Information pushing method and device thereof, and storage medium

Abstract

The application discloses an information pushing method and device and a storage medium. After determining target pushing information, a target conversion unit price of the target pushing information is acquired, a first historical conversion unit price, a first historical price adjustment coefficient and historical pushing consumption of the target pushing information are acquired, a target price adjustment coefficient of the target pushing information is obtained according to the target conversion unit price, the first historical conversion unit price, the first historical price adjustment coefficient and the historical pushing consumption, a current exposure price of the target pushing information is determined according to the target price adjustment coefficient, and whether the target pushing information is pushed to a target consumer object set is determined according to the current exposure price. The application can reasonably adjust the exposure price of the target pushing information, so that the conversion quantity of the target pushing information can reach the expectation. It can be seen that the application can be widely applied to the information pushing technology in the computer technology field.

Description

Technical Field

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

[0002] With the development of internet technology, the internet has become one of the important ways for users to obtain information. For example, users can obtain various advertising information pushed by information providers through internet platforms.

[0003] The click-through rate and conversion rate of information pushed to internet platforms are affected by the exposure price set by the information provider. If the exposure price set by the information provider is unreasonable, it may lead to a decrease in the number of times the information is exposed, resulting in a decrease in the click-through rate and conversion rate, and ultimately preventing the information provider from achieving the expected conversion rate. Summary of the Invention

[0004] The following is an overview of the subject matter described in detail herein. This overview is not intended to limit the scope of the claims.

[0005] This application provides an information push method, apparatus, and storage medium that can reasonably adjust the exposure price of target push information so that the conversion rate of the target push information can reach the expected level.

[0006] On the one hand, embodiments of this application provide an information push method, including the following steps:

[0007] Identify the target message to push;

[0008] Obtain the target conversion price of the target push information;

[0009] The first historical conversion unit price, the first historical price adjustment coefficient, and the historical push cost of the target push information are obtained, wherein the first historical conversion unit price is the conversion unit price of the target push information when it is pushed to the target consumer set during the target time period, the first historical price adjustment coefficient is the price adjustment coefficient of the target push information when it is pushed to the target consumer set during the target time period, and the historical push cost is the push cost of the target push information when it is pushed to the target consumer set during the target time period, and the target consumer set is one of multiple sets obtained by classifying all consumer objects;

[0010] The target price adjustment coefficient for the target push information is obtained based on the target conversion price, the first historical conversion price, the first historical price adjustment coefficient, and the historical push consumption.

[0011] The current exposure price of the target push information is determined based on the target price adjustment coefficient.

[0012] Based on the current exposure price, determine whether to push the target push information to the target consumer group.

[0013] On the other hand, embodiments of this application also provide an information push device, including:

[0014] Information determination unit, used to determine target push information;

[0015] The first acquisition unit is used to acquire the target conversion unit price of the target push information;

[0016] The second acquisition unit is used to acquire the first historical conversion unit price, the first historical price adjustment coefficient, and the historical push consumption of the target push information, wherein the first historical conversion unit price is the conversion unit price of the target push information being pushed to the target consumer object set during the target time period, the first historical price adjustment coefficient is the price adjustment coefficient of the target push information being pushed to the target consumer object set during the target time period, and the historical push consumption is the push consumption of the target push information being pushed to the target consumer object set during the target time period, and the target consumer object set is one of multiple sets obtained by classifying all consumer objects;

[0017] The coefficient acquisition unit is used to obtain the target price adjustment coefficient of the target push information based on the target conversion unit price, the first historical conversion unit price, the first historical price adjustment coefficient and the historical push consumption;

[0018] A price determination unit is used to determine the current exposure price of the target push information based on the target price adjustment coefficient.

[0019] The information push unit is used to determine whether to push the target push information to the target consumer group based on the current exposure price.

[0020] Optionally, the information push device further includes:

[0021] The information acquisition unit is used to acquire information on the purchasing power attributes and consumer group attributes of all consumers.

[0022] The object classification unit is used to classify all the consumer objects according to the consumption capacity attribute information and the consumer group attribute information to obtain multiple consumer object sets.

[0023] Optionally, the consumption capacity attribute information includes income level information, and the consumer group attribute information includes age information; the object classification unit includes:

[0024] The first classification unit is used to perform a first classification process on all the consumer objects according to the age information to obtain a first classification result;

[0025] The second classification unit is used to perform a second classification process on the first classification result based on the income level information to obtain multiple consumer object sets.

[0026] Optionally, the consumer objects in the first classification result include a first type of object and a second type of object. The first type of object has the income level information, and the second type of object has historical consumption amount information and historical consumption quantity information; the second classification unit includes:

[0027] The first classification subunit is used to perform a second classification process on the first type of object based on the income level information to obtain a first pre-classification result, wherein the first pre-classification result includes multiple classifications.

[0028] The first determining unit is used to determine multiple target objects in the first category of objects for each object to be classified in the second category of objects, based on the historical consumption amount information and the historical consumption quantity information of the object to be classified, wherein the distance between the target object and the object to be classified is less than a first preset threshold.

[0029] The second determining unit is used to select the category that includes the target object the most among the multiple categories as the first target category;

[0030] The first classification unit is used to classify the object to be classified into the first target category to obtain multiple consumer object sets.

[0031] Optionally, the first determining unit includes:

[0032] The first determining subunit is used to determine multiple first candidate objects in the first type of objects based on the historical consumption amount information of the object to be classified, wherein the distance between the first candidate object and the object to be classified is less than a second preset threshold.

[0033] The second determining subunit is used to determine multiple target objects from the plurality of first candidate objects based on the historical consumption quantity information of the object to be classified, wherein the second preset threshold is greater than or equal to the first preset threshold.

[0034] Optionally, the first determining unit includes:

[0035] The third determining subunit is used to determine multiple second candidate objects in the first type of objects based on the historical consumption quantity information of the object to be classified, wherein the distance between the second candidate object and the object to be classified is less than a third preset threshold.

[0036] The fourth determining subunit is used to determine multiple target objects from the multiple second candidate objects based on the historical consumption amount information of the object to be classified, wherein the third preset threshold is greater than or equal to the first preset threshold.

[0037] Optionally, the consumer objects in the first classification result include a first type of object and a third type of object, wherein the first type of object has the income level information, and the third type of object does not have the income level information; the second classification unit includes:

[0038] The second classification subunit is used to perform a second classification process on the first type of object based on the income level information to obtain a second pre-classification result, wherein the second pre-classification result includes a second target classification.

[0039] The second classification unit is used to classify the third type of object into the second target category to obtain multiple consumer object sets.

[0040] Optionally, the consumer group attribute information further includes at least one of geographical information, job level information, or job type information; the second classification unit includes:

[0041] The third classification subunit is used to perform a second classification process on the first classification result based on the income level information to obtain a second classification result.

[0042] The fourth classification subunit is used to perform a third classification process on the second classification result based on at least one of the geographical information, the job level information, or the job type information, to obtain multiple consumer object sets.

[0043] Optionally, the coefficient acquisition unit includes:

[0044] The first acquisition subunit is used to acquire the second historical conversion unit price and the second historical price adjustment coefficient of the target push information, wherein the second historical conversion unit price is the conversion unit price of the target push information when it is pushed to all the consumers during the target period, and the second historical price adjustment coefficient is the price adjustment coefficient of the target push information when it is pushed to all the consumers during the target period;

[0045] The second acquisition subunit is used to obtain the target price adjustment coefficient of the target push information based on the target conversion price, the first historical conversion price, the first historical price adjustment coefficient, the historical push consumption, the second historical conversion price, and the second historical price adjustment coefficient.

[0046] Optionally, the target time period is the time period between the first moment before the current moment and the current moment within the current time cycle; the information push device further includes:

[0047] A coefficient maintenance unit is used to maintain the target price adjustment coefficient for a second period of time, wherein the second period of time is a period of time in the current time cycle other than the target period of time.

[0048] Optionally, the price determination unit includes:

[0049] The first calculation unit is used to calculate the predicted conversion price of the target push information based on the target price adjustment coefficient and the target conversion price.

[0050] The second calculation unit is used to calculate the current exposure price of the target push information based on the predicted conversion price.

[0051] Optionally, the information push unit includes:

[0052] An adjustment unit is used to adjust the order of the target push information in the information push sequence according to the current exposure price;

[0053] The push unit is used to determine whether to push the target push information to the target consumer set based on the sorting.

[0054] On the other hand, embodiments of this application also provide an information push device, including:

[0055] At least one processor;

[0056] At least one memory for storing at least one program;

[0057] The information push method described above is implemented when at least one of the programs is executed by at least one of the processors.

[0058] On the other hand, embodiments of this application also provide a computer-readable storage medium storing a processor-executable program, which, when executed by a processor, is used to implement the information push method as described above.

[0059] On the other hand, embodiments of this application also provide a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the information push method described above.

[0060] After determining the target push information, the target conversion cost is first obtained. Then, the first historical conversion cost, the first historical price adjustment coefficient, and the historical push cost are obtained. Next, based on the target conversion cost, the first historical conversion cost, the first historical price adjustment coefficient, and the historical push cost, the target price adjustment coefficient of the target push information is obtained. Here, the first historical conversion cost, the first historical price adjustment coefficient, and the historical push cost are all historical data of the target push information being pushed to the target consumer set during the target time period. The target consumer set is one of multiple sets obtained by classifying all consumers. Therefore, the target price adjustment coefficient is more suitable for the target consumer set. Thus, determining the current exposure price of the target push information based on the target price adjustment coefficient can achieve the purpose of reasonably adjusting the exposure price of the target push information, so that the conversion volume of the target push information can reach the expectation.

[0061] Other features and advantages of this application will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the application. The objectives and other advantages of this application may be realized and obtained by means of the structures particularly pointed out in the description, claims and drawings. Attached Figure Description

[0062] The accompanying drawings are used to provide a further understanding of the technical solutions of this application and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of this application and do not constitute a limitation on the technical solutions of this application.

[0063] Figure 1 This is a schematic diagram of an implementation environment provided in an embodiment of this application;

[0064] Figure 2 This is a flowchart of an information push method provided in one embodiment of this application;

[0065] Figure 3 This is a flowchart of a method for classifying all consumer objects, provided in one embodiment of this application;

[0066] Figure 4 yes Figure 2 A flowchart of a specific method for step 140;

[0067] Figure 5 yes Figure 3 A flowchart of a specific method for step 220;

[0068] Figure 6 yes Figure 5 A flowchart of a specific method for step 222;

[0069] Figure 7 yes Figure 6 A flowchart of a specific method for step 2222;

[0070] Figure 8 yes Figure 6 A flowchart of another specific method for step 2222;

[0071] Figure 9 yes Figure 5 A flowchart of another specific method for step 222;

[0072] Figure 10 yes Figure 2 A flowchart of a specific method for step 150;

[0073] Figure 11 This is a schematic diagram of an information push device provided in one embodiment of this application;

[0074] Figure 12 This is a schematic diagram of an information push device provided in another embodiment of this application. Detailed Implementation

[0075] The present application will be further described below with reference to the accompanying drawings and specific embodiments. The described embodiments should not be considered as limitations on the present application, and all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of the present application.

[0076] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0077] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0078] Before providing a further detailed description of the embodiments of this application, the nouns and terms involved in the embodiments of this application will be explained, and the nouns and terms involved in the embodiments of this application shall be interpreted as follows.

[0079] 1) Conversion refers to the action a user performs based on the information pushed to them, as expected by the information provider. For example, when the information provider is an application provider, the pushed information could be a link to the application. The expected action from the information provider could be registration or activation of the application. If a user registers or activates the application based on the information pushed by the internet platform, a conversion can be considered complete. Similarly, when the information provider is a merchant on an e-commerce platform, the pushed information could be a link to the merchant. The expected action from the information provider could be placing an order with that merchant. If a user places an order with that merchant based on the information pushed by the internet platform, a conversion can be considered complete.

[0080] 2) Cost Per Action (CPA), also known as the unit cost of a conversion, refers to the ratio of the cost paid by an information provider for a specific push notification to the number of conversions achieved within a given time period. For example, if an advertising provider spends 1000 yuan on a particular ad push in one day and achieves 10 conversions, then the CPA for that ad is 100 yuan. Target CPA, or the information provider's expected CPA, refers to the average price offered by the information provider for a single conversion within a given time period. For example, it's the expected CPA set by an advertising provider for a given day. Historical CPA refers to the actual CPA generated by the push notification before the current moment within the current time period. Predicted CPA refers to the CPA that the push notification may generate in the future after the current moment within the current time period. The purpose of predicting CPA is to ensure that the final actual conversion cost of the push notification does not exceed the target conversion cost set by the information provider.

[0081] 3) The price adjustment factor, also known as the price adjustment coefficient or calibration factor, can be used to adjust the exposure price. By adjusting the price adjustment factor, the exposure price can be reasonably adjusted so that the actual CPA of the pushed information does not exceed the information provider's target CPA, thereby meeting the information provider's cost requirements. For example, the exposure price of an advertisement can be reasonably adjusted by adjusting the price adjustment factor so that the actual CPA of the advertisement can be kept close to the target CPA. The historical price adjustment factor refers to the average of the price adjustment factors up to the current moment within the current time period. The target price adjustment factor is the price adjustment factor that ensures that the actual CPA of the pushed information does not exceed the information provider's target CPA.

[0082] 4) Exposure refers to the price at which a push notification is sent or displayed. The price required to send or display a push notification is the exposure price. In some scenarios, for ease of statistics, the exposure price is usually expressed as the price per thousand pushes or per thousand displays. In other words, the exposure price can be considered as the cost of sending or displaying a push notification a thousand times.

[0083] 5) Push cost refers to the fee paid by the information provider for pushing push information. Therefore, the CPA of push information can also be considered as the ratio of the push cost to the conversions obtained within a time period. Historical push cost refers to the cumulative cost paid by the information provider for pushing push information up to the current moment within the current time period.

[0084] 6) Direct-sales e-commerce refers to merchants that do not have their own sales platform and need to sell their products directly on a third-party platform.

[0085] 7) Blockchain is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms. Essentially, a blockchain is a decentralized database, a chain of data blocks linked using cryptographic methods. Each data block contains information about a batch of network transactions, used to verify the validity of the information (anti-counterfeiting) and generate the next block. A blockchain can include a blockchain underlying platform, a platform product service layer, and an application service layer. The blockchain underlying platform can include processing modules such as user management, basic services, smart contracts, and operation monitoring. The user management module is responsible for managing the identity information of all blockchain participants, including maintaining public and private key generation (account management), key management, and maintaining the correspondence between user real identities and blockchain addresses (access management). Under authorization, it also monitors and audits transactions of certain real identities and provides risk control rule configuration (risk control audit). The basic service module is deployed on all blockchain node devices to verify the validity of business requests. After consensus is reached on valid requests, they are recorded in storage. For a new business request, the basic service first performs interface adaptation parsing and authentication (interface adaptation), and then encrypts the business information through a consensus algorithm (consensus management). The blockchain process involves several layers of data transmission and processing. The first layer, the blockchain service layer, provides basic capabilities and implementation frameworks for typical applications. Developers can define contract logic using a programming language and publish it to the blockchain (contract registration). The second layer provides blockchain-based application services for business stakeholders. The third layer provides application services based on blockchain solutions for use by business participants.

[0086] With the development of internet technology, the internet has become one of the most important ways for users to obtain information. For example, users can obtain information through advertisements posted online. Information providers can push various information to internet platforms through information push systems, allowing users to access a wide range of information. A typical application of information push is advertising. Both internet products and physical products have a need to push advertisements on internet platforms to attract users. For internet platforms, pushing information that users are interested in can generate profits and, to some extent, increase user stickiness. Therefore, internet platforms often interface with corresponding information push systems. Information push systems can sort pushed information based on certain mechanisms and then push information according to this sorting. Currently, information push systems generally use a bidding-based ranking method to push information provided by information providers to internet platforms. The ranking of information to be pushed to the internet platform within the information push system is affected by the exposure price set by the information provider. If the exposure price set by the information provider is unreasonable, it may lead to a decrease in the number of times the information is exposed, resulting in a decrease in the click-through rate and conversion rate, and ultimately preventing the information provider from achieving the expected conversion rate.

[0087] To reasonably adjust the exposure price of targeted push information and ensure that the conversion rate of the targeted push information reaches the expected level, this application provides an information push method, an information push device, and a computer-readable storage medium. After determining the targeted push information, the target conversion unit price of the targeted push information is first obtained. Then, the first historical conversion unit price, the first historical price adjustment coefficient, and the historical push cost of the targeted push information are obtained. Next, based on the target conversion unit price, the first historical conversion unit price, the first historical price adjustment coefficient, and the historical push cost, the target price adjustment coefficient of the targeted push information is obtained. The first historical conversion unit price, the first historical price adjustment coefficient, and the historical push cost are all historical data of the targeted push information being pushed to a target consumer set during a target time period. The target consumer set is one of multiple sets obtained by classifying all consumer objects. Therefore, the target price adjustment coefficient is more suitable for the target consumer set. Thus, determining the current exposure price of the targeted push information based on the target price adjustment coefficient can achieve the purpose of reasonably adjusting the exposure price of the targeted push information, so that the conversion rate of the targeted push information reaches the expected level.

[0088] Figure 1 This is a schematic diagram of an implementation environment provided in an embodiment of this application. (Refer to...) Figure 1 The implementation environment includes server 101 and Internet platform 102, wherein server 101 and Internet platform 102 are connected in communication.

[0089] Server 101 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms. Additionally, Server 101 can also be a node in a blockchain.

[0090] Server 101 may integrate an information push function, or server 101 may be equipped with an information push device for implementing the information push function.

[0091] Server 101 has at least the following functions: calculating the target price adjustment coefficient of the target push information, calculating the exposure price based on the target price adjustment coefficient, and determining whether to push the target push information to the target consumer set based on the exposure price. For example, when the target conversion unit price of the target push information is obtained, the server can obtain the first historical conversion unit price, the first historical price adjustment coefficient, and the historical push cost of the target push information. Then, based on the target conversion unit price, the first historical conversion unit price, the first historical price adjustment coefficient, and the historical push cost, the server can calculate the target price adjustment coefficient of the target push information. Then, based on the target price adjustment coefficient, the server can determine the current exposure price of the target push information. Finally, based on the current exposure price, the server can determine whether to push the target push information to the target consumer set.

[0092] Internet platform 102 can be any internet application platform such as an e-commerce platform, web browsing platform, payment platform, social platform, content interaction platform, education platform, video sharing platform, etc. Internet platform 102 can receive push information from server 101 and display or push the push information to the user.

[0093] In one optional implementation, in response to receiving the target conversion price of a target advertisement sent by an advertiser, server 101 obtains the first historical conversion price, the first historical price adjustment coefficient, and the historical push cost of the target advertisement. Based on the target conversion price, the first historical conversion price, the first historical price adjustment coefficient, and the historical push cost, server 101 calculates the target price adjustment coefficient of the target advertisement. Then, based on the target price adjustment coefficient, server 101 determines the current exposure price of the target advertisement and determines whether to push the target advertisement to the target consumer set. When server 101 determines to push the target advertisement to the target consumer set, server 101 pushes the target advertisement to internet platform 102. In response to receiving the target advertisement, internet platform 102 displays the target advertisement to the target consumer set.

[0094] Figure 2 This is a flowchart illustrating an information push method provided in an embodiment of this application. In this embodiment, a server is used as the execution entity for explanation. (Refer to...) Figure 2 The information push method includes, but is not limited to, steps 110 to 160.

[0095] Step 110: Determine the target push information.

[0096] In this step, the target push notification is the push notification for which the information provider wishes to adjust the exposure price. When the actual conversion price of a certain push notification does not meet the information provider's expectations, the information provider can send a price adjustment request to the server for that push notification. When the server receives the price adjustment request, it can determine the target push notification based on the request, so that subsequent steps can adjust the exposure price of the target push notification so that the conversion volume of the target push notification can meet the information provider's expectations.

[0097] It should be noted that the target push information can be of various types, such as advertising information, application download links, webpage links, etc. This embodiment does not make specific limitations on this.

[0098] Step 120: Obtain the target conversion cost of the target push information.

[0099] In this step, since the target push information was determined in step 110, the target conversion price of the target push information can be obtained, so that subsequent steps can adjust the exposure price of the target push information according to the target conversion price, so that the conversion volume of the target push information can reach the information provider's expectations.

[0100] It should be noted that there are different ways to obtain the target conversion price of the target push information, and this embodiment does not limit it in any specific way. For example, after the target push information is determined by step 110, the server can further obtain the target conversion price of the target push information from the information provider; or, the price adjustment request sent by the information provider to the server may include the identification information of the target push information and the target conversion price, and the server can determine the target push information based on the identification information and obtain the target conversion price of the target push information from the price adjustment request.

[0101] Step 130: Obtain the first historical conversion unit price, the first historical price adjustment coefficient, and the historical push consumption of the target push information.

[0102] In this step, since the target push information was determined in step 110 and the target conversion price was obtained in step 120, the first historical conversion price, the first historical price adjustment coefficient, and the historical push cost of the target push information can be obtained. This allows subsequent steps to adjust the exposure price of the target push information based on the target conversion price, the first historical conversion price, the first historical price adjustment coefficient, and the historical push cost, so that the conversion volume of the target push information can meet the expectations of the information provider.

[0103] It should be noted that the first historical conversion cost is the conversion cost of the target push information being pushed to the target consumer set during the target time period, the first historical price adjustment coefficient is the price adjustment coefficient of the target push information being pushed to the target consumer set during the target time period, and the historical push cost is the push cost of the target push information being pushed to the target consumer set during the target time period. The target consumer set is one of multiple sets obtained by classifying all consumer objects.

[0104] It should be noted that the target time period can be any time period before the current time, or a certain time period before the current time within the current time period, or the time period between the first time before the current time within the current time period and the current time. The appropriate selection can be made according to the actual application situation, and this embodiment does not impose specific limitations on it. The time period refers to a cycle of information push, which can be one day, one week, or one month, etc., and can be appropriately selected according to the actual application situation. This embodiment does not impose specific limitations on it.

[0105] In an alternative implementation, such as Figure 3 As shown, classifying all consumer objects may include, but is not limited to, steps 210 and 220.

[0106] Step 210: Obtain the spending power attribute information and consumer group attribute information of all consumer objects.

[0107] It should be noted that the consumer spending power attribute information may include income level information, social class information, etc., and this embodiment does not specifically limit it. The consumer group attribute information may include age information, geographical information, job level information, or job type information, etc., and this embodiment also does not specifically limit it.

[0108] Step 220: Classify all consumer objects according to their spending power and consumer group attributes to obtain multiple consumer object sets.

[0109] It should be noted that since the consumption capacity attribute information and consumption group attribute information of all consumer objects are obtained in step 210, all consumer objects can be classified according to the consumption capacity attribute information and the consumption group attribute information to obtain multiple consumer object sets. This allows for the storage of information such as the conversion unit price, price adjustment coefficient, and push consumption corresponding to each consumer object set, and facilitates subsequent steps to adjust the price of push information for each consumer object set separately.

[0110] Step 140: Based on the target conversion price, the first historical conversion price, the first historical price adjustment coefficient, and the historical push consumption, obtain the target price adjustment coefficient for the target push information.

[0111] In this step, since the target conversion price of the target push information was obtained in step 120, and the first historical conversion price, the first historical price adjustment coefficient, and the historical push cost of the target push information were obtained in step 130, the target price adjustment coefficient of the target push information can be obtained based on the target conversion price, the first historical conversion price, the first historical price adjustment coefficient, and the historical push cost. This allows subsequent steps to adjust the exposure price of the target push information based on the target price adjustment coefficient, so that the conversion volume of the target push information can meet the expectations of the information provider.

[0112] It should be noted that the conversion price of the push notification can be calculated according to the conversion price formula (the specific conversion price formula will be given in the later embodiments). The first historical conversion price, the first historical price adjustment coefficient, the historical push consumption, and the current price adjustment coefficient of the push notification are all input parameters in the conversion price formula, while the conversion price of the push notification is the output result of the conversion price formula. Since the target conversion price of the target push notification was obtained in step 120, the target conversion price can be substituted into the output result of the conversion price formula, and then the target price adjustment coefficient of the target push notification can be calculated in reverse.

[0113] Step 150: Determine the current exposure price of the target push information based on the target price adjustment coefficient.

[0114] In this step, since the target price adjustment coefficient for the target push information was obtained in step 140, the current exposure price of the target push information can be determined based on this coefficient. Because the current exposure price is determined based on historical data of the target push information being pushed to the target consumer set during the target time period and the target conversion rate of the target push information, this current exposure price is more suitable for the target consumer set. Therefore, by pushing the target push information according to the current exposure price, the conversion rate of the target push information can meet the information provider's expectations.

[0115] Step 160: Based on the current exposure price, determine whether to push the target information to the target consumer group.

[0116] In this step, since the current exposure price of the target push information was obtained in step 150, it can be determined whether to push the target push information to the target consumer group based on the current exposure price. Because the current exposure price is more suitable for the target consumer group, pushing the target push information according to the current exposure price can ensure that the conversion rate of the target push information meets the information provider's expectations.

[0117] It should be noted that the server uses a bidding-based ranking system to push information provided by information providers to the internet platform. Since different information providers offer different push notifications, and even the same information provider may offer different notifications, the push notifications will be ranked differently based on their corresponding exposure prices. In this embodiment, because the current exposure price of the target push notification is adjusted according to the target price adjustment coefficient, the current exposure price affects the push notification ranking. Therefore, after determining the current exposure price of the target push notification, it is necessary to re-rank the target push notification and other push notifications to determine whether to push the target push notification to the target consumer set based on the re-ranking result.

[0118] In this embodiment, by employing an information push method including steps 110 to 160 above, after determining the target push information, the target conversion unit price of the target push information is first obtained, and then the first historical conversion unit price, the first historical price adjustment coefficient, and the historical push cost of the target push information are obtained. Then, based on the target conversion unit price, the first historical conversion unit price, the first historical price adjustment coefficient, and the historical push cost, the target price adjustment coefficient of the target push information is obtained. Here, the first historical conversion unit price, the first historical price adjustment coefficient, and the historical push cost are all historical data of the target push information being pushed to the target consumer set during the target time period. Moreover, the target consumer set is one of multiple sets obtained by classifying all consumer objects. Therefore, the target price adjustment coefficient can better adapt to the target consumer set. Thus, determining the current exposure price of the target push information based on the target price adjustment coefficient can achieve the purpose of reasonably adjusting the exposure price of the target push information, so that the conversion volume of the target push information can reach the expectation.

[0119] Reference Figure 4 As shown in one embodiment of this application, step 140 is further described. Step 140 may include, but is not limited to, the following steps:

[0120] Step 141: Obtain the second historical conversion rate and the second historical price adjustment coefficient for the target push information;

[0121] Step 142: Based on the target conversion price, the first historical conversion price, the first historical price adjustment coefficient, the historical push consumption, the second historical conversion price, and the second historical price adjustment coefficient, obtain the target price adjustment coefficient for the target push information.

[0122] It should be noted that the second historical conversion cost is the conversion cost of the target push information when it is pushed to all consumers during the target period, and the second historical price adjustment coefficient is the price adjustment coefficient of the target push information when it is pushed to all consumers during the target period.

[0123] It should be noted that the target time period is the period between the first moment before the current moment and the current moment within the current time cycle. For example, assuming the current time cycle is one day, and the current moment is 10:00, and the first moment is 0:00, then the target time period is the period from 0:00 to 10:00.

[0124] In the conversion price formula, the second historical conversion price and the second historical price adjustment coefficient of the target push information are also input parameters in the conversion price formula. Therefore, after obtaining the second historical conversion price and the second historical price adjustment coefficient of the target push information in step 141, the target price adjustment coefficient of the target push information can be calculated based on the target conversion price, the first historical conversion price, the first historical price adjustment coefficient, the historical push consumption, the second historical conversion price and the second historical price adjustment coefficient.

[0125] The following detailed description of the principle and process of calculating the target price adjustment coefficient is provided with specific examples.

[0126] For a given advertisement, its conversion cost per day can be calculated using formula (1):

[0127]

[0128] Among them, CPA final This is the total cost per conversion for the ad within one day. h This is the total cost of pushing this ad up to the current time within the day. e It represents the total push cost of this ad within the current timeframe of the day, conv h It represents the total conversions of this ad up to the current time on that day. e This represents the total conversions for this ad within the current timeframe of the day. Adjusting the price of this ad involves adjusting the cost at the current moment by adjusting the price adjustment factor. e and conv e This makes CPA finalIt should be as close as possible to the target conversion cost set by the advertiser.

[0129] Based on formula (1), formula (2) can be obtained by decomposing formula (1) as follows:

[0130]

[0131] in:

[0132]

[0133] In formula (2), CPA h This is the second highest historical conversion cost per unit (CPA) for this ad. e It is the future conversion unit price after adjustment by the target price adjustment coefficient. Assuming that the conversion unit price changes linearly with the price adjustment coefficient, then we can obtain the following formula (3):

[0134]

[0135] Where λ is the price adjustment coefficient to be solved, λ h It is the second historical price adjustment factor, CPA. 1.0 It is the conversion unit price for the time period after the current moment on the same day, without price adjustment (i.e., the price adjustment coefficient is 1).

[0136] Substituting formula (3) into formula (2), we get the following formula (4):

[0137]

[0138] At this point, ω can be transformed to simplify equation (4). The transformation of ω includes the transformation of conv... e With conv h The conversion process for the ratio is as follows:

[0139]

[0140] in:

[0141]

[0142] Where, conv λ，e This is the future conversion rate after adjustment by the target price adjustment factor. This is the historical conversion volume after adjustment by the second historical price adjustment factor, cost. λ，e This is the future push consumption after adjustment by the target price adjustment factor. This is the historical push consumption after adjustment by the second historical price adjustment factor, cost. 1.0，eThis refers to the future push cost without price adjustment (i.e., a price adjustment factor of 1). 1.0，h This represents the historical push consumption without price adjustment (i.e., a price adjustment coefficient of 1), where g(t) is a function that depends only on time; specifically, g(t) = 3.314E. -7 t 2 +0.000218t;f(λ)=λ β ,f(λ h )=λ h β , β=2.5.

[0143] Substituting the result obtained after transforming ω into formula (4), we get the following formula (5):

[0144]

[0145] Where, x = λ / λ h , α=g(t).

[0146] Regarding formula (5), CPA final The value of λ can be obtained by solving formula (5) after the target conversion unit price is equal to the target conversion unit price.

[0147] It should be noted that formula (5) is for advertisements targeting all consumers. By monitoring the current cost of the entire advertisement and adjusting the price adjustment coefficient in real time, the future cost of the advertisement can be controlled, thereby keeping the daily cost of the advertisement within the expected range.

[0148] However, the aforementioned price adjustment applies to ads targeting all consumers. For some ads, the conversion cost per click (CPC) may be lower for certain consumer groups. A lower CPC results in a lower exposure cost, impacting the ad's competitiveness in bidding rankings. When an ad's competitiveness decreases, it becomes less likely to be selected for push notifications, reducing its exposure and consequently lowering conversion rates. This ultimately prevents advertisers from achieving their expected conversion rates. Research indicates that this problem arises because different consumer groups have varying interests in different ads. The level of interest influences click-through rates and purchase intentions, thus affecting conversion rates. To address this, consumer groups can be categorized. Then, for the consumer groups causing the lower CPC, the price adjustment coefficient for ads targeting these groups can be appropriately increased, raising the exposure cost and maximizing ad exposure to these groups to help advertisers achieve their desired conversion rates.

[0149] To effectively categorize consumers, specific classification criteria need to be determined based on the application scenario. For example, in the direct-sales e-commerce scenario, consumers can be categorized based on their age and historical purchasing habits. Research and analysis show that, qualitatively, consumer age is closely related to their purchasing habits. Many older consumers use online shopping apps less frequently, making direct-sales e-commerce attractive to this demographic. Quantitatively, consumers of different age groups have varying purchasing power. Table 1 below shows a data analysis of the purchase amount and number of items purchased by consumers of different age groups on direct-sales e-commerce platforms during a specific period.

[0150] Table 1

[0151]

[0152]

[0153] It should be noted that the purchase amount and number of items purchased in Table 1 have been anonymized. In the purchase amount section, the larger the value, the larger the purchase amount; in the number of items section, the larger the value, the more items purchased.

[0154] As shown in Table 1, consumers of different age groups exhibit significantly varying levels of interest in the direct-to-consumer (DTC) e-commerce model. If price adjustments are not made based on these consumer groups, it will be impossible to tailor prices to their specific characteristics, potentially leading to lower-than-expected conversion rates. However, by adjusting prices separately for different consumer groups, it's possible to increase ad exposure for those more interested in DTC e-commerce, thereby improving overall advertising effectiveness.

[0155] Since the consumer objects are classified, each advertisement can correspond to a price adjustment coefficient for each consumer object set (i.e., consumer object group). Therefore, the above formula (2) can be adjusted to formula (6):

[0156]

[0157] in:

[0158]

[0159]

[0160]

[0161]

[0162]

[0163] Among them, g p (t) is a function that depends only on time; specifically, g p (t)=3.314E -7 t 2 +0.000218t;f(λ p )=λ p β ,f(λ h )=λ h β β = 2.5; ω is the historical CPA weight; n is the number of categories in the consumer object set; ω p λ is the weight of the set of consumer objects of type p; p λ is the price adjustment coefficient for the p-th type of consumer object set to be solved (i.e., the target price adjustment coefficient in this embodiment); hp It is the historical price adjustment coefficient for the p-th type of consumer object set (i.e., the first historical price adjustment coefficient in this embodiment); CPA hp It is the historical conversion unit price of the p-th type of consumer object set (i.e., the first historical conversion unit price in this embodiment); cost hp It is the historical consumption of the p-th type of consumer object set (i.e., the historical push consumption in this embodiment); It is the future push consumption for the p-th type of consumer set after adjustment by the target price adjustment coefficient.

[0164] Therefore, for formula (6), CPA final The value is equal to the target conversion unit price. By solving formula (6), λ can be obtained. p The value of .

[0165] It should be noted that when adjusting the target price adjustment coefficient for the p-th consumer set, the price adjustment coefficients for other consumer sets remain unchanged. In other words, when adjusting the exposure price of the target ad for the p-th consumer set, the exposure price of the same target ad for other consumer sets remains unchanged.

[0166] After calculating the target price adjustment coefficient for the p-th consumer set using the above formula, the current exposure price of the target ad can be determined based on this coefficient. Furthermore, pushing the target ad at this current exposure price can ensure that the conversion rate meets the advertiser's expectations. After some experimental observation, it was found that both the cost and achievement rate of the target ad improved, with the cost increasing by approximately 6% and the achievement rate by approximately 7%. However, data analysis revealed that in some cases, the price adjustment coefficient for certain consumer sets is unstable, leading to unstable costs for those sets. This instability manifests as: underestimation during a certain period, resulting in low costs, requiring an upward adjustment of the price adjustment coefficient; and overestimation during another period, resulting in high costs, requiring a downward adjustment. This leads to oscillating instability in the price adjustment coefficient. Research analysis revealed that one reason for this problem is the unstable conversion intention of the consumer set towards direct-to-consumer e-commerce ads. Further analysis showed that there are significant differences in purchasing power among consumers with different income levels. As shown in Table 2, Table 2 is a data analysis table of the income level of consumers aged 50 to 55.

[0167] Table 2

[0168] Income level (anonymized) Purchase amount (anonymized) Number of items purchased (anonymized) Low income level 0.12 0.17 Middle income level 0.56 0.62 High income level 0.31 0.27 High income level 0.25 0.20

[0169] It should be noted that the income level, purchase amount, and number of items purchased in Table 2 have been anonymized. In the purchase amount section, the larger the value, the larger the purchase amount; in the number of items purchased section, the larger the value, the more items purchased.

[0170] As shown in Table 2, there are significant differences in purchasing power among consumers with different income levels. Ignoring this difference will lead to substantial variations in the exposure effect of advertisements across different consumer groups, resulting in unstable price adjustment coefficients and affecting the effectiveness of price adjustments. To address this issue, in one embodiment of this application, all consumer groups are first categorized based on income level and age information to obtain multiple appropriately categorized consumer group sets. Then, a target price adjustment coefficient is determined for each consumer group set. Next, based on the target price adjustment coefficient for each consumer group set, the current exposure price of each target push message pushed to each consumer group set is determined. This ensures that when each target push message is pushed at its current exposure price, the conversion rate of each target push message can meet the information provider's expectations.

[0171] The following are various embodiments of specific methods for classifying all consumer objects.

[0172] Reference Figure 5 As shown, in one embodiment, step 220 is further explained. When the consumption capacity attribute information includes income level information and the consumption group attribute information includes age information, step 220 may include, but is not limited to, steps 221 and 222.

[0173] Step 221: Perform a first classification process on all consumers based on their age information to obtain the first classification result.

[0174] In this step, when the consumption capacity attribute information includes income level information and the consumption group attribute information includes age information, all consumers can be first classified according to age information to obtain the first classification result. This allows subsequent steps to further classify the first classification result based on income level information, making the final classification result more appropriate.

[0175] It should be noted that the first classification of all consumers based on age information can be done by categorizing them by each year or by age group; this embodiment does not impose a specific limitation on this. For example, referring to the age group division in Table 1 above, all consumers can be divided into 10 categories based on age information to obtain the first classification result.

[0176] Step 222: Perform a second classification process on the first classification result based on income level information to obtain multiple consumer object sets.

[0177] In this step, since the first classification result based on age information was obtained in step 221, the first classification result can be further processed into a second classification based on income level information to obtain multiple consumer object sets. This allows subsequent steps to reasonably adjust the price of the target push information for each consumer object set, so that the conversion rate of the target push information can reach the expected level.

[0178] It should be noted that the second classification process based on income level information can be carried out by classifying the results according to different income levels, such as classifying them according to the levels of thousands, tens of thousands, hundreds of thousands, etc.; or it can be classified according to specific income levels, such as 3,000 to 6,000 yuan as one category, 7,000 to 10,000 yuan as another category, etc. This embodiment does not make specific limitations on this.

[0179] In an optional implementation, the first classification result can be divided into 4 categories based on income level information, as shown in Table 3 below.

[0180] Table 3

[0181] Income level (anonymized) Income level classification 0 to 0.25 Low income level 0.25 to 0.5 Middle income level 0.5 to 0.75 High income level 0.75 to 1 High income level

[0182] It should be noted that the income level information in Table 3 has been anonymized. In the income level information, the larger the value, the higher the income level.

[0183] In this embodiment, by classifying all consumers based on age and income level information, a more appropriate classification result can be obtained. This allows subsequent steps to reasonably adjust the price of the target push information for each consumer set, so that the conversion rate of the target push information can reach the expected level.

[0184] It should be noted that while all consumers can be categorized based on age and income level, income level information is sometimes unavailable. This prevents a secondary categorization based on the initial categorization. To address this issue, research and analysis have shown that consumers can be categorized by income level using their historical spending amount and quantity information.

[0185] The following are various embodiments of classifying consumers by income level using their historical consumption amount and quantity information.

[0186] Reference Figure 6 As shown, in one embodiment, step 222 is further explained. The consumer objects in the first classification result include a first type of object and a second type of object. The first type of object has income level information, historical consumption amount information and historical consumption quantity information. The second type of object has historical consumption amount information and historical consumption quantity information but does not have income level information. In this case, step 222 may include, but is not limited to, steps 2221 to 2224.

[0187] Step 2221: Perform a second classification process on the first category of objects based on income level information to obtain the first pre-classification result, which includes multiple categories.

[0188] In this step, since the first type of object has income level information, it can be classified into a second category based on the income level information to obtain a first pre-classification result. This allows subsequent steps to use the first pre-classification result as a basis to classify the second type of object, thereby achieving the goal of obtaining multiple consumer object sets by performing a second classification based on the income level information.

[0189] It should be noted that, when performing a second classification on the first category of objects based on income level information, the classification method in Table 3 above can be used as a reference. Based on income level information, the first category of objects can be divided into 4 categories to obtain the first pre-classification result.

[0190] Step 2222: For each object to be classified in the second category of objects, based on the historical consumption amount information and historical consumption quantity information of the object to be classified, determine multiple target objects in the first category of objects, wherein the distance between the target objects and the objects to be classified is less than a first preset threshold.

[0191] In this step, for the second category of objects that have historical consumption amount and quantity information but not income level information, the K-Nearest Neighbor (KNN) classification algorithm can be used to classify each object in the second category using the historical consumption amount and quantity information. Specifically, for each object in the second category, multiple target objects in the first category whose distance from the object to be classified is less than a first preset threshold can be identified based on the object's historical consumption amount and quantity information. This allows subsequent steps to classify the object based on these target objects.

[0192] It should be noted that there are different implementation methods for determining multiple target objects from the first category of objects based on the historical consumption amount information and historical consumption quantity information of the object to be classified, and this embodiment does not specifically limit this. For example, multiple candidate objects can be determined first from the first category of objects based on the historical consumption amount information, and then multiple target objects that are less than a first preset threshold distance from the object to be classified can be determined from these candidate objects based on the historical consumption quantity information; or, multiple candidate objects can be determined first from the first category of objects based on the historical consumption quantity information, and then multiple target objects that are less than a first preset threshold distance from the object to be classified can be determined from these candidate objects based on the historical consumption amount information.

[0193] It should be noted that the first preset threshold can be appropriately selected according to the actual application situation, and this embodiment does not impose specific limitations on it.

[0194] Step 2223: Select the category that contains the most target objects from among the multiple categories as the first target category.

[0195] In this step, since the first pre-classification result includes multiple categories, and multiple target objects were identified in step 2222, the category with the most target objects can be taken as the first target category. This allows subsequent steps to classify the object to be classified into the first target category, thus achieving the classification processing of the second type of object. For example, suppose there are 100 target objects identified in the first type of object, of which 20 belong to the low-income category, 50 belong to the medium-income category, and 30 belong to the high-income category. Then, the medium-income category can be taken as the first target category.

[0196] Step 2224: Classify the objects to be classified into the first target category to obtain multiple sets of consumer objects.

[0197] In this step, since the first target category was determined in step 2223, the object to be classified can be categorized into the first target category, resulting in multiple consumer object sets. This allows subsequent steps to reasonably adjust the price of the target push information for each consumer object set, so that the conversion rate of the target push information can reach the expected level.

[0198] Reference Figure 7 As shown, in one embodiment, step 2222 is further described, and step 2222 may include, but is not limited to, the following steps:

[0199] Step 22221: Based on the historical consumption amount information of the object to be classified, determine multiple first candidate objects in the first category of objects, wherein the distance between the first candidate object and the object to be classified is less than a second preset threshold;

[0200] Step 22222: Based on the historical consumption quantity information of the objects to be classified, determine multiple target objects from multiple first candidate objects, wherein the second preset threshold is greater than or equal to the first preset threshold.

[0201] In this embodiment, in the process of determining multiple target objects in the first category of objects based on the historical consumption amount information and historical consumption quantity information of the objects to be classified, the historical consumption amount information of the objects to be classified can be used first to use the KNN classification algorithm to determine multiple first candidate objects in the first category of objects. Then, the historical consumption quantity information of the objects to be classified can be used again to use the KNN classification algorithm to determine multiple target objects in these first candidate objects.

[0202] For example, in an optional implementation, the historical consumption amount information of the objects to be classified can be used first to use the KNN classification algorithm to determine 200 first candidate objects in the first category of objects. Then, the historical consumption quantity information of the objects to be classified can be used again to use the KNN classification algorithm to determine 100 target objects from these 200 first candidate objects.

[0203] It should be noted that the second preset threshold can be appropriately selected according to the actual application situation, and this embodiment does not impose specific limitations on it.

[0204] Reference Figure 8 As shown, in one embodiment, step 2222 is further described, and step 2222 may also include, but is not limited to, the following steps:

[0205] Step 22223: Based on the historical consumption quantity information of the object to be classified, determine multiple second candidate objects in the first category of objects, wherein the distance between the second candidate object and the object to be classified is less than a third preset threshold;

[0206] Step 22224: Based on the historical consumption amount information of the objects to be classified, determine multiple target objects from multiple second candidate objects, wherein the third preset threshold is greater than or equal to the first preset threshold.

[0207] It should be noted that steps 22223 and 22224 in this embodiment are the same as those in the previous embodiment. Figure 7 Steps 22221 and 22222 in the illustrated embodiment are parallel technical solutions.

[0208] In this embodiment, in the process of determining multiple target objects in the first category of objects based on the historical consumption amount information and historical consumption quantity information of the objects to be classified, the historical consumption quantity information of the objects to be classified can be used first to use the KNN classification algorithm to determine multiple second candidate objects in the first category of objects. Then, the historical consumption amount information of the objects to be classified can be used again to use the KNN classification algorithm to determine multiple target objects in these second candidate objects.

[0209] For example, in an optional implementation, the historical consumption quantity information of the objects to be classified can be used first to identify 150 second candidate objects in the first category of objects using the KNN classification algorithm. Then, the historical consumption amount information of the objects to be classified can be used again to identify 100 target objects from these 150 second candidate objects using the KNN classification algorithm.

[0210] It should be noted that the third preset threshold can be appropriately selected according to the actual application situation, and this embodiment does not impose specific limitations on it.

[0211] The preceding embodiments provided methods and steps for classifying the income level of consumers using their historical consumption amount and quantity information. However, for some newly registered consumers, there is no historical consumption amount and quantity information. Therefore, the preceding embodiments for classifying the income level of consumers using their historical consumption amount and quantity information are not applicable to these newly registered consumers. In order to reasonably classify newly registered consumers, some embodiments of this application propose specific methods.

[0212] Reference Figure 9 As shown, in one embodiment, step 222 is further explained. In the case that the consumer objects in the first classification result include a first type of object and a third type of object, wherein the first type of object has income level information and the third type of object does not have income level information, step 222 may also include, but is not limited to, steps 2225 and 2226.

[0213] Step 2225: Perform a second classification process on the first category of objects based on income level information to obtain a second pre-classification result, which includes the second target classification.

[0214] In this step, since the first type of object has income level information, it can be classified into a second category based on the income level information to obtain a second pre-classification result. This allows subsequent steps to use the second pre-classification result as a basis to classify the third type of object, thereby achieving the goal of obtaining multiple consumer object sets by performing a second classification based on the income level information.

[0215] It should be noted that, in an optional implementation, the second pre-classification result of this step can be the first pre-classification result in step 2221. In this case, the second target classification will be included among the multiple classifications of the first pre-classification result.

[0216] Step 2226: Classify the third type of object into the second target category to obtain multiple sets of consumer objects.

[0217] In this step, for the third type of object that does not have income level information (such as newly registered consumers), the third type of object can be directly classified into the second target category to obtain multiple consumer object sets. This allows subsequent steps to reasonably adjust the price of the target push information for each consumer object set, so that the conversion rate of the target push information can reach the expected level.

[0218] In another embodiment, step 222 is further described as follows: if the consumer group attribute information also includes at least one of geographic information, job level information, or job type information, step 222 may also include, but is not limited to, the following steps:

[0219] The first category result is processed into a second category based on income level information to obtain the second category result.

[0220] Based on at least one of the following: geographic information, job level information, or job type information, the second classification result is processed into a third classification to obtain multiple sets of consumer objects.

[0221] In this embodiment, since the consumer group attribute information also includes at least one of the following: regional information, job level information, or job type information, the first classification result can be processed by the second classification based on the income level information to obtain the second classification result. Then, the second classification result can be processed by the third classification based on at least one of the regional information, job level information, or job type information to obtain multiple consumer object sets. This makes the classification of the consumer object sets more appropriate, so that subsequent steps can reasonably adjust the price of the target push information for each consumer object set, so that the conversion rate of the target push information can reach the expected level.

[0222] It should be noted that geographical information may include residential information or workplace information, etc., and this embodiment does not specifically limit it.

[0223] In another embodiment, the information push method is further described as follows: when the target time period is the time period between the first moment before the current moment in the current time cycle and the current moment, after determining the current exposure price of the target push information in step 150, the information push method may also include, but is not limited to, the following steps:

[0224] The target price adjustment coefficient will be maintained for the second period, where the second period is the time period in the current time cycle excluding the target period.

[0225] In this embodiment, after determining the current exposure price of the target push information in step 150, the target price adjustment coefficient obtained in step 140 can be maintained for the second time period, so that the current exposure price of the target push information can remain stable in the second time period, so as to effectively obtain the click-through rate and conversion volume of the target push information, thereby providing an effective data source for the next price adjustment process.

[0226] Reference Figure 10 As shown, in one embodiment, step 150 is further described, and step 150 may include, but is not limited to, steps 151 and 152.

[0227] Step 151: Calculate the predicted conversion price of the target push information based on the target price adjustment coefficient and the target conversion price.

[0228] In this step, since the target conversion price of the target push information was obtained in step 120 and the target price adjustment coefficient of the target push information was obtained in step 140, the predicted conversion price of the target push information can be calculated based on the target price adjustment coefficient and the target conversion price, so that subsequent steps can calculate the current exposure price of the target push information based on the predicted conversion price.

[0229] It should be noted that predicting the conversion cost per unit (CCU) ensures that the final actual conversion cost per unit of the targeted push message does not exceed the target conversion cost set by the information provider.

[0230] It should be noted that the predicted conversion cost per unit of the target push information can be calculated using the following formula (7):

[0231]

[0232] Where SmartBid is the predicted conversion unit price to be solved; CPA Target Target conversion unit price; The billing ratio is the ratio between the actual cost of a single click on the target message and the bid made by the information provider for that single click; λ p This is the target price adjustment factor for the p-th type of consumer set. Due to CPA Target , and λ p Both are available, so the predicted conversion unit price of the target push information can be calculated according to formula (7), so that the current exposure price of the target push information can be calculated in subsequent steps based on the predicted conversion unit price.

[0233] Step 152: Calculate the current exposure price of the target push information based on the predicted conversion cost per unit.

[0234] In this step, since the predicted conversion cost per unit of the target push information was calculated in step 151, the current exposure cost of the target push information can be calculated based on the predicted conversion cost per unit.

[0235] It should be noted that the current exposure cost of the target push information can be calculated using the following formula (8):

[0236] eCPM=SmartBid·pCVR·pCTR·1000 (8)

[0237] Wherein: eCPM is the cost per thousand impressions (i.e., the current exposure cost in this step); pCVR is the estimated conversion rate obtained by the conversion rate prediction model based on the conversion data provided by the information provider, which can be obtained through conversion logs; pCTR is the estimated click-through rate obtained by the click data of the target push information. Both the conversion rate prediction model and the click-through rate prediction model are common models in this field. Therefore, for a detailed description of the principles of the conversion rate prediction model and the click-through rate prediction model, please refer to the relevant descriptions of conversion rate prediction models and click-through rate prediction models in related technologies, which will not be repeated here.

[0238] In another embodiment, step 160 is further described, and step 160 may include, but is not limited to, the following steps:

[0239] Adjust the order of target push information in the information push sequence based on the currently exposed price;

[0240] The decision on whether to push the target message to the target consumer set is based on the sorting.

[0241] In this embodiment, since the current exposure price of the target push information is determined in step 150, the order of the target push information in the information push sequence can be adjusted according to the current exposure price first, and then the order of the target push information in the information push sequence can be used to determine whether to push the target push information to the target consumer set.

[0242] It should be noted that when servers push information to internet platforms, they generally use a bidding-based ranking method to determine the information to be pushed. The specific process is as follows: the server first obtains the exposure price of each push notification, then sorts these prices to form an information push sequence, for example, sorting the push notifications in descending order of exposure price. Next, the server selects a certain number of push notifications from this sequence as the information to be pushed, and then pushes these to the internet platform. Since the current exposure price of the target push notification has been adjusted according to the target price adjustment factor, this may change the order of the target push notification in the information push sequence. Therefore, after adjusting the order of the target push notification in the information push sequence according to the current exposure price, it is necessary to determine whether the target push notification can be pushed to the target consumer group based on this order.

[0243] It should be noted that since the current exposure price of the target push information is determined based on the target price adjustment coefficient, and this target price adjustment coefficient is adapted to the target consumer set, the current exposure price of the target push information after adjustment can be more suitable for the target consumer set. Therefore, when the target push information is pushed to the target consumer set according to the current exposure price, the conversion rate of the target push information can reach the expectations of the information provider.

[0244] Reference Figure 11 This application also discloses an information push device, which can be used as such Figure 1 The server 101 in the illustrated embodiment, or deployed in such a location Figure 1 The server 101 in the illustrated embodiment implements the information push method as described in the previous embodiments. The information push device 300 includes:

[0245] Information determination unit 301 is used to determine target push information;

[0246] The first acquisition unit 302 is used to acquire the target conversion unit price of the target push information;

[0247] The second acquisition unit 303 is used to acquire the first historical conversion unit price, the first historical price adjustment coefficient, and the historical push consumption of the target push information. The first historical conversion unit price is the conversion unit price of the target push information when it is pushed to the target consumer object set during the target time period. The first historical price adjustment coefficient is the price adjustment coefficient of the target push information when it is pushed to the target consumer object set during the target time period. The historical push consumption is the push consumption of the target push information when it is pushed to the target consumer object set during the target time period. The target consumer object set is one of multiple sets obtained by classifying all consumer objects.

[0248] The coefficient acquisition unit 304 is used to obtain the target price adjustment coefficient of the target push information based on the target conversion unit price, the first historical conversion unit price, the first historical price adjustment coefficient and the historical push consumption.

[0249] Price determination unit 305 is used to determine the current exposure price of the target push information based on the target price adjustment coefficient;

[0250] The information push unit 306 is used to determine whether to push target information to the target consumer group based on the current exposure price.

[0251] In one embodiment, the information push device 300 further includes:

[0252] The information acquisition unit is used to acquire information on the purchasing power attributes and consumer group attributes of all consumers.

[0253] The object classification unit is used to classify all consumer objects according to their purchasing power and consumer group attributes, resulting in multiple consumer object sets.

[0254] In one embodiment, when the consumption capacity attribute information includes income level information and the consumer group attribute information includes age information, the object classification unit includes:

[0255] The first classification unit is used to perform a first classification process on all consumers based on age information to obtain the first classification result;

[0256] The second classification unit is used to perform a second classification process on the results of the first classification based on income level information, resulting in multiple sets of consumer objects.

[0257] In one embodiment, where the consumer objects in the first classification result include a first category of objects and a second category of objects, and the first category of objects has income level information, while the second category of objects has historical consumption amount information and historical consumption quantity information, the second classification unit includes:

[0258] The first classification subunit is used to perform a second classification process on the first type of object based on income level information to obtain a first pre-classification result, which includes multiple classifications.

[0259] The first determining unit is used to determine multiple target objects in the first category of objects for each object to be classified in the second category of objects, based on the historical consumption amount information and historical consumption quantity information of the object to be classified, wherein the distance between the target object and the object to be classified is less than a first preset threshold.

[0260] The second determining unit is used to select the category that includes the most target objects among multiple categories as the first target category;

[0261] The first classification unit is used to classify the objects to be classified into the first target category, resulting in multiple sets of consumer objects.

[0262] In one embodiment, the first determining unit includes:

[0263] The first determining subunit is used to determine multiple first candidate objects in the first category of objects based on the historical consumption amount information of the objects to be classified, wherein the distance between the first candidate objects and the objects to be classified is less than a second preset threshold.

[0264] The second determining subunit is used to determine multiple target objects from multiple first candidate objects based on the historical consumption quantity information of the objects to be classified, wherein the second preset threshold is greater than or equal to the first preset threshold.

[0265] In one embodiment, the first determining unit includes:

[0266] The third determining subunit is used to determine multiple second candidate objects in the first category of objects based on the historical consumption quantity information of the object to be classified, wherein the distance between the second candidate object and the object to be classified is less than the third preset threshold.

[0267] The fourth determining subunit is used to determine multiple target objects from multiple second candidate objects based on the historical consumption amount information of the object to be classified, wherein the third preset threshold is greater than or equal to the first preset threshold.

[0268] In one embodiment, where the consumer objects in the first classification result include a first category of objects and a third category of objects, and the first category of objects has income level information while the third category of objects does not, the second classification unit includes:

[0269] The second classification subunit is used to perform a second classification process on the first category of objects based on income level information to obtain a second pre-classification result, which includes the second target classification.

[0270] The second classification unit is used to classify the third type of object into the second target category, resulting in multiple sets of consumer objects.

[0271] In one embodiment, when the consumer group attribute information further includes at least one of geographic information, job level information, or job type information, the second classification unit includes:

[0272] The third classification subunit is used to perform a second classification process on the first classification result based on income level information to obtain the second classification result.

[0273] The fourth classification subunit is used to perform third classification processing on the second classification results based on at least one of the following: regional information, job level information, or job type information, to obtain multiple sets of consumer objects.

[0274] In one embodiment, the coefficient acquisition unit 304 includes:

[0275] The first acquisition subunit is used to acquire the second historical conversion unit price and the second historical price adjustment coefficient of the target push information, wherein the second historical conversion unit price is the conversion unit price of the target push information pushed to all consumers during the target period, and the second historical price adjustment coefficient is the price adjustment coefficient of the target push information pushed to all consumers during the target period.

[0276] The second acquisition subunit is used to obtain the target price adjustment coefficient of the target push information based on the target conversion price, the first historical conversion price, the first historical price adjustment coefficient, the historical push consumption, the second historical conversion price, and the second historical price adjustment coefficient.

[0277] In one embodiment, when the target time period is the time period between the first moment before the current moment and the current moment within the current time cycle, the information push device 300 further includes:

[0278] The coefficient maintenance unit is used to maintain the target price adjustment coefficient for a second period, where the second period is the time period in the current time cycle excluding the target period.

[0279] In one embodiment, the price determination unit 305 includes:

[0280] The first calculation unit is used to calculate the predicted conversion price of the target push information based on the target price adjustment coefficient and the target conversion price.

[0281] The second calculation unit is used to calculate the current exposure price of the target push information based on the predicted conversion cost.

[0282] In one embodiment, the information push unit 306 includes:

[0283] The adjustment unit is used to adjust the order of target push information in the information push sequence based on the current exposure price;

[0284] The push unit is used to determine whether to push target push information to the target consumer set based on the sorting.

[0285] It should be noted that since the information push device 300 of this embodiment can implement the information push method with the server as the execution subject in the previous embodiment, the information push device 300 of this embodiment and the information push method with the server as the execution subject in the previous embodiment have the same technical principle and the same beneficial effect. In order to avoid repetition, it will not be described again here.

[0286] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0287] Reference Figure 12 This application also discloses an information push device, the information push device 400 comprising:

[0288] At least one processor 401;

[0289] At least one memory 402 is used to store at least one program;

[0290] When at least one of the programs is executed by at least one of the processors 401, the information push method as described in any of the preceding embodiments is implemented.

[0291] This application also discloses a computer-readable storage medium storing a processor-executable program, which, when executed by a processor, is used to implement the information push method as described in any of the preceding embodiments.

[0292] This application also discloses a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the information push method described in any of the preceding embodiments.

[0293] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatuses.

[0294] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0295] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, apparatuses, or units, and may be electrical, mechanical, or other forms.

[0296] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0297] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0298] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0299] The step numbers in the above method embodiments are set only for ease of explanation and do not limit the order of the steps. The execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

Claims

1. An information push method, characterized in that, Includes the following steps: Identify the target message to push; Obtain the target conversion price of the target push information; The first historical conversion unit price, the first historical price adjustment coefficient, and the historical push cost of the target push information are obtained, wherein the first historical conversion unit price is the conversion unit price of the target push information when it is pushed to the target consumer set during the target time period, the first historical price adjustment coefficient is the price adjustment coefficient of the target push information when it is pushed to the target consumer set during the target time period, and the historical push cost is the push cost of the target push information when it is pushed to the target consumer set during the target time period, and the target consumer set is one of multiple sets obtained by classifying all consumer objects; Obtain the second historical conversion unit price and the second historical price adjustment coefficient of the target push information, wherein the second historical conversion unit price is the conversion unit price of the target push information when it is pushed to all the consumers during the target period, and the second historical price adjustment coefficient is the price adjustment coefficient of the target push information when it is pushed to all the consumers during the target period; The target conversion price, the first historical conversion price, the first historical price adjustment coefficient, the historical push consumption, the second historical conversion price, and the second historical price adjustment coefficient are substituted into the conversion price formula to calculate the target price adjustment coefficient of the target push information. The billing ratio coefficient is obtained by comparing the actual cost of a single click on the target push information with the bid offered by the information provider for that single click. The predicted conversion price of the target push information is obtained by multiplying the target price adjustment coefficient, the target conversion price, and the billing ratio coefficient. The current exposure price of the target push information is calculated based on the predicted conversion cost per unit. Based on the current exposure price, determine whether to push the target push information to the target consumer group.

2. The information push method according to claim 1, characterized in that, The process of classifying all consumer objects includes: Obtain spending power attribute information and consumer group attribute information for all consumers; Based on the consumption capacity attribute information and the consumption group attribute information, all the consumption objects are classified to obtain multiple sets of consumption objects.

3. The information push method according to claim 2, characterized in that, The consumption capacity attribute information includes income level information, and the consumption group attribute information includes age information; The step of classifying all consumer objects based on the consumption capacity attribute information and the consumer group attribute information yields multiple consumer object sets, including: Based on the age information, all the consumers are classified into a first category to obtain a first category result; The first classification result is processed by a second classification based on the income level information to obtain multiple sets of consumer objects.

4. The information push method according to claim 3, characterized in that, The consumer objects in the first classification result include a first type of object and a second type of object. The first type of object has the income level information, and the second type of object has historical consumption amount information and historical consumption quantity information. The second classification process, based on the income level information, yields multiple sets of consumer objects, including: Based on the income level information, the first type of object is subjected to a second classification process to obtain a first pre-classification result, which includes multiple classifications. For each object to be classified in the second category of objects, based on the historical consumption amount information and the historical consumption quantity information of the object to be classified, multiple target objects are determined in the first category of objects, wherein the distance between the target objects and the object to be classified is less than a first preset threshold. The category that includes the target object the most among the multiple categories is selected as the first target category; The objects to be classified are categorized into the first target category to obtain multiple sets of consumer objects.

5. The information push method according to claim 4, characterized in that, The step of determining multiple target objects in the first category of objects based on the historical consumption amount information and the historical consumption quantity information of the objects to be classified includes: Based on the historical consumption amount information of the object to be classified, multiple first candidate objects are determined in the first type of object, wherein the distance between the first candidate object and the object to be classified is less than a second preset threshold. Based on the historical consumption quantity information of the object to be classified, multiple target objects are determined from the plurality of first candidate objects, wherein the second preset threshold is greater than or equal to the first preset threshold.

6. The information push method according to claim 4, characterized in that, The step of determining multiple target objects in the first category of objects based on the historical consumption amount information and the historical consumption quantity information of the objects to be classified includes: Based on the historical consumption quantity information of the object to be classified, multiple second candidate objects are determined in the first type of object, wherein the distance between the second candidate object and the object to be classified is less than a third preset threshold. Based on the historical consumption amount information of the object to be classified, multiple target objects are determined from the multiple second candidate objects, wherein the third preset threshold is greater than or equal to the first preset threshold.

7. The information push method according to claim 3, characterized in that, The consumer objects in the first classification result include a first type of object and a third type of object. The first type of object has the income level information, while the third type of object does not have the income level information. The second classification process, based on the income level information, yields multiple sets of consumer objects, including: Based on the income level information, the first type of object is subjected to a second classification process to obtain a second pre-classification result, which includes a second target classification. The third type of object is classified into the second target category to obtain multiple sets of consumer objects.

8. The information push method according to claim 3, characterized in that, The consumer group attribute information also includes at least one of the following: geographic information, job level information, or job type information; The second classification process, based on the income level information, yields multiple sets of consumer objects, including: Based on the income level information, the first classification result is processed into a second classification result; Based on at least one of the geographical information, job level information, or job type information, the second classification result is subjected to a third classification process to obtain multiple sets of consumer objects.

9. The information push method according to claim 1, characterized in that, The target time period is the time period between the first moment before the current moment and the current moment within the current time cycle; After determining the current exposure price of the target push information based on the target price adjustment coefficient, the information push method further includes: The target price adjustment coefficient is maintained for a second period, wherein the second period is the time period in the current time cycle excluding the target period.

10. The information push method according to claim 1, characterized in that, The step of determining whether to push the target push information to the target consumer group based on the current exposure price includes: Adjust the order of the target push information in the information push sequence according to the current exposure price; Based on the sorting, determine whether to push the target push information to the target consumer set.

11. An information push device, characterized in that, include: Information determination unit, used to determine target push information; The first acquisition unit is used to acquire the target conversion unit price of the target push information; The second acquisition unit is used to acquire the first historical conversion unit price, the first historical price adjustment coefficient, and the historical push consumption of the target push information, wherein the first historical conversion unit price is the conversion unit price of the target push information being pushed to the target consumer object set during the target time period, the first historical price adjustment coefficient is the price adjustment coefficient of the target push information being pushed to the target consumer object set during the target time period, and the historical push consumption is the push consumption of the target push information being pushed to the target consumer object set during the target time period, and the target consumer object set is one of multiple sets obtained by classifying all consumer objects; The coefficient acquisition unit is used to acquire the second historical conversion unit price and the second historical price adjustment coefficient of the target push information, and substitute the target conversion unit price, the first historical conversion unit price, the first historical price adjustment coefficient, the historical push consumption, the second historical conversion unit price and the second historical price adjustment coefficient into the conversion unit price formula to calculate the target price adjustment coefficient of the target push information. The second historical conversion unit price is the conversion unit price of the target push information pushed to all the consumers in the target period, and the second historical price adjustment coefficient is the price adjustment coefficient of the target push information pushed to all the consumers in the target period. The pricing unit is used to obtain a billing ratio coefficient based on the ratio between the actual cost of a single click on the target push information and the bid offered by the information provider for the single click; multiply the target price adjustment coefficient, the target conversion price, and the billing ratio coefficient to obtain the predicted conversion price of the target push information; and calculate the current exposure price of the target push information based on the predicted conversion price. The information push unit is used to determine whether to push the target push information to the target consumer group based on the current exposure price.

12. An information push device, characterized in that, include: At least one processor; At least one memory for storing at least one program; The information push method as described in any one of claims 1 to 10 is implemented when at least one of the programs is executed by at least one of the processors.

13. A computer-readable storage medium, characterized in that, It stores a processor-executable program, which, when executed by the processor, is used to implement the information push method as described in any one of claims 1 to 10.

Images