Message delivery method and device, electronic equipment and readable storage medium

Abstract

The application discloses a message delivery method and device, electronic equipment and readable storage medium, applied to the field of artificial intelligence technology, comprising: obtaining the configuration condition corresponding to the message delivery party and the historical message bit information of each preset message bit; according to the configuration condition and the historical message bit information, output the recommended score of each preset message bit through a preset gradient boosting tree model; according to the configuration condition and the recommended score of each preset message bit, generate a message delivery scheme, wherein the message delivery scheme comprises each target message bit selected from each preset message bit and the delivery schedule corresponding to each target message bit. The application solves the technical problem of low selection efficiency of the message delivery scheme.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence, and in particular to a message placement method and device, electronic equipment and a readable storage medium. BACKGROUND

[0002] With the rapid development of technology, artificial intelligence technology is also becoming more and more mature. At present, a message placement direction message placement party provides various message positions to enable the message placement party to place messages on the message positions. However, in most cases, the number of message positions is massive, and the message placement party has many restrictions, so that it takes a long time to provide a suitable message placement scheme for the message placement party, thereby resulting in low efficiency of selecting a message placement scheme. SUMMARY

[0003] The main purpose of the present application is to provide a message placement method, device, electronic equipment and readable storage medium, which aims to solve the technical problem of low efficiency of selecting a message placement scheme in the prior art.

[0004] To achieve the above-mentioned purpose, the present application provides a message placement method applied to a message placement device, the message placement method comprising:

[0005] obtaining configuration conditions corresponding to a message placement party and historical message position information of each preset message position;

[0006] outputting, by a preset gradient boosting tree model, a recommended score of each preset message position according to the configuration conditions and the historical message position information;

[0007] generating a message placement scheme according to the configuration conditions and the recommended score of each preset message position, wherein the message placement scheme comprises each target message position selected from each preset message position and a placement schedule corresponding to each target message position.

[0008] To achieve the above-mentioned purpose, the present application also provides a message placement device, comprising:

[0009] an obtaining module configured to obtain configuration conditions corresponding to a message placement party and historical message position information of each preset message position;

[0010] an output module configured to output, by a preset gradient boosting tree model, a recommended score of each preset message position according to the configuration conditions and the historical message position information;

[0011] a generating module configured to generate a message placement scheme according to the configuration conditions and the recommended score of each preset message position, wherein the message placement scheme comprises each target message position selected from each preset message position and a placement schedule corresponding to each target message position.

[0012] The application further provides an electronic device, comprising a memory, a processor, and a program of the message delivery method stored in the memory and executable on the processor, which can realize the steps of the message delivery method as described above when executed by the processor.

[0013] The application further provides a computer readable storage medium, which stores a program of the message delivery method, which can realize the steps of the message delivery method as described above when executed by the processor.

[0014] The application further provides a computer program product, comprising a computer program, which can realize the steps of the message delivery method as described above when executed by the processor.

[0015] The application provides a message delivery method and device, an electronic device, and a readable storage medium. The message delivery method comprises the following steps: obtaining configuration conditions corresponding to a message delivery party and historical message bit information of each preset message bit; outputting a recommended score of each preset message bit through a preset gradient boosting tree model according to the configuration conditions and the historical message bit information; and generating a message delivery scheme according to the configuration conditions and the recommended scores of each preset message bit, wherein the message delivery scheme comprises each target message bit selected from each preset message bit and a delivery schedule corresponding to each target message bit. The recommended score of each preset message bit is generated based on the configuration conditions and the historical message bit information of the preset message bit, and then the message delivery scheme is generated based on the recommended score and the configuration conditions, so that the message delivery party is provided with a suitable message delivery scheme automatically, and the selection efficiency of the message delivery scheme is improved. BRIEF DESCRIPTION OF DRAWINGS

[0016] The accompanying drawings, which are incorporated herein and constitute part of the specification, illustrate embodiments consistent with the application and, together with the description, serve to explain the principles of the application.

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the accompanying drawings needed to be used in the embodiments or prior art description will be briefly introduced as follows. Obviously, for those skilled in the art, other drawings can also be obtained based on these drawings without any creative work.

[0018] Figure 1 FIG. 1 is a flowchart of a first embodiment of the message delivery method of the application;

[0019] Figure 2 FIG. 2 is a flowchart of a second embodiment of the message delivery method of the application;

[0020] Figure 3 A structure diagram of a message bit tree involved in the message delivery method in the embodiments of the present application;

[0021] Figure 4 A structure diagram of an apparatus involved in the message delivery method in the embodiments of the present application;

[0022] Figure 5 A structure diagram of a device of a hardware running environment involved in the message delivery method in the embodiments of the present application.

[0023] The purposes, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0024] In order to make the above purposes, features and advantages of the present application more apparent, clear and complete, the technical solutions in the embodiments of the present application will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of the present application.

[0025] Embodiment One

[0026] The embodiments of the present application provide a message delivery method. In a first embodiment of the message delivery method of the present application, with reference to Figure 1 , the message delivery method comprises:

[0027] Step S10: obtaining configuration conditions corresponding to a message delivery party and historical message bit information of each preset message bit;

[0028] In the embodiment, it should be noted that the message delivery party is a subject of delivering a message, and the message is a subject in various forms carrying information, which includes but is not limited to a text form, a picture form and a video form. The message can be an advertisement, and when the message is an advertisement, the message delivery party is an advertiser or an advertisement delivery party. The configuration conditions are set by the message delivery party and are used to represent the needs of the message delivery party. The preset message bit is a selectable message bit provided by a message bit delivery party for the message delivery party.

[0029] In an implementation, the configuration condition includes at least one of an amount condition, a delivery date condition, a message type condition, a page type condition, and a delivery location condition. The amount condition can be a limited amount value or a limited amount range. The delivery date condition can be a delivery limit date (for example, July 20, 2023) or a delivery limit date range (for example, July 20, 2023 to July 24, 2023, or within July 2023). The message type condition is a limit on the proportion of message types. The message type can be a text type, which can be a text chain type or an image type, which can be a dynamic image type or a static image type. When the message is an advertisement, the message type can also be a full-screen type. For example, the message type condition is a limit on the proportion of image types, which is 50%, and a limit on the proportion of text types, which is 50%. The delivery location condition is a message delivery limit location and / or a message delivery limit location proportion. When the message is an advertisement, the delivery location condition can be an advertisement delivery limit location, which is a limit location of an advertisement delivery page. The limit location can be a first screen, a second screen, or the like. The message delivery limit location proportion is a limit on the proportion of limit locations. For example, the proportion of the first screen is 60%, and the proportion of the second screen is 40%. The page type condition is a delivery limit page type, which can be a home page type, a forum type, a configuration page type, or other types.

[0030] In step S20, a recommendation score of each preset message position is output based on the configuration condition and the historical message position information by using a preset gradient boosting tree model.

[0031] For example, message position features of each preset message position are determined based on the configuration condition and the historical message position information, and each message position feature is mapped to a recommendation score of each preset message position by using the preset gradient boosting tree model.

[0032] In step S20, the step of outputting a recommendation score of each preset message position based on the configuration condition and the historical message position information by using a preset gradient boosting tree model includes:

[0033] In step S21, a page condition feature is extracted based on the configuration condition.

[0034] In an embodiment, if the configuration condition comprises a message type condition, a message type condition feature is extracted based on the message type condition, and / or if the configuration condition comprises a page type condition, a page type condition is extracted based on the page type condition, and / or if the configuration condition comprises a delivery position condition, a delivery position condition feature is extracted based on the delivery position condition; if the number of categories of the features extracted based on the configuration condition is one, the features extracted based on the configuration condition are taken as the page condition feature, and if the number of categories of the features extracted based on the configuration condition is more than one, the features extracted based on the configuration condition are spliced to obtain the page condition feature.

[0035] In another embodiment, the features are unique hot encoding vectors, and if the number of the features is more than one, the unique hot encoding vectors are added to obtain a page condition feature vector as the page condition feature.

[0036] In step S22, message bit information features corresponding to each of the preset message bits are extracted based on the historical message bit information.

[0037] In this embodiment, it should be noted that the historical message bit information comprises historical information of each of the preset message bits, wherein the historical information comprises at least one of selection times, exposure information and click information, the selection times comprise a total selection times of the preset message bit and / or a cumulative selection times of the message delivery party for the preset message bit; the exposure information comprises a total exposure amount and / or an average exposure amount, the average exposure amount can be a daily average exposure amount, a monthly average exposure amount or an annual average exposure amount, which is not limited herein; the click information comprises a total click amount and / or an average click amount, the average click amount can be a daily average click amount, a monthly average click amount or an annual average click amount, which is not limited herein.

[0038] In an example embodiment, if the historical message bit information comprises the number of selections, a number-of-selection feature corresponding to each of the preset message bits is extracted based on the number of selections; and / or, if the historical message bit information comprises exposure information, exposure information features corresponding to each of the preset message bits are extracted based on the exposure information; and / or, if the historical message bit information comprises click information, click information features corresponding to each of the preset message bits are extracted based on the click information; if the number of categories of the features extracted from the historical message bit information is one, the features extracted from the historical message bit information are taken as the message bit information features of the preset message bits; if the number of categories of the features extracted from the historical message bit information is more than one, the features extracted from the historical message bit information are concatenated as the message bit information features of the preset message bits.

[0039] In step S23, the message bit information features corresponding to each of the preset message bits are concatenated with the page condition features respectively to obtain message bit features corresponding to each of the preset message bits.

[0040] In an example embodiment, the message bit information features comprise message bit information encoding vectors, and the page condition features comprise page condition encoding vectors. The message bit information encoding vectors of each of the preset message bits are added to the page condition encoding vectors respectively to obtain message bit encoding vectors as the message bit features corresponding to each of the preset message bits.

[0041] In step S24, the preset gradient boosting tree model is used to output the recommendation scores of each of the preset message bits according to the message bit features corresponding to each of the preset message bits.

[0042] In an example embodiment, the preset gradient boosting tree model is used to map the message bit features to the recommendation scores of each of the preset message bits.

[0043] In step S20, before the step of outputting the recommendation scores of each of the preset message bits by the preset gradient boosting tree model according to the configuration condition and the historical message bit information, the method further comprises:

[0044] In step S01, an already-selected selection message bit is obtained.

[0045] It can be understood that the message publisher can have message bits that must be selected, which are taken as the already-selected selection message bits.

[0046] In an example embodiment, the selection message bits selected by the message publisher are obtained.

[0047] In step S20, the step of extracting message bit information features corresponding to each preset message bit based on the historical message bit information includes:

[0048] In step S02, message bit feature information corresponding to each preset message bit is extracted based on the selected message bit and the historical message bit information, wherein the message bit feature information is obtained by splicing at least one of selection frequency feature, exposure information feature, click information feature, relationship feature with the selected message bit, and page information feature.

[0049] In an embodiment, the historical message bit information further includes relationship information between each preset message bit and the selected message bit, and the relationship information includes a total number of times that each preset message bit and the selected message bit are in the same historical message placement scheme. Based on the relationship information between each preset message bit and the selected message bit, a relationship feature between each preset message bit and the selected message bit is extracted.

[0050] In step S30, a message placement scheme is generated according to the configuration condition and the recommendation scores of each preset message bit, wherein the message placement scheme includes each target message bit selected from each preset message bit and a placement schedule corresponding to each target message bit.

[0051] In this embodiment, it should be noted that the placement schedule is a placement arrangement duration corresponding to a message bit, and the unit of the placement schedule can be day or hour, which is not limited here.

[0052] For example, a message placement scheme is generated according to the recommendation scores of each preset message bit, which satisfies the configuration condition and makes the recommendation scores of each target message bit selected satisfy a preset condition.

[0053] In step S30, the step of generating a message placement scheme according to the configuration condition and the recommendation scores of each preset message bit includes:

[0054] In step S03, the message placement scheme is generated according to the selected message bit, the configuration condition, and the recommendation scores of each preset message bit.

[0055] For example, a message placement scheme is generated according to the recommendation scores of each preset message bit, which satisfies the configuration condition and makes the recommendation scores of each target message bit selected satisfy a preset condition, wherein the selected message bit belongs to each target message bit.

[0056] In step S20, before the step of outputting the recommended score of each preset message bit by the preset gradient boosting tree model according to the configuration condition and the historical message bit information, the step further comprises:

[0057] In step A10, a plurality of training samples and a gradient boosting tree model to be trained are obtained, wherein the plurality of training samples comprise at least one positive sample and at least one negative sample, the positive sample comprises first input feature data and a first real label corresponding to the first input feature data, the first input feature data is message bit information features of a first training message bit selected for message delivery and page condition features corresponding to the first training message bit, message bit features of the first training message bit obtained by splicing, and the first real label is a recommended score determined by the number of selections of the first training message bit, and the more the number of selections, the higher the recommended score; the negative sample comprises second input feature data and a second real label corresponding to the second input feature data, the second input feature data is message bit features of a second training message bit not selected for message delivery and page condition features corresponding to the second training message bit, message bit features of the second training message bit obtained by splicing, and the second real label is a recommended score of a preset value.

[0058] It can be understood that, in order to ensure the model prediction accuracy of the preset gradient boosting tree model obtained by training, the training samples of the gradient boosting tree model to be trained are usually sufficient, for example, one thousand training samples can be obtained, one million training samples can be obtained, and the present application does not limit the same.

[0059] In step A20, the gradient boosting tree model to be trained is iteratively optimized according to the plurality of training samples to obtain the preset gradient boosting tree model.

[0060] For example, the input feature data is mapped to a predicted score by the gradient boosting tree model to be trained, the difference between the predicted score and the real label is obtained, the loss corresponding to the gradient boosting tree model to be trained is constructed according to the difference, it is judged whether the loss corresponding to the gradient boosting tree model to be trained converges, if the loss converges, the gradient boosting tree model to be trained is taken as the preset gradient boosting tree model, if the loss does not converge, the gradient boosting tree model to be trained is updated according to the gradient calculated based on the loss corresponding to the gradient boosting tree model to be trained, and the step of obtaining a plurality of training samples is executed until the loss converges.

[0061] The embodiment of the application provides a message delivery method, by obtaining configuration conditions corresponding to a message delivery party and historical message bit information of each preset message bit; according to the configuration conditions and the historical message bit information, outputting recommended scores of each preset message bit through a preset gradient boosting tree model; and according to the configuration conditions and the recommended scores of each preset message bit, generating a message delivery scheme, wherein the message delivery scheme comprises each target message bit selected from each preset message bit and a delivery schedule corresponding to each target message bit. The recommended scores of each preset message bit are generated based on the configuration conditions and the historical message bit information of the preset message bit, and then the message delivery scheme is generated based on the recommended scores and the configuration conditions, so that the message delivery party is automatically provided with a suitable message delivery scheme, and therefore the selection efficiency of the message delivery scheme is improved.

[0062] Embodiment two

[0063] Further, based on the first embodiment of the application, in another embodiment of the application, the same or similar contents as the above embodiment one can refer to the above introduction, and the following will not be repeated. On this basis, wherein, referring to Figure 2 In step S30, the step of generating a message delivery scheme according to the configuration conditions and the recommended scores of each preset message bit comprises:

[0064] In step S31, a message bit tree is constructed according to the historical message bit information of each preset message bit and the configuration conditions, wherein each node in the message bit tree is each preset message bit and a delivery schedule duration corresponding to each preset message bit, and a path from a first node to a last node in the message bit tree constitutes a recommended scheme.

[0065] Exemplarily, each preset message bit is combined to obtain at least one recommended scheme satisfying the configuration conditions according to the message bit information of each preset message bit, and the message bit tree is constructed according to each recommended scheme.

[0066] In step S31, the message bit information comprises a message bit deliverable date and a message bit delivery cost,

[0067] The step of constructing a message bit tree according to the historical message bit information of each preset message bit and the configuration conditions comprises:

[0068] In step B10, a limit amount and a limit date corresponding to the message delivery scheme are determined according to the configuration conditions.

[0069] In an example embodiment, the configuration condition includes an amount condition and a delivery date condition, if the amount condition includes a limited amount value, the limited amount value is taken as the limited amount, if the amount condition includes a limited amount range, any value in the limited amount range is taken as the limited amount, or the limited amount range is taken as the limited amount; if the delivery date condition includes a delivery limit date, the delivery limit date is taken as the limit date, if the delivery date condition includes a delivery limit date range, the delivery date limit range is taken as the limit date, or any date in the delivery date limit range is taken as the limit date.

[0070] Step B20, selecting at least one candidate message bit in the preset message bit, which corresponds to the message bit delivery date matching the limit date.

[0071] Step B30, generating each of the recommended schemes according to the message bit delivery fee of each of the candidate message bits and the limited amount.

[0072] In the example embodiment, it is to be noted that the message bit delivery fee is the average delivery fee of the message bit, which can be the average daily delivery fee or the average hourly delivery fee.

[0073] Exemplarily, each of the candidate message bits is combined with each delivery schedule to obtain at least one candidate scheme, the delivery amount of each of the candidate schemes is calculated according to the message bit delivery fee of each of the candidate message bits, each scheme corresponding to the delivery amount less than or equal to the limited amount is selected from each of the candidate schemes as each of the recommended schemes.

[0074] Step B40, constructing the message bit tree according to each of the recommended schemes.

[0075] In an example embodiment, a first message bit with the highest frequency of occurrence is selected from the message bits in each of the recommended schemes, the first message bit and the delivery schedule corresponding to the first message bit are taken as the first head node of the message bit tree; each branch node corresponding to the first head node is determined according to each scheme in which the first message bit is located in the recommended schemes; if there are first schemes not containing the first message bit in the recommended schemes, a second message bit with the highest frequency of occurrence is selected from the message bits in each of the first schemes, the second message bit and the delivery schedule corresponding to the second message bit are taken as the second head node of the message bit tree; each branch node corresponding to the second head node is determined according to each scheme in which the second message bit is located in the recommended schemes, and the above steps are recursively cycled until each of the recommended schemes is displayed by the path from each head node to the tail node of the message bit tree.

[0076] In an embodiment, referring to Figure 3 , Figure 3 FIG. 1 is a structural diagram of a message bit tree involved in a message delivery method in an embodiment of the present application, Figure 3 comprises a first head node composed of a message bit A and a delivery schedule a corresponding to the message bit A, a first branch node corresponding to the first head node and connected to the first head node, the first branch node being composed of a message bit B and a delivery schedule b corresponding to the message bit B, a second branch node corresponding to the first head node and connected to the first branch node, the second branch node being composed of a message bit C and a delivery schedule c corresponding to the message bit C, the message bit A, the delivery schedule a, the message bit B, the delivery schedule b, the message bit C and the delivery schedule c constituting a recommendation scheme; a third branch node corresponding to the first head node and connected to the first head node, the third branch node being composed of a message bit D and a delivery schedule d corresponding to the message bit D, a fourth branch node corresponding to the first head node and connected to the third branch node, the fourth branch node being composed of a message bit E and a delivery schedule e corresponding to the message bit E, the message bit A, the delivery schedule a, the message bit D, the delivery schedule d, the message bit E and the delivery schedule e constituting a recommendation scheme; a second head node composed of a message bit F and a delivery schedule f corresponding to the message bit F, a fifth branch node corresponding to the second head node and connected to the second head node, the fifth branch node being composed of a message bit G and a delivery schedule g corresponding to the message bit G, the message bit F, the delivery schedule f, the message bit G and the delivery schedule g constituting a recommendation scheme.

[0077] In step S32, the message delivery scheme is selected from the message bit tree according to the recommendation scores of the preset message bits.

[0078] In step S32, the message delivery scheme is selected from the message bit tree according to the recommendation scores of the preset message bits.

[0079] In step C10, the total score corresponding to each recommendation scheme in the message bit tree is determined according to the recommendation scores of the preset message bits.

[0080] In an embodiment, for any target recommendation scheme in each recommendation scheme, the recommendation score of each message bit in the target recommendation scheme is multiplied by the delivery schedule duration corresponding to the message bit, respectively, to obtain a schedule score, and the sum of the schedule scores is taken as the total score of the target recommendation scheme.

[0081] In step C20, the recommendation scheme with the highest total score is selected from the recommendation schemes as the message delivery scheme.

[0082] The embodiment of the application provides a message delivery method, comprising the following steps: obtaining a first array and a second array to be compared, wherein the array comprises at least one group of key-value pairs; adjusting the order of the key-value pairs in each group in the first array and / or the second array; rendering the difference key-value pairs between the first array and the second array after sorting, to obtain a comparison result; and adjusting the order of the key-value pairs in each group in the first array and / or the second array, thereby avoiding the misjudgment of the difference key-value pairs caused by the non-uniform order of the key-value pairs in the first array and the second array, and improving the accuracy of message delivery.

[0083] Embodiment three

[0084] The embodiment of the application further provides a message delivery device, referring to Figure 4 , the message delivery device comprises:

[0085] An acquisition module is configured to acquire configuration conditions corresponding to a message delivery party and historical message bit information of each preset message bit.

[0086] An output module is configured to output a recommended score of each preset message bit by a preset gradient boosting tree model according to the configuration conditions and the historical message bit information.

[0087] A generation module is configured to generate a message delivery scheme according to the configuration conditions and the recommended score of each preset message bit, wherein the message delivery scheme comprises each target message bit selected from each preset message bit and a delivery schedule corresponding to each target message bit.

[0088] Optionally, the output module is further configured to:

[0089] extract page condition features based on the configuration conditions;

[0090] extract message bit information features corresponding to each preset message bit based on the historical message bit information;

[0091] splice the message bit information features corresponding to each preset message bit and the page condition features respectively, to obtain message bit features corresponding to each preset message bit;

[0092] output the recommended score of each preset message bit by the preset gradient boosting tree model according to the message bit features corresponding to each preset message bit.

[0093] Optionally, before the step of outputting the recommended score of each preset message bit by the preset gradient boosting tree model according to the configuration conditions and the historical message bit information, the message delivery device further comprises:

[0094] acquiring a selected message bit;

[0095] The output module is further configured to:

[0096] Based on the selected message bit and the historical message bit information, message bit feature information corresponding to each of the preset message bits is extracted, wherein the message bit feature information is obtained by splicing at least one of selection frequency feature, exposure information feature, click information feature, relationship feature with the selected message bit, and page information feature;

[0097] The generation module is further configured to:

[0098] According to the selected message bit, the configuration condition, and the recommendation score of each of the preset message bits, the message delivery scheme is generated.

[0099] Optionally, before the step of outputting the recommendation score of each of the preset message bits according to the configuration condition and the historical message bit information through the preset gradient boosting tree model, the message delivery device further comprises:

[0100] A plurality of training samples and a gradient boosting tree model to be trained are obtained, wherein the plurality of training samples include at least one positive sample and at least one negative sample, a positive sample includes first input feature data and a first real label corresponding to the first input feature data, the first input feature data is message bit information feature of a first training message bit selected for message delivery and page condition feature corresponding to the first training message bit, message bit feature of the first training message bit spliced therefrom, and the first real label is a recommendation score determined by the selection frequency of the first training message bit, the more the selection frequency, the higher the recommendation score; a negative sample includes second input feature data and a second real label corresponding to the second input feature data, the second input feature data is message bit feature of a second training message bit not selected for message delivery and page condition feature corresponding to the second training message bit, message bit feature of the second training message bit spliced therefrom, and the second real label is a preset value of the recommendation score;

[0101] According to the plurality of training samples, the gradient boosting tree model to be trained is iteratively optimized to obtain the preset gradient boosting tree model.

[0102] Optionally, the generation module is further configured to:

[0103] According to the historical message bit information of each of the preset message bits and the configuration condition, a message bit tree is constructed, wherein each node in the message bit tree is each of the preset message bits and delivery schedule length corresponding to each of the preset message bits, and a path from a head node to a tail node in the message bit tree constitutes a recommendation scheme.

[0104] According to the recommendation scores of each of the preset message bit, the message bit tree is selected to obtain the message delivery scheme.

[0105] Optionally, the message bit information includes a message bit deliverable date and a message bit delivery cost, and the generating module is further configured to:

[0106] According to the configuration condition, a limit amount and a limit date corresponding to the message delivery scheme are determined.

[0107] At least one candidate message bit in the preset message bit is selected, wherein the candidate message bit is matched with the limit date.

[0108] According to the message bit delivery cost of each of the candidate message bit and the limit amount, each of the recommendation schemes is generated.

[0109] According to each of the recommendation schemes, the message bit tree is constructed.

[0110] Optionally, the generating module is further configured to:

[0111] According to the recommendation scores of each of the preset message bit, a total score corresponding to each of the recommendation schemes in the message bit tree is determined.

[0112] In each of the recommendation schemes, a scheme with the highest total score is selected as the message delivery scheme.

[0113] The message delivery device provided in the present application adopts the message delivery method in the above-mentioned embodiments, and solves the technical problem of low efficiency of selecting a message delivery scheme. Compared with the prior art, the message delivery device provided in the present application has the same beneficial effects as the message delivery method provided in the above-mentioned embodiments, and other technical features in the message delivery device are the same as the features disclosed in the above-mentioned embodiments, which will not be repeated here.

[0114] Embodiment Four

[0115] The present application provides an electronic device, which comprises at least one processor and a memory in communication connection with the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the message delivery method in the above-mentioned embodiments.

[0116] The following refers to Figure 5The diagram illustrates a structural schematic of an electronic device suitable for implementing embodiments of the present disclosure. The electronic devices in the embodiments of the present disclosure may include, but are not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers (PDAs), tablet computers, portable media players (PMPs), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 5 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0117] like Figure 5 As shown, an electronic device may include a processing unit (such as a central processing unit, graphics processing unit, etc.) that can perform various appropriate actions and processes based on programs stored in ROM (Read-Only Memory) or programs loaded from storage devices into RAM (Random Access Memory). RAM also stores various programs and data required for the operation of the electronic device. The processing unit, ROM, and RAM are interconnected via a bus. Input / output (I / O) ports are also connected to the bus.

[0118] Typically, the following systems can be connected to I / O ports: input devices including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices including, for example, magnetic tapes, hard disks, etc.; and communication devices. Communication devices allow electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although electronic devices with various systems are shown in the figures, it should be understood that it is not required to implement or possess all the systems shown. More or fewer systems may be implemented alternatively.

[0119] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device, or installed from a ROM. When the computer program is executed by a processing device, it performs the functions defined above in the methods of embodiments of this disclosure.

[0120] The electronic device provided in the present application adopts the message delivery method in the above embodiments, and solves the technical problem of low selection efficiency of the message delivery scheme. Compared with the prior art, the electronic device provided in the embodiments of the present application has the same beneficial effects as the message delivery method provided in the above embodiments, and other technical features in the electronic device are the same as the features disclosed in the above method embodiments, which will not be repeated here.

[0121] It should be understood that parts of the present disclosure can be realized by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0122] The above merely provides a specific implementation of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0123] Embodiment five

[0124] The present embodiment provides a computer-readable storage medium having computer-readable program instructions stored thereon, the computer-readable program instructions being used to execute the message delivery method in the above embodiments.

[0125] The computer-readable storage medium provided in the embodiments of the present application may, for example, be a U disk, but is not limited to an electric, magnetic, optical, electromagnetic, infrared, or semiconductor system, system or device, or any combination thereof. More specific examples of the computer-readable storage medium can include, but are not limited to: an electric connection with one or more conductive wires, a portable computer disk, a hard disk, RAM, ROM, EPROM (Erasable Programmable Read Only Memory) or flash memory, an optical fiber, a CD-ROM (compact disc read-only memory), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present embodiment, the computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by or in conjunction with an instruction execution system, system or device. The program code contained on the computer-readable storage medium can be transmitted by any suitable medium, including but not limited to: an electric wire, an optical cable, an RF (Radio Frequency) cable, etc., or any suitable combination of the above.

[0126] The computer readable storage medium described above can be included in an electronic device; or can exist separately from the electronic device.

[0127] The computer readable storage medium described above carries one or more programs, when the one or more programs are executed by the electronic device, cause the electronic device to: acquire a configuration condition corresponding to a message publisher and historical message slot information of each preset message slot; output a recommended score of each preset message slot through a preset gradient boosting tree model according to the configuration condition and the historical message slot information; and generate a message publishing scheme according to the configuration condition and the recommended score of each preset message slot, wherein the message publishing scheme includes each target message slot selected from each preset message slot and a publishing schedule corresponding to each target message slot.

[0128] Computer program code for carrying out operations of the present disclosure can be written in one or more programming languages or combinations of languages including object oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a LAN or a WAN, or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0129] The flow diagrams and the block diagrams in the drawings are illustrations of possible architectures, functions, and operations for systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flow diagrams and the block diagrams can represent a module, a segment, or a portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that in some alternative implementations, the functions noted in the block can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or the blocks may

[0130] The modules described in the embodiments of the present disclosure can be implemented in the form of software or in the form of hardware. In some cases, the name of the module does not constitute a limitation on the module itself.

[0131] The computer readable storage medium provided by the present application stores computer readable program instructions for executing the message delivery method described above, and solves the technical problem of low efficiency of selecting a message delivery scheme. Compared with the prior art, the beneficial effects of the computer readable storage medium provided by the embodiments of the present application are the same as those of the message delivery method provided by the above embodiments, and are not described here.

[0132] Embodiment six

[0133] The present application also provides a computer program product, comprising a computer program, which is executed by a processor to implement the steps of the message delivery method as described above.

[0134] The computer program product provided by the present application solves the technical problem of low efficiency of selecting a message delivery scheme. Compared with the prior art, the beneficial effects of the computer program product provided by the embodiments of the present application are the same as those of the message delivery method provided by the above embodiments, and are not described here.

[0135] The above is only the preferred embodiment of the present application, and does not limit the patent scope of the present application, and any equivalent structure or equivalent flow transformation using the content of the present application specification and drawings, or direct or indirect application in other related technical fields, are also included in the patent processing scope of the present application.

Claims

1. A message delivery method, characterized in that, The message delivery method includes: Obtain the configuration conditions corresponding to the message delivery party and the historical message bit information of each preset message bit; Based on the configuration conditions and the historical message position information, a recommendation score for each preset message position is output using a preset gradient boosting tree model. The step of outputting the recommendation score for each preset message position using the preset gradient boosting tree model includes: extracting page condition features based on the configuration conditions; extracting message position information features corresponding to each preset message position based on the historical message position information; concatenating the message position information features corresponding to each preset message position with the page condition features to obtain message position features corresponding to each preset message position; and outputting the recommendation score for each preset message position based on the message position features corresponding to each preset message position using the preset gradient boosting tree model. A message delivery plan is generated based on the configuration conditions and the recommendation scores of each preset message position. The message delivery plan includes target message positions selected from each preset message position and the delivery schedule corresponding to each target message position. The step of generating the message delivery plan based on the configuration conditions and the recommendation scores of each preset message position includes: Based on the historical message information of each preset message bit and the configuration conditions, a message bit tree is constructed, wherein each node in the message bit tree represents each preset message bit and the corresponding delivery schedule duration, and a path from the first node to the last node in the message bit tree constitutes a recommendation scheme; based on the recommendation score of each preset message bit, the message delivery scheme is selected from the message bit tree.

2. The message delivery method as described in claim 1, characterized in that, Before the step of outputting the recommendation score for each preset message bit based on the configuration conditions and the historical message bit information using a preset gradient boosting tree model, the method further includes: Retrieve the selected selection message bits; The step of extracting message bit information features corresponding to each preset message bit based on the historical message bit information includes: Based on the selected message position and the historical message position information, message position feature information corresponding to each preset message position is extracted, wherein the message position feature information is obtained by concatenating at least one of the following: selection count feature, exposure information feature, click information feature, relationship feature with the selected message position, and page information feature; The step of generating a message delivery plan based on the configuration conditions and the recommendation scores of each preset message position includes: The message delivery scheme is generated based on the selected message positions, the configuration conditions, and the recommendation scores of each preset message position.

3. The message delivery method as described in claim 1, characterized in that, Before the step of outputting the recommendation score for each preset message bit based on the configuration conditions and the historical message bit information using a preset gradient boosting tree model, the method further includes: Multiple training samples and a gradient boosting tree model to be trained are obtained. The multiple training samples include at least one positive sample and at least one negative sample. A positive sample includes first input feature data and a first true label corresponding to the first input feature data. The first input feature data is the message position feature of the first training message position, obtained by concatenating the message position information feature of the first training message position selected for message delivery and the page condition feature corresponding to the first training message position. The first true label is a recommendation score determined by the number of times the first training message position is selected; the more times it is selected, the higher the recommendation score. A negative sample includes second input feature data and a second true label corresponding to the second input feature data. The second input feature data is the message position feature of the second training message position, obtained by concatenating the message position feature information of the second training message position not selected for message delivery and the page condition feature corresponding to the second training message position. The second true label is a recommendation score with a preset value. The gradient boosting tree model to be trained is iteratively optimized based on the multiple training samples to obtain the preset gradient boosting tree model.

4. The message delivery method as described in claim 1, characterized in that, The message slot information includes the date the message slot can be delivered and the delivery cost for the message slot. The step of constructing a message bit tree based on the historical message bit information of each preset message bit and the configuration conditions includes: Based on the configuration conditions, determine the limit amount and limit date corresponding to the message delivery scheme; Select at least one candidate message bit from the preset message bits whose delivery date matches the restricted date; Based on the message placement cost of each candidate message position and the limit amount, generate each of the recommended schemes; The message bit tree is constructed based on each of the recommended schemes.

5. The message delivery method as described in claim 1, characterized in that, The step of selecting the message delivery scheme from the message bit tree based on the recommendation scores of each preset message bit includes: Based on the recommendation scores of each preset message bit, determine the total score corresponding to each recommendation scheme in the message bit tree; The scheme with the highest total score among all the recommended schemes is selected as the message delivery scheme.

6. A message delivery device, characterized in that, The message delivery device includes: The acquisition module is used to acquire the configuration conditions corresponding to the message delivery party and the historical message information of each preset message position; The output module is used to output the recommendation score of each preset message bit according to the configuration conditions and the historical message bit information through a preset gradient boosting tree model; The output module is further configured to: extract page condition features based on the configuration conditions; extract message position information features corresponding to each preset message position based on the historical message position information; concatenate the message position information features corresponding to each preset message position with the page condition features to obtain message position features corresponding to each preset message position; and output the recommendation score of each preset message position based on the message position features corresponding to each preset message position through the preset gradient boosting tree model. The generation module is used to generate a message delivery plan based on the configuration conditions and the recommendation scores of each preset message position. The message delivery plan includes each target message position selected from each preset message position and the delivery schedule corresponding to each target message position. The generation module is further configured to construct a message bit tree based on the historical message bit information of each preset message bit and the configuration conditions, wherein each node in the message bit tree represents each preset message bit and the corresponding delivery schedule duration, and a path from the first node to the last node in the message bit tree constitutes a recommendation scheme; and the message delivery scheme is selected from the message bit tree based on the recommendation score of each preset message bit.

7. An electronic device, characterized in that, The electronic device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the steps of the message delivery method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that implements the message delivery method, which is executed by a processor to implement the steps of the message delivery method as described in any one of claims 1 to 5.

Images