Virtual resource pushing method, electronic equipment and computer program product

Abstract

The embodiment of the invention is suitable for the technical field of computers, and provides a virtual resource pushing method, electronic equipment and a computer program product, and the method comprises the steps: inputting to-be-pushed behavior data corresponding to a user into a preset prediction algorithm, and generating demand prediction information corresponding to the user; determining at least one candidate resource based on the demand prediction information and the behavior data; inputting all the candidate resources into a preset decision algorithm, and determining at least one virtual resource; the decision algorithm is used for determining the virtual resource from all the candidate resources according to operation information of at least one pushed resource corresponding to the user; and pushing the virtual resource to the user. According to the method provided by the embodiment of the invention, the pushing accuracy of the virtual resources can be improved.

Description

Technical Field

[0001] The embodiments of the present application belong to the field of computer technology, and particularly relate to a method for pushing virtual resources, an electronic device, and a computer program product. Background Art

[0002] With the rapid development of information technology, various online service platforms have emerged like mushrooms after a spring rain, bringing great convenience to many aspects of people's lives, studies, entertainments, etc. In order to meet the diverse needs of users and improve the user experience, online service platforms often regularly push virtual resources used on the service platform to users, such as coupons used for shopping. However, in the prior art, online service platforms often adopt a unified push mode, pushing the same virtual resources to all users. For example, when the online service platform is an e-commerce platform, the e-commerce platform can push a discount coupon with a fixed amount to a user on the user's birthday. However, this method of pushing the same virtual resources to all users easily leads to a large number of virtual resources that do not meet the user's needs accumulating in the user's account, and it is difficult for the user to obtain truly useful virtual resources from them. Thus, it can be seen that the existing virtual resource push methods are single and do not perform personalized push according to user needs, thereby reducing the push accuracy of virtual resources. Summary of the Invention

[0003] In view of this, the embodiments of the present application provide a method for pushing virtual resources, an electronic device, and a computer program product to improve the push accuracy of virtual resources.

[0004] The first aspect of the embodiments of the present application provides a method for pushing virtual resources, including:

[0005] Inputting the behavior data corresponding to the user to be pushed into a preset prediction algorithm to generate demand prediction information corresponding to the user;

[0006] Determining at least one candidate resource based on the demand prediction information and the behavior data;

[0007] Inputting all the candidate resources into a preset decision algorithm to determine at least one virtual resource; the decision algorithm is used to determine the virtual resource from all the candidate resources according to the operation information of at least one pushed resource corresponding to the user;

[0008] Pushing the virtual resource to the user.

[0009] In a possible implementation manner of the first aspect, the demand prediction information includes an expected usage object and an expected consumption resource corresponding to the expected usage object;

[0010] The determining at least one candidate resource based on the demand prediction information and the behavior data includes:

[0011] Input the behavior data into a feature extraction algorithm to determine the user identifiers corresponding to the user in different identification dimensions.

[0012] According to the user identifiers corresponding to each of the identification dimensions and the current time, determine at least one candidate resource type corresponding to the user; the user identifier is used, together with the judgment condition corresponding to the user identifier, to determine one candidate resource type corresponding to the user; the current time is used, together with the relationship with holiday information, to determine one candidate resource type corresponding to the user.

[0013] For any candidate resource type, according to the expected consumed resources corresponding to the expected usage object, the average consumed resources corresponding to the expected usage object, the operation information of at least one pushed resource corresponding to the user, and the preset range of any candidate resource type, determine at least one candidate resource quantity corresponding to any candidate resource type.

[0014] According to the candidate resource type and the candidate resource quantity, determine the candidate resources corresponding to the user.

[0015] In a possible implementation manner of the first aspect, the inputting all the candidate resources into a preset decision algorithm to determine at least one virtual resource includes:

[0016] Input the at least one pushed resource corresponding to the user and the operation information corresponding to the pushed resource into a preset objective function to determine the weight coefficients corresponding to each candidate resource; the objective function is used to determine the weight coefficients corresponding to the candidate resources according to the operation information of the pushed resources corresponding to the candidate resources.

[0017] Input the weight coefficients corresponding to all the candidate resources into a preset policy function to determine the at least one virtual resource.

[0018] In a possible implementation manner of the first aspect, the objective function includes a conversion rate objective function, a retention rate objective function, and a resource objective function; the operation information of the pushed resources includes the receiving status of the pushed resources, the usage status of the pushed resources, the browsing information of the user regarding the applicable objects corresponding to the pushed resources, and the first transaction information of the user regarding the pushed resources.

[0019] The inputting the operation information of at least one pushed resource corresponding to the user into a preset objective function to determine the weight coefficients corresponding to each virtual resource includes:

[0020] For any candidate resource type, input the usage status of the pushed resources corresponding to the any candidate resource type into the conversion rate objective function to determine the first weight coefficient corresponding to the any candidate resource type and the second weight coefficients of the respective candidate resource quantities corresponding to the any candidate resource type;

[0021] For any candidate resource type, input the receiving status of the pushed resources corresponding to the any candidate resource type and the browsing information into the retention rate objective function to determine the third weight coefficient corresponding to the any candidate resource type and the fourth weight coefficients of the respective candidate resource quantities corresponding to the any candidate resource type;

[0022] For any candidate resource type, input the historical consumed resources corresponding to each usage object in the first transaction information corresponding to the any candidate resource type and the average consumed resources corresponding to the usage object into the resource objective function to determine the fifth weight coefficient corresponding to the any candidate resource type and the sixth weight coefficients of the respective candidate resource quantities corresponding to the any candidate resource type;

[0023] Correspondingly, the step of inputting the weight coefficients corresponding to all the virtual resources into a preset policy function to determine the at least one virtual resource includes:

[0024] Input the first weight coefficient, the third weight coefficient, and the fifth weight coefficient into a first policy function to determine the target resource type from the candidate resource types;

[0025] Input the second weight coefficient, the fourth weight coefficient, and the sixth weight coefficient into a second policy function to determine the target resource quantity from all the candidate resource quantities corresponding to the target resource type;

[0026] Determine the virtual resource according to the target resource type and the target resource quantity.

[0027] In a possible implementation manner of the first aspect, the demand prediction information includes the expected usage probability and expected usage frequency of the virtual resources for any user:

[0028] The step of determining at least one candidate resource based on the demand prediction information and the behavior data includes:

[0029] If the expected usage probability is greater than a preset first probability threshold, determine at least one resource push time period corresponding to the user according to the expected usage frequency and holiday information;

[0030] Within the resource push time period, determine at least one candidate resource based on the demand prediction information and the behavior data.

[0031] In a possible implementation of the first aspect, pushing the virtual resource to the user includes:

[0032] Determining the push time corresponding to the virtual resource according to the resource push period, the user's historical online period, and the holiday information;

[0033] Determining the applicable scope corresponding to the virtual resource based on the demand prediction information, the user's transaction information, and the push time;

[0034] At the push time, pushing the virtual resource recording the applicable scope to any one of the users.

[0035] In a possible implementation of the first aspect, the demand prediction information includes the expected usage objects and the expected transaction probabilities corresponding to each of the expected usage objects; the applicable scope includes the applicable time and the applicable objects;

[0036] The pushing the virtual resource recording the applicable scope to any one of the users at the push time includes:

[0037] If the expected transaction probability of any one of the expected usage objects is greater than a preset second probability threshold, determining the any one of the expected usage objects as the applicable object of the virtual resource;

[0038] Obtaining the second transaction information of any one of the users about the applicable object;

[0039] Determining the applicable time corresponding to the virtual resource according to the time stamps in each of the second transaction information, the user's usage frequency of the pushed resources, the user's historical online period, and the holiday information;

[0040] At the push time, pushing the virtual resource recording the applicable object and the applicable time to any one of the users.

[0041] In a possible implementation of the first aspect, the behavior data includes at least one of the user's transaction information, historical online period, login frequency, registration duration, browsing information about the usage object, and operation information about the pushed resources.

[0042] A second aspect of the embodiments of the present application provides a virtual resource push device, including:

[0043] A demand prediction module, configured to input the behavior data corresponding to the user to be pushed into a preset prediction algorithm to generate the demand prediction information corresponding to the user;

[0044] A candidate resource determination module, configured to determine at least one candidate resource based on the demand prediction information and the behavior data;

[0045] A virtual resource determination module, configured to input all the candidate resources into a preset decision algorithm to determine at least one virtual resource; the decision algorithm is used to determine the virtual resource from all the candidate resources according to the operation information of at least one pushed resource corresponding to the user;

[0046] A push module, configured to push the virtual resource to the user.

[0047] A third aspect of the embodiments of the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for pushing virtual resources as described in the first aspect above is implemented.

[0048] A fourth aspect of the embodiments of the present application provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the method for pushing virtual resources as described in the first aspect above is implemented.

[0049] A fifth aspect of the embodiments of the present application provides a computer program product, and when the computer program product runs on a computer, the computer is enabled to execute the method for pushing virtual resources as described in the first aspect above.

[0050] Compared with the prior art, the embodiments of the present application have the following advantages:

[0051] In the embodiments of the present application, an electronic device can first perform computational analysis on the behavior data of a user through a prediction algorithm, excavate the potential demands behind the user's behavior, and generate demand prediction information corresponding to the user. By combining the generated demand prediction information and the user's behavior data, at least one candidate resource is determined. It can be seen that in this embodiment, the candidate resources are determined by the electronic device according to the user's demands and past behaviors, meeting the user's demands and behaviors. Further, the electronic device can determine the virtual resources corresponding to the user from the candidate resources through a decision algorithm, and the decision algorithm not only considers the candidate resources themselves but also combines the operation information of the user on the pushed resources. This means that the decision algorithm can dynamically adjust the screening method for the current candidate resources according to the user's past feedback on the pushed resources, such as whether to receive or use them. Therefore, compared with the unified resource push method in the prior art, the method provided in this embodiment can enable the virtual resources pushed by the electronic device to accurately match the actual demands and preferences of the user, thereby improving the accuracy of the virtual resources pushed by the electronic device. Description of the Drawings

[0052] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0053] Figure 1 It is a schematic diagram of a method for pushing virtual resources provided by an embodiment of the present application;

[0054] Figure 2 It is a schematic diagram of another method for pushing virtual resources provided by an embodiment of the present application;

[0055] Figure 3 It is a schematic diagram of another method for pushing virtual resources provided by an embodiment of the present application;

[0056] Figure 4 It is a schematic diagram of the pushing process of coupons on an e-commerce platform provided by an embodiment of the present application;

[0057] Figure 5 It is a schematic diagram of a device for pushing virtual resources provided by an embodiment of the present application;

[0058] Figure 6 It is a schematic diagram of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0059] In the following description, specific details such as specific system structures and technologies are presented for the purpose of illustration rather than limitation, so as to thoroughly understand the embodiments of the present application. However, those skilled in the art should clearly understand that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present application.

[0060] The following uses specific embodiments to illustrate the technical solutions of the present application.

[0061] Referring to Figure 1 , a schematic diagram of a method for pushing virtual resources provided by an embodiment of the present application is shown. This method can be applied to any electronic device such as a computer, a server, and a mobile terminal. The above method for pushing virtual resources can specifically include the following steps:

[0062] S101. Input the behavior data corresponding to the user to be pushed into a preset prediction algorithm to generate demand prediction information corresponding to the user.

[0063] In this embodiment, the electronic device can obtain the corresponding behavior data of the user to be pushed at intervals according to the demand prediction period preset by the R & D personnel. Among them, the user to be pushed can be any user on the platform who allows the platform to push virtual resources. Among them, the virtual resources can be the resources consumed by the user when making an electronic payment on the platform. The electronic device can obtain various different behavior data of the user. Specifically, the behavior data obtained by the electronic device can include at least one of the user's transaction information, historical online time period, login frequency, registration duration, browsing information of the user about the usage object, and operation information of the user on the pushed resources.

[0064] Furthermore, the operation information of the pushed resources can include the receiving status of the pushed resources, the usage status of the pushed resources, the browsing information of the user about the applicable object corresponding to the pushed resources, and the first transaction information of the user using the pushed resources for transactions. The usage object can be any object that the user needs to consume resources to obtain on the platform. Exemplarily, when the platform is an e-commerce platform, the usage object can be a commodity. The user's transaction information can include information such as the total historical consumption in the historical transaction corresponding to the transaction information, the usage object obtained by the user in this historical transaction, the historical consumption resources consumed by the user when obtaining each usage object, and the time stamp of this historical transaction. Exemplarily, when the platform is an e-commerce platform, a certain transaction information of the user can include the total payment amount in the historical transaction corresponding to the transaction information, the commodities obtained by the user in this historical transaction, the payment amount of the user for each commodity, and the user's payment time.

[0065] After the electronic device obtains the behavior data of the user to be pushed, it can input all the obtained behavior data into the prediction algorithm preset by the R & D personnel to generate the demand prediction information corresponding to this user. Among them, the prediction algorithm can infer whether the user will use virtual resources at a certain future time point, the usage object that the user may obtain, and the amount of resources that the user may consume by analyzing the user's behavior data. Specifically, the demand prediction information generated by the prediction algorithm can include, but is not limited to, the expected usage probability of the user for virtual resources, the expected usage frequency, the expected usage object of the user for virtual resources, and the expected consumption resources of the user for the expected usage object. It should be noted that the prediction algorithm in the electronic device can be any deep learning algorithm well-known to those skilled in the art, such as Long-Short Term Memory (LSTM), Multi-Layer Perceptron (MLP), Gated Recurrent Unit (GRU), etc. The embodiments of the present application do not specifically limit the prediction algorithm.

[0066] In a possible implementation, the electronic device may include multiple prediction algorithms, and each prediction algorithm can be used to generate a corresponding demand prediction information for the user. Specifically, the electronic device may include a usage object prediction algorithm, a usage probability prediction algorithm, and a resource consumption prediction algorithm. Among them, the usage object prediction algorithm can be used to predict whether the user has a demand for obtaining a certain type of usage object. Therefore, the usage object prediction algorithm can be used to generate an expected usage object according to the user's behavior data. The probability prediction algorithm can be used to generate the expected usage probability of the user for virtual resources according to the behavior data. The resource consumption prediction algorithm can be used to generate the total expected resource consumption corresponding to the user or the expected resource consumption of the user for a certain expected usage object according to the behavior data.

[0067] Exemplarily, when the platform is an e-commerce platform and the virtual resource is a coupon, the expected usage object generated by the usage object prediction algorithm can be a certain type of commodity that the user needs to purchase; the expected usage probability generated by the probability prediction algorithm can be the usage probability of the user for the coupon; the total expected resource consumption generated by the resource consumption prediction algorithm can be the total expected amount that the user may consume; the expected resource consumption corresponding to the expected usage object generated by the resource consumption prediction algorithm is the expected amount that the user may consume for a certain type of commodity.

[0068] S102. Determine at least one candidate resource based on the demand prediction information and the behavior data.

[0069] In this embodiment, after the electronic device determines the demand prediction information of the user, it can determine at least one candidate resource type corresponding to the user from multiple preset resource types according to the demand prediction information and the behavior data corresponding to the user, and determine the candidate resource amount corresponding to the user from the multiple preset resource amounts corresponding to the candidate resource type. Then, the terminal device can determine at least one candidate resource corresponding to the user according to the candidate resource type and the candidate resource amount. Exemplarily, when the platform is an e-commerce platform, the candidate resource can be a candidate coupon.

[0070] In a possible implementation, the demand prediction information may include the expected usage probability and the expected usage frequency of the user for the virtual resources. After generating the demand prediction information of the user, the electronic device may determine whether the expected usage probability in the demand prediction information is greater than a preset first probability threshold. If the electronic device determines that the expected usage probability is less than or equal to the first probability threshold, the electronic device may not generate candidate resources corresponding to the user and may not push virtual resources to the user. If the electronic device determines that the expected usage probability is greater than the first probability threshold, the electronic device may determine at least one resource push time period corresponding to the user according to the expected usage frequency and the holiday information in the demand prediction information. Then, within the resource push time period, the electronic device may determine at least one candidate resource of the user based on the demand prediction information corresponding to the user and the behavior data corresponding to the user, so as to push virtual resources to the user.

[0071] Through the method provided in this embodiment, since the electronic device can determine whether to push virtual resources to the user according to the expected usage probability, the method provided in this embodiment can reduce the phenomenon that the electronic device pushes virtual resources to users who do not use virtual resources, thereby reducing resource waste and improving the utilization rate of virtual resources.

[0072] S103. Input all candidate resources into a preset decision algorithm to determine at least one virtual resource.

[0073] In this embodiment, after determining all candidate resources corresponding to the user, the electronic device may input all the determined candidate resources into a decision algorithm preset by the R & D personnel, so as to determine at least one virtual resource corresponding to the user from all candidate resources through the decision algorithm. Exemplarily, the decision algorithm may be a reinforcement learning algorithm.

[0074] S104. Push the virtual resource to the user.

[0075] In this embodiment, after determining the virtual resource, the electronic device may send the determined virtual resource to the user terminal logged in with the user account, so as to push the determined virtual resource to the user.

[0076] In a possible implementation, after determining the virtual resource, the electronic device may determine the push time corresponding to the virtual resource according to the resource push time period corresponding to the virtual resource, the user's historical online time period, and the holiday information. The electronic device may also determine the applicable scope corresponding to the virtual resource according to the demand prediction information corresponding to the user, the transaction information of the user, and the push time corresponding to the virtual resource. After determining the push time and the applicable scope, the electronic device may push the virtual resource including the applicable scope to the user at the push time.

[0077] Specifically, virtual resources that do not contain applicable time by themselves. The applicable scope of virtual resources can include the applicable time and applicable objects corresponding to the virtual resources. After the electronic device determines the push time corresponding to the virtual resources, it can determine whether there is any expected transaction probability of any expected user object in the demand prediction information that is greater than a preset second probability threshold. For virtual resources that do not contain applicable objects by themselves, if the in-vehicle terminal determines that the expected transaction probabilities of all expected user objects in the demand prediction information are less than or equal to the second probability threshold, the electronic device can determine all user objects in the platform as the applicable objects corresponding to the virtual resources. The electronic device can obtain all first transaction information of the user regarding the pushed resources, and determine the applicable time corresponding to the virtual resources based on the time stamps of each first transaction information, the usage frequency of the user for the pushed resources, the user's historical online periods, and holiday information. If the in-vehicle terminal determines that there is at least one expected user object in the demand prediction information whose expected transaction probability is greater than the second probability threshold, the electronic device can determine all expected user objects with expected transaction probabilities greater than the second probability threshold as the applicable objects corresponding to the virtual resources. The electronic device can obtain all second transaction information of the user regarding the applicable objects, and determine the applicable time corresponding to the virtual resources based on the time stamps of each second transaction information, the usage frequency of the user for the pushed resources, the user's historical online periods, and holiday information. Exemplarily, the holiday information can be such as Double Eleven, Mid-Autumn Festival, Spring Festival, etc.

[0078] Through the method provided in this embodiment, since the electronic device can generate virtual resources corresponding to the user personalized according to the user's behavior data, holiday information, and the user's transaction information, therefore, the method provided in this embodiment can improve the push accuracy of virtual resources, and further improve the user conversion rate and the platform revenue.

[0079] Figure 2 Fig. shows the specific implementation flowchart of a virtual resource push method S102 provided in the second embodiment of the present application. Refer to Figure 2 , compared with Figure 1 the embodiment described above, S102 in a virtual resource push method provided in this embodiment includes: S1021 - S1024, which are specifically described in detail as follows:

[0080] S1021. Input the behavior data into the feature extraction algorithm to determine the user identifiers corresponding to the user in different identification dimensions.

[0081] In this embodiment, before generating candidate resources corresponding to the user, the electronic device can first input all behavior data corresponding to the user into a preset feature extraction algorithm to generate user identifiers corresponding to the user in different identification dimensions through the feature extraction algorithm.

[0082] In a possible implementation, the identification dimension may include an object preference dimension, a resource sensitivity dimension, and an activity dimension. For the user preference dimension, the feature extraction algorithm may obtain the historical transaction frequency of the user for each usage object according to the transaction information in the user behavior data. The electronic device may determine whether there is a historical transaction frequency corresponding to any usage object that is greater than a preset third frequency threshold. If the electronic device determines that the historical transaction frequency corresponding to any one usage object is greater than the third frequency threshold, the electronic device may generate a preference identifier corresponding to the usage object based on the usage object, and use the generated preference identifier as the identifier of the user in the user preference dimension. Among them, the preference identifier corresponding to a certain usage object can be used to indicate that the user prefers to obtain the usage object. If the electronic device determines that the historical transaction frequencies of the user for all usage objects are less than or equal to the third frequency threshold, the electronic device may not generate a preference identifier for the user preference dimension.

[0083] For the resource sensitivity dimension, the electronic device may obtain the historical consumed resources of the user for each usage object in each historical transaction according to the transaction information in the user behavior data. Then, the electronic device may determine the user identifier of the user in the resource sensitivity dimension according to the relationship between the historical consumed resources of the user for each usage object and the average consumed resources corresponding to the usage object. Specifically, the identifier of the resource sensitivity dimension may include a high sensitivity identifier and a low sensitivity identifier. For a certain usage object, the electronic device may determine the average historical consumption of the user for the usage object according to all the historical consumed resources of the user for the usage object. Then, the electronic device may determine the consumption level of the user in the usage object according to the difference between the average historical consumption of the user for the usage object and the average consumed resources corresponding to the usage object. Then, the electronic device may determine the user identifier of the user in the resource sensitivity dimension according to the consumption levels of the user for all usage objects. Exemplarily, when the platform is an e-commerce platform, for usage objects of clothing, if the average payment amount of the user for clothing products is 300 yuan, which is less than the average amount of clothing products, the electronic device may determine that the user identifier of the user in the resource sensitivity dimension is a low sensitivity identifier.

[0084] For the activity dimension, the electronic device can determine the user identification in the resource sensitivity dimension according to the historical online periods in the user behavior data. Specifically, the identification in the activity dimension can include a high-activity identification and a low-activity identification. The electronic device can determine the online frequency and online duration corresponding to the user according to the user's historical online periods. When the electronic device determines that the user's online frequency is greater than a preset fourth frequency threshold and / or the online duration is greater than a preset duration threshold, the electronic device can determine that the user identification in the activity dimension is the high-activity identification. When the electronic device determines that the user's online frequency is less than or equal to the fourth frequency threshold and the online duration is less than or equal to the duration threshold, the electronic device can determine that the user identification in the activity dimension is the low-activity identification.

[0085] S1022. Determine at least one candidate resource type corresponding to the user according to the user identification corresponding to each identification dimension and the current time.

[0086] In this embodiment, after the electronic device determines the user identification corresponding to the user in each identification dimension, it can determine at least one candidate resource type corresponding to the user according to the user identification corresponding to the user in each identification dimension and the current time. Among them, different resource types can be used to determine the usage method of virtual resources. For example, when the resource type corresponding to a certain virtual resource is a discount type, the user can use the virtual resource for electronic payment according to a preset ratio. Among the above, the resource type can also be used to define the usage scenario of virtual resources. For example, when the resource type corresponding to a certain virtual resource is a full reduction type, the user can use the virtual resource for electronic payment according to the reduction amount when the user's consumed resources meet the reduction conditions. For any user identification of the user, the electronic device can determine at least one candidate resource type of the user in the behavior data dimension according to the judgment condition corresponding to the identification dimension of the user identification. For the current time, the electronic device can determine the candidate resource type of the user in the time dimension according to the relationship between the current time and the holiday information. Then, the electronic device can determine all the candidate resource types in the behavior data dimension and all the candidate resource types in the time dimension as the candidate resource types corresponding to the user.

[0087] In a possible implementation, for the resource sensitivity dimension, if the user identifier of the user under this dimension is a high-sensitivity identifier, the electronic device may determine the discount type and / or the full reduction type as the candidate resource types of the user under the resource sensitivity dimension; if the user identifier of the user under this dimension is a low-sensitivity identifier, the electronic device may determine the fixed amount type as the candidate resource types of the user under the resource sensitivity dimension. Among them, the candidate resource quantity corresponding to the discount type may represent a reduction ratio value, and the virtual resource of the discount type may reduce the resources that the user needs to consume when obtaining the usage object according to a preset ratio (resource quantity). Among them, the candidate resource quantity corresponding to the virtual resource of the full reduction type represents a reduction condition and a reduction amount, and the virtual resource of the full reduction type may reduce the resource quantity that the user needs to consume according to the reduction amount when the user's consumed resources meet the resource condition. The candidate resource quantity corresponding to the fixed amount type may represent a reduction value, and the virtual resource of the fixed amount type may reduce the resources that the user needs to consume according to the reduction value when the user obtains the usage object.

[0088] For the activity dimension, if the user identifier of the user under this dimension is a high-activity identifier, the electronic device may determine the candidate resource type of the user under the activity dimension as the integral type; if the user identifier of the user under this dimension is a low-activity identifier, the electronic device may determine the candidate resource type of the user under the activity dimension as the discount type including the applicable time and / or the fixed amount type including the usage times condition. The candidate resource quantity corresponding to the integral type may represent an integral value, and the virtual resource of the fixed amount type may convert the user's consumed resources into the integral corresponding to the integral value when the user obtains the usage object.

[0089] For the user preference dimension, if the user has any preference identifier, the electronic device may determine the discount type including the applicable object and / or the preferential type including the applicable object as the candidate resource types corresponding to the user; if the user does not have a preference identifier, the electronic device may not generate the candidate resource type of the user under this identifier dimension. Exemplarily, when the platform is an e-commerce platform, if the user has a preference identifier corresponding to a certain category of goods or a certain brand, the electronic device may determine the discount type applicable to the category of goods or the brand as the candidate resource type.

[0090] In a possible implementation, if the distance value between the current time and any holiday in the holiday information is less than or equal to a preset distance threshold, the electronic device may determine the discount type including the applicable time as the candidate resource type; if the distance values between the current time and all holidays in the holiday information are greater than the distance threshold, the electronic device may generate the candidate resource type corresponding to the current time. For example, when the platform is an e-commerce platform, the electronic device may determine the discount type applicable during the Spring Festival as the candidate resource type corresponding to the user during the Spring Festival.

[0091] S1023. For any candidate resource type, determine at least one candidate resource quantity corresponding to any candidate resource type according to the expected consumed resources corresponding to the expected usage object, the average consumed resources corresponding to the expected usage object, the operation information of at least one pushed resource corresponding to the user, and the preset range of any candidate resource type.

[0092] In this embodiment, after the electronic device determines all candidate resource types corresponding to the user, for any one of the determined candidate resource types, the electronic device may determine at least one candidate resource quantity corresponding to the candidate resource type according to the expected consumed resources corresponding to the expected usage object in the demand prediction information of the user, the average consumed resources of the expected usage object in the platform, the operation information of the user for the pushed resources, and the preset range corresponding to the candidate resource type. Exemplarily, when the platform is an e-commerce platform and the candidate resource type is the discount type, the candidate resource quantity determined by the electronic device may be multiple different discount amounts; when the candidate resource type is the full reduction type, the candidate resource quantity determined by the electronic device may be multiple different full reduction amounts; when the candidate resource type is the fixed amount type, the candidate resource quantity determined by the electronic device may be multiple different exemption amounts.

[0093] S1024. Determine the candidate resources corresponding to the user according to the candidate resource type and the candidate resource quantity.

[0094] In this embodiment, after the electronic device determines the candidate resource type and the candidate resource quantity, it may determine at least one candidate resource corresponding to the user according to each candidate resource type and the candidate resource quantity corresponding to each candidate resource type respectively.

[0095] Through the method provided in this embodiment, since the electronic device can determine the candidate resource type according to the user identifier of the user in different dimensions, and determine the candidate resource quantity according to the operation information of the user for the pushed resources and the demand prediction information, therefore, the method provided in this embodiment can improve the accuracy of the candidate resources determined by the electronic device.

[0096] Figure 3 The specific implementation flowchart of a virtual resource pushing method S103 provided in the third embodiment of the present application is shown. Refer to Figure 3 , compared with Figure 1 the above embodiment, S103 in a virtual resource pushing method provided in this embodiment includes: S1031 to S1032, which are specifically described in detail as follows:

[0097] S1031. Input at least one pushed resource corresponding to the user and the operation information corresponding to the pushed resource into a preset objective function to determine the weight coefficient corresponding to each candidate resource.

[0098] In this embodiment, after the electronic device determines all candidate resources corresponding to the user, it may input at least one pushed resource corresponding to the user and the operation information corresponding to the pushed resource into a preset objective function to determine the weight coefficients corresponding to each candidate resource. The objective function is used to determine the weight coefficient corresponding to the candidate resource according to the operation information of the user for the pushed resource associated with the candidate resource. The candidate resource may be determined by the candidate resource type and the candidate resource quantity. The objective function may be used to determine the weight coefficients of the candidate resource type and the candidate data quantity corresponding to the candidate resource type respectively according to the operation information of the user for the pushed resource corresponding to the candidate resource type.

[0099] In a possible implementation manner, the objective function may include a conversion rate objective function, a retention rate objective function, and a resource objective function. The operation information of the user for a certain pushed resource may include the receiving status of the pushed resource, the usage status of the pushed resource, the browsing information of the user for the applicable object associated with the pushed resource, and the first transaction information related to the pushed resource in all the transaction information of the user. The browsing information of the user for a certain applicable object may include, but is not limited to, the browsing frequency and browsing duration of the user for a certain applicable object.

[0100] For any candidate resource type, the electronic device may input the usage status of the pushed resource corresponding to the candidate resource type into the conversion rate objective function to determine the first weight coefficient corresponding to the candidate resource type and the second weight coefficients corresponding to each candidate resource quantity of the candidate resource type. Specifically, for a certain candidate resource type, the conversion rate objective function may determine the usage rate corresponding to the candidate resource type according to the usage status of each pushed resource with the same resource type as the candidate resource type. Then, the conversion rate objective function may determine the first weight corresponding to the candidate resource type according to the difference between the usage rate corresponding to the candidate resource type and a preset first conversion rate threshold. Similarly, for a certain candidate resource quantity, the conversion rate objective function may determine the usage rate corresponding to the candidate resource quantity according to the usage status of each pushed resource with the same resource quantity as the candidate resource quantity. Then, the conversion rate objective function may determine the second weight corresponding to the candidate resource quantity according to the difference between the usage rate corresponding to the candidate resource quantity and a preset second conversion rate threshold.

[0101] For any candidate resource type, the electronic device may input the receiving status and browsing information of the pushed resource corresponding to the candidate resource type into the retention rate objective function to determine the third weight coefficient of the candidate resource type and the fourth weight coefficients corresponding to each candidate resource quantity of the candidate resource type through the retention rate objective function.

[0102] Specifically, for a certain candidate resource type, the retention rate objective function can determine the redemption rate corresponding to the candidate resource type based on the redemption status of each pushed resource with the same resource type as the candidate resource type; and determine the view volume corresponding to the candidate resource type based on the viewing information of each pushed resource with the same resource type as the candidate resource type. Then, the retention rate objective function can determine the third weight corresponding to the candidate resource type based on the difference between the redemption rate corresponding to the candidate resource type and a preset first redemption rate threshold, and the difference between the view volume and a preset first view volume threshold. Similarly, for a certain candidate resource quantity, the retention rate objective function can determine the redemption rate corresponding to the candidate resource quantity based on the redemption status of each pushed resource with the same resource quantity as the candidate resource quantity; and determine the view volume corresponding to the resource quantity based on the viewing information of each pushed resource with the same resource quantity as the candidate resource quantity. Then, the retention rate objective function can determine the fourth weight corresponding to the candidate resource type based on the difference between the redemption rate corresponding to the candidate resource type and a preset second redemption rate threshold, and the difference between the view volume and a preset second view volume threshold.

[0103] For any candidate resource type, the electronic device can input the historical consumed resources corresponding to each usage object and the average consumed resources corresponding to the usage object in the first transaction information corresponding to the any candidate resource type into the resource objective function, so as to determine the fifth weight coefficient of the any candidate resource type and the sixth weight coefficient of each candidate resource quantity corresponding to the any candidate resource type through the resource objective function. Among them, the historical consumed resources can be the amount of resources consumed by the user to obtain the usage object. The average consumed resources can be the average amount of resources consumed by all users on the platform to obtain the usage object. For a certain any candidate resource type, the electronic device can input the usage status of the pushed resources corresponding to the any candidate resource type into the conversion rate objective function to determine the first weight coefficient of the any candidate resource type and the second weight coefficient of each candidate resource quantity corresponding to the any candidate resource type.

[0104] Specifically, for a certain candidate resource type, the conversion rate objective function can determine the fifth weight corresponding to the candidate resource type based on the difference between the historical consumed resources of each usage object and the average consumed resources corresponding to the usage object in the first transaction information corresponding to the candidate resource type. Similarly, for a certain candidate resource quantity, the resource objective function can determine the sixth weight corresponding to the candidate resource quantity based on the difference between the historical consumed resources of each usage object and the average consumed resources corresponding to the usage object in the first transaction information corresponding to the candidate resource quantity.

[0105] S1032. Input the weight coefficients corresponding to all candidate resources into a preset policy function to determine at least one virtual resource.

[0106] In this embodiment, the electronic device may input the weight coefficients corresponding to all candidate resources, the third transaction data within a preset time period, and the user's historical online time periods into a preset policy function to determine at least one virtual resource. The policy function may be used to determine at least one virtual resource corresponding to the user from all candidate resources according to the user's status information, such as the third transaction data of the user within a preset time period and the user's historical online time periods, as well as the weight coefficients of the candidate resources.

[0107] Specifically, the electronic device may input the first weight coefficient, the third weight coefficient, the fifth weight coefficient, the third transaction data within a preset time period, and the user's historical online time periods into the first policy function to determine the target resource type from the candidate resource types of the first policy function. Then, the electronic device may input the second weight coefficient, the fourth weight coefficient, the sixth weight coefficient, the third transaction data within a preset time period, and the user's historical online time periods into the second policy function to determine the target resource quantity from all candidate resource quantities corresponding to the target resource type through the second policy function. Finally, the electronic device may determine the virtual resource corresponding to the user according to the target resource type and the target resource quantity.

[0108] Through the method provided in this embodiment, since the electronic device can continuously adjust the weight coefficients of the candidate resources according to the platform's goals (such as increasing the conversion rate, raising the average order value, reducing the churn rate, etc.) and the user's operation information on the pushed resources, that is, the user's feedback on the pushed resources, the method provided in this embodiment can form a closed-loop feedback mechanism to dynamically optimize the policy algorithm in real time. It can be seen that the method provided in this embodiment can improve the accuracy of the policy algorithm, and further improve the accuracy of the virtual resources determined by the policy algorithm.

[0109] See Figure 4 , which shows a schematic diagram of the push process of coupons on an e-commerce platform provided by an embodiment of the present application. As Figure 4As shown, after the electronic device obtains the user's behavior data, it can input the behavior data into the object prediction algorithm, the usage probability prediction algorithm, and the resource consumption prediction algorithm respectively to predict the user's needs, and generate the user's expected usage object (expected consumption category), the expected usage probability of the user for virtual resources, and the expected resource consumption (expected consumption amount) of the user for the expected usage object. The electronic device can determine at least one candidate resource type according to the current time and the user identification of the user. Specifically, the electronic device can determine at least one candidate resource type according to the user identification of the user in the object preference dimension, the user identification of the user in the resource sensitivity dimension, the user identification of the user in the activity dimension, and the relationship between the current time and the holiday information. After the electronic device determines the candidate resource type, it can determine at least one candidate resource amount corresponding to each candidate resource type according to the operation information of the user for the pushed resources, and the expected usage object (expected consumption category) and the expected resource consumption (expected consumption amount) of the user for the expected usage object in the demand prediction information. Then, the electronic device can input the candidate resources into the decision algorithm, and determine at least one virtual resource from all the candidate resources through the decision algorithm. Specifically, the policy algorithm can continuously update the weights of each candidate resource according to the operation information of the user for the pushed resources and multiple objective functions to select the virtual resource that can maximize the objective.

[0110] It should be noted that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution. The order of execution of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0111] Referring to Figure 5 , a schematic diagram of a virtual resource push device provided by an embodiment of the present application is shown, which may specifically include a demand prediction module 501, a candidate resource determination module 502, a virtual resource determination module 503, and a push module 504, where:

[0112] The demand prediction module 501 is configured to input the behavior data corresponding to the user to be pushed into a preset prediction algorithm to generate the demand prediction information corresponding to the user;

[0113] The candidate resource determination module 502 is configured to determine at least one candidate resource based on the demand prediction information and the behavior data;

[0114] The virtual resource determination module 503 is configured to input all the candidate resources into a preset decision algorithm to determine at least one virtual resource; the decision algorithm is configured to determine the virtual resource from all the candidate resources according to the operation information of at least one pushed resource corresponding to the user;

[0115] A push module 504, configured to push the virtual resources to the user.

[0116] The candidate resource determination module may further be configured to input the behavior data into a feature extraction algorithm to determine user identifiers corresponding to the user in different identification dimensions; determine at least one candidate resource type corresponding to the user according to the user identifiers corresponding to the respective identification dimensions and the current time; use the user identifier to determine, according to a judgment condition corresponding to the user identifier, one candidate resource type corresponding to the user; use the current time to determine, according to a relationship with holiday information, one candidate resource type corresponding to the user; for any candidate resource type, determine at least one candidate resource quantity corresponding to the any candidate resource type according to the expected consumed resources corresponding to the expected usage object, the average consumed resources corresponding to the expected usage object, operation information of at least one pushed resource corresponding to the user, and a preset range of the any candidate resource type; and determine the candidate resources corresponding to the user according to the candidate resource type and the candidate resource quantity.

[0117] The virtual resource determination module may further be configured to input at least one pushed resource corresponding to the user and operation information corresponding to the pushed resource into a preset objective function to determine weight coefficients corresponding to the respective candidate resources; the objective function is configured to determine the weight coefficients corresponding to the candidate resources according to the operation information of the pushed resources corresponding to the candidate resources; and input the weight coefficients corresponding to all the candidate resources into a preset policy function to determine the at least one virtual resource.

[0118] The virtual resource determination module can also be used to input the usage status of the pushed resources corresponding to any candidate resource type into the conversion rate objective function for any candidate resource type, to determine the first weight coefficient corresponding to any candidate resource type and the second weight coefficient of each candidate resource quantity corresponding to any candidate resource type; for any candidate resource type, to input the receiving status of the pushed resources corresponding to any candidate resource type and the browsing information into the retention rate objective function, to determine the third weight coefficient corresponding to any candidate resource type and the fourth weight coefficient of each candidate resource quantity corresponding to any candidate resource type; for any candidate resource type, to input the historical consumed resources corresponding to each usage object in the first transaction information corresponding to any candidate resource type and the average consumed resources corresponding to the usage object into the resource objective function, to determine the fifth weight coefficient corresponding to any candidate resource type and the sixth weight coefficient of each candidate resource quantity corresponding to any candidate resource type; to input the first weight coefficient, the third weight coefficient, and the fifth weight coefficient into the first policy function to determine the target resource type from the candidate resource types; to input the second weight coefficient, the fourth weight coefficient, and the sixth weight coefficient into the second policy function to determine the target resource quantity from all candidate resource quantities corresponding to the target resource type; and to determine the virtual resource according to the target resource type and the target resource quantity.

[0119] The candidate resource determination module can also be used to determine at least one resource push period corresponding to the user according to the expected usage frequency and holiday information if the expected usage probability is greater than a preset first probability threshold; and to determine at least one candidate resource based on the demand prediction information and the behavior data during the resource push period.

[0120] The push module can also be used to determine the push time corresponding to the virtual resource according to the resource push period, the user's historical online period, and the holiday information; to determine the applicable range corresponding to the virtual resource based on the demand prediction information, the user's transaction information, and the push time; and to push the virtual resource recording the applicable range to any user at the push time.

[0121] The push module can also be used to determine any desired usage object for which the desired transaction probability is greater than a preset second probability threshold as the applicable object for the virtual resource; obtain second transaction information of the user regarding the applicable object; determine the applicable time corresponding to the virtual resource according to the timestamps in each of the second transaction information, the usage frequency of the user for the pushed resources, the user's historical online periods, and the holiday information; and push the virtual resource recording the applicable object and the applicable time to any user at the push time.

[0122] The behavioral data in the demand prediction module includes at least one of the user's transaction information, historical online periods, login frequency, registration duration, browsing information regarding the usage object, and operation information regarding the pushed resources.

[0123] For the device embodiments, since they are basically similar to the method embodiments, the description is relatively simple. For the relevant parts, refer to the description in the method embodiments section.

[0124] Refer to Figure 6 , which shows a schematic diagram of an electronic device provided by an embodiment of the present application. As Figure 6 shown, the electronic device 600 in the embodiment of the present application includes: a processor 610, a memory 620, and a computer program 621 stored in the memory 620 and executable on the processor 610. When the processor 610 executes the computer program 621, it implements the steps in each of the above embodiments of the push method for virtual resources, such as Figure 1 the steps S101 to S104 shown. Alternatively, when the processor 610 executes the computer program 621, it implements the functions of each module / unit in each of the above device embodiments, such as Figure 3 the functions of the modules 301 to 305 shown.

[0125] Exemplarily, the computer program 621 can be divided into one or more modules / units. The one or more modules / units are stored in the memory 620 and executed by the processor 610 to complete the present application. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, and these instruction segments can be used to describe the execution process of the computer program 621 in the electronic device 600. For example, the computer program 621 can be divided into a demand prediction module, a candidate resource determination module, a virtual resource determination module, and a push module. The specific functions of each module are as follows:

[0126] The demand prediction module is used to input the behavioral data corresponding to the user to be pushed into a preset prediction algorithm to generate demand prediction information corresponding to the user;

[0127] A candidate resource determination module, configured to determine at least one candidate resource based on the demand prediction information and the behavior data;

[0128] A virtual resource determination module, configured to input all the candidate resources into a preset decision algorithm to determine at least one virtual resource; the decision algorithm is used to determine the virtual resource from all the candidate resources according to the operation information of at least one pushed resource corresponding to the user;

[0129] A push module, configured to push the virtual resource to the user.

[0130] The electronic device 600 may be a computing device such as a desktop computer or a cloud server. The electronic device 600 may include, but is not limited to, a processor 610 and a memory 620. Those skilled in the art can understand that Figure 6 This is only an example of the electronic device 600 and does not constitute a limitation on the electronic device 600. It may include more or fewer components than shown in the figure, or combine certain components, or different components. For example, the electronic device 600 may further include input / output devices, network access devices, buses, etc.

[0131] The processor 610 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.

[0132] The memory 620 may be an internal storage unit of the electronic device 600, such as a hard disk or memory of the electronic device 600. The memory 620 may also be an external storage device of the electronic device 600, such as a plug-in hard disk equipped on the electronic device 600, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, and so on. Further, the memory 620 may also include both the internal storage unit of the electronic device 600 and an external storage device. The memory 620 is used to store the computer program 621 and other programs and data required by the electronic device 600. The memory 620 may also be used to temporarily store data that has been output or is to be output.

[0133] An embodiment of the present application also discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for pushing virtual resources as described in the foregoing various embodiments is implemented.

[0134] An embodiment of the present application also discloses a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the method for pushing virtual resources as described in the foregoing various embodiments is implemented.

[0135] An embodiment of the present application also discloses a computer program product, and when the computer program product runs on a computer, the computer is caused to execute the method for pushing virtual resources as described in the foregoing various embodiments.

[0136] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included within the protection scope of the present application.

Claims

1. A method for pushing virtual resources, characterized in that: include: Inputting the behavior data corresponding to the user to be pushed into a preset prediction algorithm to generate demand prediction information corresponding to the user; determining at least one candidate resource based on the demand forecast information and the behavior data; Inputting all the candidate resources into a preset decision algorithm to determine at least one virtual resource; The decision algorithm is used to determine the virtual resource from all the candidate resources according to the operation information of at least one pushed resource corresponding to the user; The virtual resource is pushed to the user.

2. The method according to claim 1, characterized in that The demand forecast information includes expected usage objects and expected consumption resources corresponding to the expected usage objects; The determining at least one candidate resource based on the demand forecast information and the behavior data includes: Inputting the behavior data into a feature extraction algorithm to determine the user identifiers corresponding to the users in different identifier dimensions; Determine at least one candidate resource type corresponding to the user according to the user identifier corresponding to each of the identification dimensions and the current time; the user identifier is used to determine a candidate resource type corresponding to the user according to a judgment condition corresponding to the user identifier; the current time is used to determine a candidate resource type corresponding to the user in relation to holiday information; For any candidate resource type, determine at least one candidate resource amount corresponding to any candidate resource type according to the expected consumption resources corresponding to the expected usage object, the average consumption resources corresponding to the expected usage object, the operation information of at least one pushed resource corresponding to the user, and the preset range of any candidate resource type; The candidate resource corresponding to the user is determined according to the candidate resource type and the candidate resource amount.

3. The method according to claim 2, characterized in that The step of inputting all the candidate resources into a preset decision algorithm to determine at least one virtual resource includes: Inputting at least one pushed resource corresponding to the user and operation information corresponding to the pushed resource into a preset objective function to determine the weight coefficient corresponding to each of the candidate resources; the objective function is used to determine the weight coefficient corresponding to the candidate resource according to the operation information of the pushed resource corresponding to the candidate resource; The weight coefficients corresponding to all the candidate resources are input into a preset policy function to determine the at least one virtual resource.

4. The method according to claim 3, characterized in that The objective function includes a conversion rate objective function, a retention rate objective function and a resource objective function; the operation information of the pushed resource includes the collection status of the pushed resource, the usage status of the pushed resource, the browsing information of the user for the applicable object corresponding to the pushed resource and the first transaction information of the user regarding the pushed resource; The step of inputting the operation information of at least one pushed resource corresponding to the user into a preset objective function to determine the weight coefficient corresponding to each virtual resource includes: For any candidate resource type, inputting the usage status of the pushed resource corresponding to any candidate resource type into the conversion rate objective function, and determining a first weight coefficient corresponding to any candidate resource type and a second weight coefficient of each candidate resource amount corresponding to any candidate resource type; For any candidate resource type, input the receipt status of the pushed resource corresponding to any candidate resource type and the browsing information into the retention rate objective function, and determine a third weight coefficient of any candidate resource type and a fourth weight coefficient of each candidate resource amount corresponding to any candidate resource type; For any candidate resource type, input the historical consumption resources corresponding to each usage object in the first transaction information corresponding to the any candidate resource type and the average consumption resources corresponding to the usage object into the resource objective function, and determine the fifth weight coefficient of the any candidate resource type and the sixth weight coefficient of each candidate resource amount corresponding to the any candidate resource type; Accordingly, inputting the weight coefficients corresponding to all the virtual resources into a preset policy function to determine the at least one virtual resource includes: Inputting the first weight coefficient, the third weight coefficient and the fifth weight coefficient into a first strategy function, and determining the target resource type from the candidate resource types; Inputting the second weight coefficient, the fourth weight coefficient and the sixth weight coefficient into a second strategy function, and determining a target resource amount from all the candidate resource amounts corresponding to the target resource type; The virtual resource is determined according to the target resource type and the target resource amount.

5. The method according to any one of claims 1 to 4, characterized in that: The demand forecast information includes the user's expected usage probability and expected usage frequency for virtual resources: The determining at least one candidate resource based on the demand forecast information and the behavior data includes: If the expected usage probability is greater than a preset first probability threshold, determining at least one resource push time period corresponding to the user according to the expected usage frequency and holiday information; During the resource push period, at least one candidate resource is determined based on the demand forecast information and the behavior data.

6. The method according to claim 5, characterized in that The pushing the virtual resource to the user comprises: Determine the push time corresponding to the virtual resource according to the resource push time period, the user's historical online time period and the holiday information; Determine the applicable scope of the virtual resource based on the demand forecast information, the transaction information of the user and the push time; At the push time, the virtual resource having the applicable scope recorded therein is pushed to any one of the users.

7. The method according to claim 6, characterized in that The demand forecast information includes expected usage objects and expected transaction probabilities corresponding to each of the expected usage objects; the applicable scope includes applicable time and applicable objects; The pushing of the virtual resource having the applicable scope recorded therein to any one of the users at the pushing time includes: If the expected transaction probability of any expected use object is greater than a preset second probability threshold, determining the any expected use object as a suitable object for the virtual resource; Acquiring second transaction information of the user regarding the applicable object; Determine the applicable time corresponding to the virtual resource according to the timestamp in each of the second transaction information, the frequency of use of the pushed resource by the user, the historical online time period of the user, and the holiday information; At the push time, the virtual resource recording the applicable object and the applicable time is pushed to any one of the users.

8. The method according to any one of claims 1 to 4, characterized in that: The behavior data includes at least one of the user's transaction information, historical online time periods, login frequency, registration duration, browsing information about usage objects, and operation information about pushed resources.

9. An electronic device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the electronic device implements the method for pushing virtual resources as described in any one of claims 1 to 8.

10. A computer program product, characterized in that The invention comprises a computer program, and when the computer program is executed, the method for pushing virtual resources according to any one of claims 1 to 8 is executed.

Images