A method and apparatus for information push

Abstract

The specification discloses a method and device for information pushing. First, historical data of a merchant to which to-be-recommended information belongs is acquired. Second, a type of business demand corresponding to the merchant is determined as a target type according to the historical data. Third, a business model matched with the target type is determined as a target model from pre-trained business models. Fourth, the historical data is input into the target model to determine an amount of resources consumed by the merchant when the to-be-recommended information is pushed. Finally, the to-be-recommended information is pushed according to the amount of resources. The method determines the amount of resources consumed by the merchant when the business demand of the merchant is met according to the business demand corresponding to different merchants, thereby improving the information pushing effect of the merchant.

Description

Technical Field

[0001] This specification relates to the field of computer technology, and in particular to a method and apparatus for information push. Background Technology

[0002] In practical applications, the same model is typically used to determine the amount of resources consumed by different merchants when pushing advertisements. However, since different merchants have different business needs, some merchants using this model may consume more resources, thereby reducing the effectiveness of their information push.

[0003] Therefore, determining the appropriate amount of resources consumed by a merchant is an urgent problem to be solved. Summary of the Invention

[0004] This specification provides a method, apparatus, storage medium, and electronic device for information push, in order to partially solve the aforementioned problems existing in the prior art.

[0005] The following technical solution is adopted in this specification:

[0006] This manual provides a method for information push, including:

[0007] Obtain historical data of the merchants whose information is to be recommended;

[0008] Based on the historical data, the type of business need corresponding to the merchant is determined as the target type;

[0009] From the pre-trained business models, identify the business model that matches the target type and use it as the target model;

[0010] The historical data is input into the target model to determine the amount of resources consumed by the merchant when pushing the recommended information;

[0011] Based on the amount of resources, the recommended information is pushed out.

[0012] Optionally, before inputting the historical data into the target model, the method further includes:

[0013] Obtain the merchant's historical data across various information display channels and the merchant's total resources;

[0014] Based on the merchant's historical data across various information display channels, determine the merchant's resource allocation coefficient across each information display channel;

[0015] Based on the resource allocation coefficient of the merchant under each information display channel, the total amount of resources corresponding to the merchant is allocated to obtain the amount of resources of the merchant under each information display channel.

[0016] Optionally, the target model is trained, specifically including:

[0017] Acquire historical data and auxiliary data for each merchant, including: the predicted click-through rate and the predicted conversion rate for each merchant.

[0018] The auxiliary data corresponding to the merchant and the historical data of the merchant corresponding to the target type are input into the target model. The target model determines the amount of resources consumed by the merchant under the premise of maximizing the information push revenue of the merchant, which is the first amount of resources. The target model also determines the amount of resources consumed by the merchant after the user executes the historical recommendation information service, which is the second amount of resources.

[0019] Based on the first resource quantity and the second resource quantity, the amount of resources consumed by the merchant when pushing the historical recommendation information is determined and used as the target resource quantity;

[0020] Based on the information push revenue corresponding to the merchant and the target resource quantity, a target reward value is determined, and the target model is trained to maximize the target reward value.

[0021] Optionally, before determining the amount of resources consumed by the merchant when pushing the historical recommendation information based on the first resource amount and the second resource amount, and using this as the target resource amount, the method further includes:

[0022] The amount of resources that merchants expect to consume when obtaining historical recommendation information is considered as a third resource.

[0023] The historical data, auxiliary data, and the third resource quantity corresponding to the merchant are input into the target model to adjust the second resource quantity and determine the reference resource quantity;

[0024] Based on the first resource quantity and the second resource quantity, the resource quantity consumed by the merchant when pushing the historical recommendation information is determined as the target resource quantity, specifically including:

[0025] Based on the first resource quantity and the reference resource quantity, the resource quantity consumed by the merchant when pushing the historical recommendation information is determined and used as the target resource quantity.

[0026] Optionally, the target model includes: a weighted sub-model;

[0027] Based on the first resource quantity and the reference resource quantity, the resource quantity consumed by the merchant when pushing the historical recommendation information is determined as the target resource quantity, specifically including:

[0028] The historical data corresponding to the merchant is input into the weight sub-model so that the weight sub-model can determine the weight of the merchant under the premise of maximizing the information push revenue of the merchant and maximizing the amount of resources consumed by the merchant.

[0029] Based on the first resource quantity, the reference resource quantity, and the weight corresponding to the merchant, the resource quantity consumed by the merchant when pushing the historical recommendation information is determined and used as the target resource quantity.

[0030] Optionally, the target reward value is determined based on the information push revenue corresponding to the merchant and the target resource volume, specifically including:

[0031] The first reward value is determined based on the target type corresponding to the merchant and the information push revenue of the merchant to which the historical recommendation information belongs;

[0032] The second reward value is determined based on the target resource quantity;

[0033] The target reward value is determined based on the first reward value and the second reward value.

[0034] Optionally, based on the amount of resources, the recommended information can be pushed, specifically including:

[0035] The amount of resources a merchant expects to consume when retrieving historical recommendation information;

[0036] The information to be recommended is determined based on the amount of resources consumed by the merchant when pushing the information to be recommended, as determined by the target model, and the amount of resources the merchant expects to consume when pushing historical recommendation information.

[0037] This specification provides an information push device, including:

[0038] The acquisition module is used to obtain historical data of the merchants to which the information to be recommended belongs;

[0039] The determination module is used to determine the type of business needs corresponding to the merchant based on the historical data, and use it as the target type;

[0040] The matching module is used to determine the business model that matches the target type from the pre-trained business models, and use it as the target model;

[0041] An input module is used to input the historical data into the target model to determine the amount of resources consumed by the merchant when pushing the information to be recommended.

[0042] The push module is used to push the recommended information based on the amount of resources.

[0043] This specification provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned information push method.

[0044] This specification provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the aforementioned information push method.

[0045] The above-mentioned technical solutions adopted in this specification can achieve the following beneficial effects:

[0046] The information push method provided in this manual first obtains historical data of the merchant to which the information to be recommended belongs. Second, based on the historical data, the type of business need corresponding to the merchant is determined as the target type. Then, from a pre-trained set of business models, a business model matching the target type is selected as the target model. Next, the historical data is input into the target model to determine the amount of resources consumed by the merchant when pushing the information to be recommended. Finally, based on the resource consumption, the information to be recommended is pushed.

[0047] As can be seen from the above method, this approach can identify the business model that matches the merchant's specific business needs from pre-trained business models and use it as the target model. Historical data is then input into the target model to determine the amount of resources consumed by the merchant when pushing recommended information. Compared to existing technologies, this method determines the amount of resources consumed by a merchant while meeting their specific business needs, thereby improving the effectiveness of information push for that merchant. Attached Figure Description

[0048] The accompanying drawings, which are included to provide a further understanding of this specification and form part of this specification, illustrate exemplary embodiments and are used to explain this specification, but do not constitute an undue limitation thereof. In the drawings:

[0049] Figure 1 A flowchart illustrating the information push method provided in the embodiments of this specification;

[0050] Figure 2 This is a flowchart illustrating the process of determining the target resource quantity provided in the embodiments of this specification;

[0051] Figure 3 This is a schematic diagram of the structure of the information push device provided in the embodiments of this specification;

[0052] Figure 4 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this specification. Detailed Implementation

[0053] To make the objectives, technical solutions, and advantages of this specification clearer, the technical solutions of this specification will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this specification, and not all of them. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this specification.

[0054] The technical solutions provided in the various embodiments of this specification are described in detail below with reference to the accompanying drawings.

[0055] Figure 1 This is a flowchart illustrating the information push method provided in the embodiments of this specification, which specifically includes the following steps:

[0056] S100: Obtain historical data of the merchant to which the information to be recommended belongs.

[0057] In the embodiments described in this specification, the execution entity of the information push method can be a server or a terminal device such as a desktop computer. For ease of description, the information push method provided in this specification will be described below using only a server as the execution entity.

[0058] In the embodiments of this specification, the server can obtain historical data of the merchant to which the information to be recommended belongs. The information to be recommended here can be advertising information displayed to the user while the user is browsing information, or advertising information displayed to the user when the user opens a client or application (App). The historical data of the merchant includes: the merchant's click-through rate (CTR), conversion rate (CVR), return on investment (ROI), gross merchandise volume (GMV), and total resources over a past period.

[0059] The historical data for each merchant mentioned here also includes historical average data for each merchant. For example, the average amount of resources consumed by each merchant, the average click-through rate, and the average conversion rate.

[0060] The terminal device used by the user to browse information can be a mobile phone, tablet computer, or other terminal device. Of course, the entity that executes the process of obtaining advertisements can also be a client, application (App) installed on the terminal device, or a browser on the terminal device or client.

[0061] S102: Based on the historical data, determine the type of business needs corresponding to the merchant as the target type.

[0062] In practical applications, a merchant's historical data can reflect their corresponding business needs. For example, if a merchant has a low historical click-through rate but a high historical conversion rate, then their corresponding business need might be to improve the click-through rate. Based on this, the server can determine the merchant's corresponding business needs using their historical data.

[0063] In the embodiments described in this specification, the server can determine the type of business needs corresponding to the merchant based on historical data, and use this as the target type.

[0064] Specifically, the server can determine the type of business needs of a merchant based on historical data such as the merchant's historical click-through rate, historical conversion rate, historical return on investment, and total resources, and use this data as the target type.

[0065] S104: From the pre-trained business models, identify the business model that matches the target type and use it as the target model.

[0066] In practical applications, different merchants have different business needs. If the same model is used to determine the amount of resources consumed by different merchants when pushing ads, it may result in some merchants consuming more resources. Therefore, the server can determine a business model that matches the specific business needs of each merchant to improve the effectiveness of their information push.

[0067] In the embodiments of this specification, the server can determine the business model that matches the target type from the pre-trained business models and use it as the target model.

[0068] In practical applications, merchants whose information is to be recommended often push that information across multiple display channels simultaneously. Currently, merchants typically determine the resource allocation for each display channel manually. This method may result in unreasonable resource allocations for each channel, leading to lower message delivery effectiveness on some channels. Therefore, servers can allocate resources more rationally across different display channels based on the merchant's historical data.

[0069] In the embodiments described in this specification, the server can obtain historical data of the merchant across various information display channels and the merchant's total resources. The information display channels mentioned here can include: channels that display information to the user during the user's browsing process, channels that display information to the user when the user opens a client or application, or channels that display information to the user after the user searches for keywords.

[0070] Secondly, the server can determine the resource allocation coefficient for each information display channel based on the merchant's historical data across those channels.

[0071] Specifically, the server can input the merchant's historical data across various information display channels into the resource allocation model to determine the merchant's resource allocation coefficient across each information display channel.

[0072] Finally, the server can allocate the merchant's total resources across each information display channel based on the merchant's resource allocation coefficients. The specific formula is as follows:

[0073] bugget ij =γ ij ·bugget i

[0074] In the above formula, bugget ij This can be used to represent the resource quantity of the i-th merchant under the j-th information display channel. γ ij It can be used to represent resource allocation coefficients. (bugget) i It can be used to represent the total amount of resources of the i-th merchant.

[0075] γ ij =Agent b (f ij )

[0076] In the above formula, f ij It can be used to represent the historical data of the i-th merchant under the j-th information display channel. b () can be used to represent resource allocation models.

[0077]

[0078] As can be seen from the above formula, the resource allocation coefficient of the i-th merchant under the j-th information display channel needs to satisfy the sum of the resource allocation coefficients of all information display channels being 1.

[0079] S106: Input the historical data into the target model to determine the amount of resources consumed by the merchant when pushing the information to be recommended.

[0080] In the embodiments described in this specification, the server can input historical data into the target model to determine the amount of resources consumed by the merchant when pushing the recommended information.

[0081] In the embodiments of this specification, determining the amount of resources consumed by the merchant when pushing recommended information based on historical data requires relying on a pre-trained target model. The process of training the target model will be introduced below.

[0082] First, the server can obtain historical data and auxiliary data from each merchant. This auxiliary data includes the merchant's predicted click-through rate (CTR) and predicted conversion rate. It's important to note that the predicted CTR refers to the probability that a user will click on a service offered by the merchant after the service has been shown to them. Similarly, the predicted conversion rate refers to the probability that a user will actually use a service offered by the merchant after the service has been shown to them.

[0083] The server can retrieve the information content corresponding to historical recommendations. This includes the information type (food, travel, etc.), display format (videos, images, etc.), and quality information of the historical recommendations. The server can also retrieve user behavior preference data.

[0084] For example, during the process of browsing historical recommendations, users will generate various behavioral data. This data can be used to analyze user preferences while browsing historical recommendations. The acquired behavioral preference data may include: the type of information the user actually clicked on (food, travel, etc.), the page location clicked, the user's location and time period, and the duration of browsing the clicked historical recommendations. Through these methods, the server can determine certain preference characteristics reflected by the user while browsing historical recommendations.

[0085] The server can determine auxiliary data based on the information content corresponding to historical recommendations and user behavior preference data.

[0086] It should be noted that auxiliary features can be calculated in real time by the target model or obtained from other models.

[0087] Secondly, the server can input the auxiliary data corresponding to the merchant and the historical data of the merchant corresponding to the target type into the target model, so as to determine the amount of resources consumed by the merchant under the premise of maximizing the information push revenue of the merchant, as the first resource amount, and determine the amount of resources consumed by the merchant after the user executes the business of historical recommendation information, as the second resource amount.

[0088] Then, the server can determine the amount of resources consumed by the merchant when pushing historical recommendation information based on the first resource amount and the second resource amount, and use this as the target resource amount.

[0089] Finally, the server can determine the target reward value based on the merchant's information push revenue and target resource quantity, and train the target model to maximize the target reward value.

[0090] In practical applications, some merchants determine the amount of resources consumed when pushing recommended information through target models, while others determine the amount of resources consumed when pushing recommended information on their own.

[0091] If, in determining the amount of resources consumed when pushing recommended information, only the merchants whose resource consumption is determined through the target model are considered, collusion may occur among these merchants. In other words, these merchants may all determine a lower amount of resources, which would result in a reduction in the amount of resources consumed by the merchants after the user who received the recommended information performs the recommended information's business.

[0092] Based on this, the server can adjust the amount of the second resource according to the historical data and auxiliary data of each merchant, thereby improving the information push effect of each merchant.

[0093] In the embodiments described in this specification, the server can obtain the amount of resources that the merchant expects to consume when pushing historical recommendation information, as a third resource amount. The amount of resources that the merchant expects to consume when pushing historical recommendation information mentioned here can refer to the amount of resources that the merchant determines themselves will consume when pushing historical recommendation information.

[0094] Secondly, the server can input the merchant's historical data, auxiliary data, and third resource quantity into the target model to adjust the second resource quantity and determine the reference resource quantity.

[0095] Specifically, the target model also includes a reference model. The server can input the merchant's historical data, auxiliary data, and third resource quantity into the reference model of the target model to adjust the second resource quantity and determine the reference resource quantity.

[0096] Finally, the server can determine the amount of resources consumed by the merchant when pushing historical recommendation information based on the first resource quantity and the reference resource quantity, and use this as the target resource quantity.

[0097] For example, the target model can compare the historical and auxiliary data of merchants who consume resources when pushing historical recommendations based on their own determination with the historical and auxiliary data of merchants who consume resources when pushing recommendations based on the target model's determination. If the predicted click-through rate and predicted conversion rate of merchants who consume resources when pushing historical recommendations based on their own determination are higher, the target model can reduce the second resource amount and determine the reference resource amount. This ensures that merchants who consume resources when pushing historical recommendations based on their own determination have the opportunity to push historical recommendations.

[0098] In the embodiments of this specification, the target model includes a weighted sub-model. To ensure that the target resource quantity improves the effectiveness of the merchant's information push while simultaneously increasing the resource consumption of the merchant after the user executes the historical recommendation information service, the server can input the merchant's historical data into the weighted sub-model. The weighted sub-model then determines the merchant's weight while maximizing both the merchant's information push revenue and the merchant's resource consumption.

[0099] Then, the server can determine the amount of resources consumed by the merchant when pushing the historical recommendation information, based on the first resource amount, the reference resource amount, and the merchant's corresponding weight, and use this as the target resource amount. The specific formula is as follows:

[0100] bid=λ·bid i +(1-λ)bid a

[0101] In the above formula, `bid` can be used to represent the target resource quantity. `in` can be used to represent the weight of the merchant determined by the weighted sub-model. i It can be used to represent the first resource quantity. (bid) a This can be used to represent the amount of reference resources. As can be seen from the formula above, the server can determine the weight of the merchant through the weighted sub-model, thereby improving the effectiveness of the merchant's information push while increasing the amount of resources consumed by the merchant after the user executes the historical recommendation information service. Specifically, as follows... Figure 2 As shown.

[0102] Figure 2 This is a flowchart illustrating the process of determining the target resource quantity as provided in the embodiments of this specification.

[0103] exist Figure 2In this process, the server can use historical data of merchants whose resource consumption is determined when pushing recommended information through the target model to categorize them into different business types using a clustering algorithm. Based on each merchant's business type, a corresponding business model is determined, and the historical data for each merchant is input into a weighted sub-model to determine the weights for each merchant. For each merchant, its historical data and auxiliary data are input into its corresponding business model to determine the initial resource quantity for that merchant, and the historical data and auxiliary data for each merchant are input into a reference model to determine the reference resource quantity. The server can then determine the target resource quantity based on the initial resource quantity, the reference resource quantity, and the weights.

[0104] Finally, the server can determine the target reward value based on the merchant's information push revenue and target resource quantity, and train the target model to maximize the target reward value.

[0105] In practical applications, the reward value for information push revenue is determined solely based on the revenue obtained by the merchant. This method of determining the reward value cannot meet the diverse business needs of different merchants. Therefore, the server can determine a primary reward value based on the information push revenue of merchants under different business needs.

[0106] In the embodiments of this specification, the server can determine the first reward value based on the target type corresponding to the merchant and the information push revenue of the merchant to which the recommended information belongs.

[0107] The higher the metric for the merchant's target type, the higher the first reward value. The higher the information push revenue of the merchant to whom the recommended information belongs, the higher the first reward value.

[0108] Secondly, the server can determine the second reward value based on the target resource quantity.

[0109] The higher the target resource quantity, the higher the second reward value.

[0110] Finally, the server can determine the target reward value based on the first reward value and the second reward value.

[0111] S108: Push the recommended information based on the amount of resources.

[0112] In the embodiments described in this specification, the server can push recommended information based on the amount of resources available.

[0113] In the embodiments described in this specification, the server can obtain the amount of resources that the merchant expects to consume when pushing historical recommendation information.

[0114] Secondly, the server can determine the information to be recommended based on the amount of resources consumed by the merchant when pushing the information to be recommended, as determined by the target model, and the amount of resources that the merchant expects to consume when pushing historical recommendation information.

[0115] It should be noted that this method uses a second-price sealed-bid auction, also known as a Vickrey auction. In a second-price sealed-bid auction, each merchant independently bids in a sealed manner, and the opportunity to push the information to the user is given to the merchant who consumes the highest amount of resources. However, the merchant to whom the recommended information belongs consumes the second highest amount of resources among all those consumed.

[0116] As can be seen from the above process, this method can identify the business model that matches the merchant's specific business needs from pre-trained business models and use it as the target model. Historical data is then input into the target model to determine the amount of resources consumed by the merchant when pushing recommended information. This method determines the amount of resources consumed by a merchant to meet their specific business needs, thereby improving the effectiveness of information delivery for that merchant.

[0117] The above describes one or more embodiments of the information push method provided in this specification. Based on the same idea, this specification also provides corresponding information push devices, such as... Figure 3 As shown.

[0118] Figure 3 This is a schematic diagram of the structure of the information push device provided in the embodiments of this specification, specifically including:

[0119] Module 300 is used to obtain historical data of the merchants to which the information to be recommended belongs.

[0120] The determining module 302 is used to determine the type of business demand corresponding to the merchant based on the historical data, as the target type;

[0121] The matching module 304 is used to determine, from the pre-trained business models, the business model that matches the target type, and use it as the target model;

[0122] The input module 306 is used to input the historical data into the target model to determine the amount of resources consumed by the merchant when pushing the information to be recommended;

[0123] The push module 308 is used to push the information to be recommended based on the amount of resources.

[0124] Optionally, the matching module 304 is further configured to: obtain the merchant's historical data and total resources under each information display channel; determine the merchant's resource allocation coefficient under each information display channel based on the merchant's historical data under each information display channel; and allocate the merchant's total resources under each information display channel based on the merchant's resource allocation coefficient under each information display channel, thereby obtaining the merchant's resource quantity under each information display channel.

[0125] Optionally, the matching module 304 is specifically used to acquire historical data and auxiliary data of each merchant, the auxiliary data including: the predicted click-through rate and the predicted conversion rate of the merchant, inputting the auxiliary data of the merchant and the historical data of the merchant corresponding to the target type into the target model, so as to determine the amount of resources consumed by the merchant under the premise of maximizing the information push revenue of the merchant, as the first resource amount, and the amount of resources consumed by the merchant after the user executes the historical recommendation information business, as the second resource amount, and based on the first resource amount and the second resource amount, determine the amount of resources consumed by the merchant when pushing the historical recommendation information, as the target resource amount, and determine the target reward value based on the information push revenue of the merchant and the target resource amount, and train the target model to maximize the target reward value.

[0126] Optionally, the matching module 304 is further configured to: obtain the amount of resources that the merchant expects to consume when pushing historical recommendation information, as a third resource amount; input the merchant's corresponding historical data, auxiliary data, and the third resource amount into the target model to adjust the second resource amount; determine a reference resource amount; and determine the amount of resources consumed by the merchant when pushing the historical recommendation information based on the first resource amount and the reference resource amount, as a target resource amount.

[0127] Optionally, the target model includes: a weighted sub-model;

[0128] The matching module 304 is specifically used to input the historical data corresponding to the merchant into the weight sub-model, so as to determine the weight corresponding to the merchant under the premise of maximizing the information push revenue of the merchant and maximizing the amount of resources consumed by the merchant through the weight sub-model, and determine the amount of resources consumed by the merchant when pushing the historical recommendation information based on the first amount of resources, the reference amount of resources and the weight corresponding to the merchant, as the target amount of resources.

[0129] Optionally, the matching module 304 is specifically used to determine a first reward value based on the target type corresponding to the merchant and the information push revenue of the merchant to which the historical recommendation information belongs, determine a second reward value based on the target resource quantity, and determine a target reward value based on the first reward value and the second reward value.

[0130] Optionally, the matching module 304 is specifically used to obtain the amount of resources that the merchant expects to consume when pushing historical recommendation information, and determine the information to be recommended based on the amount of resources consumed by the merchant when pushing the information to be recommended through the target model and the amount of resources that the merchant expects to consume when pushing historical recommendation information.

[0131] This specification also provides a computer-readable storage medium storing a computer program that can be used to execute the above-described... Figure 1 The method for providing information push.

[0132] This instruction manual also provides Figure 4 The diagram shows the structure of the electronic device. Figure 4 At the hardware level, the electronic device includes a processor, internal bus, network interface, memory, and non-volatile memory, and may also include other hardware required for the business operations. The processor reads the corresponding computer program from the non-volatile memory into memory and then runs it to achieve the above-mentioned functions. Figure 1 The method for pushing information described herein. Of course, in addition to software implementation, this specification does not exclude other implementation methods, such as logic devices or a combination of hardware and software, etc. In other words, the execution subject of the following processing flow is not limited to individual logic units, but can also be hardware or logic devices.

[0133] It should be noted that all actions involving the acquisition of signals, information, or data in this application are carried out in compliance with the relevant data protection laws and policies of the country where the application is located, and with the authorization granted by the owner of the relevant device.

[0134] In the 1990s, improvements to a technology could be clearly distinguished as either hardware improvements (e.g., improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the methodology). However, with technological advancements, many methodological improvements today can be considered direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved methodology into the hardware circuit. Therefore, it cannot be said that a methodological improvement cannot be implemented using hardware physical modules. For example, a Programmable Logic Device (PLD) (such as a Field Programmable Gate Array (FPGA)) is such an integrated circuit whose logic function is determined by the user programming the device. Designers can program and "integrate" a digital system onto a PLD themselves, without needing chip manufacturers to design and manufacture dedicated integrated circuit chips. Furthermore, nowadays, instead of manually manufacturing integrated circuit chips, this programming is mostly implemented using "logic compiler" software. Similar to the software compiler used in program development, the original code before compilation must be written in a specific programming language, called a Hardware Description Language (HDL). There are many HDLs, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, and RHDL (Ruby Hardware Description Language). Currently, the most commonly used are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should understand that by simply performing some logic programming on the method flow using one of these hardware description languages ​​and programming it into an integrated circuit, the hardware circuit implementing the logical method flow can be easily obtained.

[0135] The controller can be implemented in any suitable manner. For example, it can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicon Labs C8051F320. A memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also recognize that, in addition to implementing the controller in purely computer-readable program code form, the same functionality can be achieved by logically programming the method steps to make the controller take the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0136] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0137] For ease of description, the above devices are described in terms of function, divided into various units. Of course, in implementing this specification, the functions of each unit can be implemented in one or more software and / or hardware components.

[0138] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0139] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0140] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0141] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0142] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0143] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0144] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0145] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0146] Those skilled in the art will understand that the embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, this specification may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this specification may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0147] This specification can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This specification can also be practiced in distributed computing environments, where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0148] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0149] The above description is merely an embodiment of this specification and is not intended to limit this specification. Various modifications and variations can be made to this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this specification should be included within the scope of the claims of this specification.

Claims

1. A method for information push, characterized in that, include: Obtain historical data of the merchants whose information is to be recommended; Based on the historical data, the type of business need corresponding to the merchant is determined as the target type; From the pre-trained business models, identify the business model that matches the target type and use it as the target model; Obtain historical data and auxiliary data from each merchant, including: Predicted click-through rate and predicted conversion rate for the merchant; The auxiliary data corresponding to the merchant and the historical data of the merchant corresponding to the target type are input into the target model. The target model determines the amount of resources consumed by the merchant under the premise of maximizing the information push revenue of the merchant, which is the first amount of resources. The target model also determines the amount of resources consumed by the merchant after the user executes the historical recommendation information service, which is the second amount of resources. Based on the first resource quantity and the second resource quantity, the amount of resources consumed by the merchant when pushing the historical recommendation information is determined and used as the target resource quantity; Based on the information push revenue corresponding to the merchant and the target resource volume, a target reward value is determined, and the target model is trained to maximize the target reward value. The historical data is input into the target model to determine the amount of resources consumed by the merchant when pushing the recommended information; The recommended information is pushed based on the amount of resources consumed by the merchant.

2. The method as described in claim 1, characterized in that, Before inputting the historical data into the target model, the method further includes: Obtain the merchant's historical data across various information display channels and the merchant's total resources; Based on the merchant's historical data across various information display channels, determine the merchant's resource allocation coefficient across each information display channel; Based on the resource allocation coefficient of the merchant under each information display channel, the total amount of resources corresponding to the merchant is allocated to obtain the amount of resources of the merchant under each information display channel.

3. The method as described in claim 1, characterized in that, Before determining the amount of resources consumed by the merchant when pushing the historical recommendation information based on the first resource amount and the second resource amount, and using this as the target resource amount, the method further includes: The amount of resources that merchants expect to consume when obtaining historical recommendation information is considered as a third resource. The historical data, auxiliary data, and the third resource quantity corresponding to the merchant are input into the target model to adjust the second resource quantity and determine the reference resource quantity; Based on the first resource quantity and the second resource quantity, the resource quantity consumed by the merchant when pushing the historical recommendation information is determined as the target resource quantity, specifically including: Based on the first resource quantity and the reference resource quantity, the resource quantity consumed by the merchant when pushing the historical recommendation information is determined and used as the target resource quantity.

4. The method as described in claim 3, characterized in that, The target model includes: Weighted sub-model; Based on the first resource quantity and the reference resource quantity, the resource quantity consumed by the merchant when pushing the historical recommendation information is determined as the target resource quantity, specifically including: The historical data corresponding to the merchant is input into the weight sub-model so that the weight sub-model can determine the weight of the merchant under the premise of maximizing the information push revenue of the merchant and maximizing the amount of resources consumed by the merchant. Based on the first resource quantity, the reference resource quantity, and the weight corresponding to the merchant, the resource quantity consumed by the merchant when pushing the historical recommendation information is determined and used as the target resource quantity.

5. The method as described in claim 4, characterized in that, The target reward value is determined based on the information push revenue corresponding to the merchant and the target resource volume, specifically including: The first reward value is determined based on the target type corresponding to the merchant and the information push revenue of the merchant to which the historical recommendation information belongs; The second reward value is determined based on the target resource quantity; The target reward value is determined based on the first reward value and the second reward value.

6. The method as described in claim 1, characterized in that, Based on the resource volume, the recommended information is pushed, specifically including: The amount of resources a merchant expects to consume when retrieving historical recommendation information; The information to be recommended is determined based on the amount of resources consumed by the merchant when pushing the information to be recommended, as determined by the target model, and the amount of resources the merchant expects to consume when pushing historical recommendation information.

7. An information push device, characterized in that, include: The acquisition module is used to obtain historical data of the merchants to which the information to be recommended belongs; The determination module is used to determine the type of business needs corresponding to the merchant based on the historical data, and use it as the target type; The matching module is used to determine the business model that matches the target type from the pre-trained business models, and use it as the target model; The training module is used to acquire historical data and auxiliary data from each merchant. The auxiliary data includes: Predicted click-through rate and predicted conversion rate for the merchant; The auxiliary data corresponding to the merchant and the historical data of the merchant corresponding to the target type are input into the target model. The target model determines the amount of resources consumed by the merchant under the premise of maximizing the information push revenue of the merchant, which is the first amount of resources. The target model also determines the amount of resources consumed by the merchant after the user executes the historical recommendation information service, which is the second amount of resources. Based on the first resource quantity and the second resource quantity, the amount of resources consumed by the merchant when pushing the historical recommendation information is determined and used as the target resource quantity; Based on the information push revenue corresponding to the merchant and the target resource volume, a target reward value is determined, and the target model is trained to maximize the target reward value. An input module is used to input the historical data into the target model to determine the amount of resources consumed by the merchant when pushing the information to be recommended. The push module is used to push the recommended information based on the amount of resources consumed by the merchant.

8. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the method described in any one of claims 1 to 6.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method described in any one of claims 1 to 6.

Images