figure 7 It is a schematic structural diagram of the computer device provided by the embodiment of the present application, as Figure 7 shown, in an embodiment of the present application, the computer device 10 may be a tablet computer, an augmented reality (AR) / virtual reality (VR) device, a laptop computer, a netbook, or other devices, and the embodiment of the present application does not impose any restrictions on the specific type of the computer device 10.

[0113] As Figure 7 shown, the computer device 10 may include, but is not limited to, a display screen 1000, a communication module 1001, a memory 1002, a processor 1003, an input / output (I / O) interface 1004, and a bus 1005. The processor 1003 is respectively coupled to the display screen 1000, the communication module 1001, the memory 1002, and the I / O interface 1004 through the bus 1005.

[0114] Those skilled in the art can understand that the schematic diagram is only an example of the computer device 10, and does not constitute a limitation on the computer device 10. It may include more or fewer components than shown in the figure, or combine some components, or different components. For example, the computer device 10 may further include a network access device, etc.

[0115] The display screen 1000 may be a touch screen, which is an inductive touchable liquid crystal display device. Alternatively, the display screen 1000 may also be a non-touch screen. The display screen 1000 is used to display a human-computer interaction interface so that the computer device 10 can interact with the user.

[0116] The communication module 1001 may include a wired communication module and / or a wireless communication module. The wired communication module may provide one or more of the solutions for wired communication such as Universal Serial Bus (USB), Controller Area Network (CAN), etc. The wireless communication module may provide one or more of the solutions for wireless communication such as Wireless Fidelity (Wi-Fi), Bluetooth (BT), mobile communication network, Frequency Modulation (FM), near field communication (NFC), Infrared (IR) technology, etc.

[0117] The memory 1002 can be used to store computer-readable instructions and / or modules. By running or executing the computer-readable instructions and / or modules stored in the memory 1002, and by invoking the data stored in the memory 1002, the processor 1003 realizes various functions of the computer device 10. The memory 1002 may mainly include a program storage area and a data storage area. Among them, the program storage area may store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area may store data created according to the use of the computer device 10. The memory 1002 may include non-volatile and volatile memories, such as: hard disk, memory, plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, at least one magnetic disk storage device, flash device, or other storage devices.

[0118] The memory 1002 can be an external memory and / or an internal memory of the computer device 10. Further, the memory 1002 can be a memory in physical form, such as a memory stick, a TF card (Trans-flash Card), etc.

[0119] The processor 1003 may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor, or the processor may also be any conventional processor, etc. The processor 1003 is the computing core and control center of the computer device 10, connecting various parts of the entire computer device 10 through various interfaces and lines, and executing the operating system of the computer device 10 and various installed application programs, program codes, etc.

[0120] Exemplarily, the computer-readable instructions may be divided into one or more modules / sub-modules / units. One or more modules / sub-modules / units are stored in the memory 1002 and executed by the processor 1003 to complete this application. One or more modules / sub-modules / units may be a series of computer-readable instruction segments capable of performing specific functions, and these computer-readable instruction segments are used to describe the execution process of the computer-readable instructions in the computer device 10.

[0121] If the modules / units integrated in the computer device 10 are implemented in the form of software functional units and sold or used as independent products, they may be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above-mentioned embodiment methods of this application, it may also be completed by instructing relevant hardware through computer-readable instructions. The computer-readable instructions may be stored in a computer-readable storage medium. When the computer-readable instructions are executed by the processor, the steps of the above-mentioned various method embodiments may be implemented.

[0122] Among them, the computer-readable instructions include computer-readable instruction codes, and the computer-readable instruction codes may be in the form of source code, object code, executable files, or some intermediate forms, etc. The computer-readable medium may include: any entity or device capable of carrying the computer-readable instruction codes, recording media, USB flash drives, mobile hard disks, magnetic disks, optical disks, computer memories, Read-Only Memories (ROMs), Random Access Memories (RAMs).

[0123] In combination with Figures 2 to 5, the memory 1002 in the computer device 10 stores computer-readable instructions, and the processor 1003 can execute the computer-readable instructions stored in the memory 1002 to implement the information push method as Figures 2 to 5 shown. Specifically, for the specific implementation method of the processor 1003 for the above computer-readable instructions, reference can be made to Figures 2 to 5 the description of the relevant steps in the corresponding embodiments, which will not be elaborated here.

[0124] The I / O interface 1004 is used to provide a channel for user input or output. For example, the I / O interface 1004 can be used to connect various input and output devices, such as a mouse, a keyboard, a touch device, a display screen, etc., so that the user can input information or visualize information.

[0125] The bus 1005 is at least used to provide a communication channel between the display screen 1000, the communication module 1001, the memory 1002, the processor 1003, and the I / O interface 1004 in the computer device 10.

[0126] In several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of modules is only a logical function division, and there can be other division methods in actual implementation.

[0127] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0128] In addition, in each embodiment of the present application, the various functional modules can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated unit can be implemented in the form of hardware or in the form of a hardware plus a software functional module.

[0129] Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present application is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present application. Any associated drawing marks in the claims should not be regarded as limiting the claimed rights.

[0130] In addition, it is obvious that the term "including" does not exclude other units or steps, and the singular form does not exclude the plural form. Multiple units or devices can also be implemented by one unit or device through software or hardware. Terms such as first and second are used to denote names and do not represent any particular order.

[0131] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and not to limit them. Although the present application has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present application can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present application.

Claims

1. An information push method, applied to a computer device, characterized in that, The method includes: Obtaining the registration information and interaction data corresponding to the user, where the interaction data includes the information interaction data between the user and the information, and the device interaction data between the user and the computer device; Determining key features based on the interaction data and a preset hierarchical decision model; Determining the user type of the user based on the registration information; Determining the push frequency of pushing information to the user based on the key features and the user type.

2. The information pushing method according to claim 1, wherein The determining key features based on the interaction data and a preset hierarchical decision model includes: Obtaining the number of information clicks, the information stay duration, the payment data, the number of times the user uses the functions of the computer device, and the number of information sharing times from the interaction data; Determining the payment probability based on the payment data and a preset payment probability prediction model; Determining the usage probability based on the number of times of use; Obtaining a level score value according to the first sub-model of the hierarchical decision model, the payment probability, the usage probability, and the number of information sharing times; Determining the user level in the key features through the level score value; Determining the information complaint rate and the click-through rate deviation based on the number of information clicks; Determining the information sensitivity in the key features according to the second sub-model of the hierarchical decision model, the information complaint rate, the click-through rate deviation, and the information stay duration.

3. The information pushing method according to claim 2, wherein The determining the user level in the key features through the level score value includes: If the level score value is within the first preset range, determining that the user level is the first level; If the level score value is within the second preset range, determining that the user level is the second level, where the upper limit value of the second preset range is less than the lower limit value of the first preset range; If the level score value is within the third preset range, determining that the user level is the third level, where the upper limit value of the third preset range is less than the lower limit value of the second preset range.

4. The information push method according to claim 2, wherein After determining the push frequency of pushing information to the user, the method further includes: If the information complaint rate is greater than a preset threshold within a preset statistical period, reducing the push frequency.

5. The information push method according to claim 1, wherein The determining the push frequency of pushing information to the user based on the key features and the user type includes: If the registration duration of the user is greater than a preset duration, determining that the user type of the user is the first type; For the users of the first type, determining the push frequency according to the user level in the key features.

6. The information push method according to claim 5, wherein The determining the push frequency according to the user level in the key features includes: If the user level is the first level, determining the push frequency based on a preset benchmark push frequency, a preset level weight parameter, and the level score value corresponding to the first level; If the user level is the second level, determining the push frequency based on the benchmark push frequency, the level weight parameter, the level score value corresponding to the second level, the information sensitivity in the key features, and a preset sensitivity weight parameter; If the user level is the third level, determine the push frequency based on the reference push frequency, the level weight parameter, and the level score value corresponding to the third level.

7. The information pushing method according to claim 5, wherein The method further includes: If the registration duration of the user is less than or equal to the preset duration, determine that the user type of the user is the second type; For the users of the second type, determine the push frequency according to the preset frequency and the number of messages pushed each time in the key features.

8. The information push method according to claim 1, wherein The information includes advertisements, and the method further includes determining the type of information pushed to the user, including: If the user type is the first type and the user level is the first level, push brand advertisements to the user; If the user type is the first type, the user level is the second level, and the information sensitivity is low sensitivity, push effect advertisements to the user; If the user type is the second type, push the brand advertisements to the user.

9. An information push device, applied to a computer device, characterized in that, The information push device includes: An acquisition module, configured to acquire registration information and interaction data corresponding to a user, where the interaction data includes information interaction data between the user and information, and device interaction data between the user and the computer device; A data processing module, configured to determine key features based on the interaction data and a preset hierarchical decision model; An information determination module, configured to determine the user type of the user based on the registration information; An information push module, configured to determine the push frequency of the information pushed to the user based on the key features and the user type.

10. A computer device, characterized in that, including: A memory; and A processor; the processor executes computer-readable instructions stored in the memory to implement the information push method according to any one of claims 1 to 8.

Images