Content pushing method and device, computer device and storage medium

Abstract

The application relates to a content pushing method and device, computer equipment and a storage medium. The method comprises the following steps: obtaining a target content interaction record set corresponding to a target pushing object; the content interaction record comprises an interaction behavior type of a content interaction subject to target pushing content; obtaining an object approval stage sequence corresponding to the target pushing object; the object approval stage sequence comprises a plurality of object approval stages arranged according to an object approval degree, and the object approval stage sequence is obtained according to an interaction behavior type corresponding to an object conversion link of the target pushing object; obtaining a standard interaction behavior type corresponding to the object approval stage; determining a content interaction subject with a matched interaction behavior type and standard interaction behavior type according to the target content interaction record set; and obtaining a target content interaction subject corresponding to the object approval stage based on the content interaction subject with the matched interaction behavior type and standard interaction behavior type. The method can improve the accuracy of content pushing.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computers, in particular to a content pushing method and device, computer equipment and a storage medium. BACKGROUND

[0002] With the development of computer technology and Internet technology, more and more content is pushed through the network, for example, through the network to push advertisements, and advertisers place their advertisements on the platform of the traffic master through the platform of the traffic master, thereby realizing the use of network to push advertisements.

[0003] In the traditional content pushing method, when pushing content, the content is usually determined according to the heat of the content, for example, the content with high heat is pushed, however, the content with high heat is not necessarily the content that the user is interested in, so the traditional content pushing method cannot accurately push the content, resulting in low accuracy of content pushing. SUMMARY

[0004] Therefore, it is necessary to provide a content pushing method, device, computer equipment and storage medium capable of improving the accuracy of content pushing.

[0005] A content pushing method, the method comprising: obtaining a target content interaction record set corresponding to a target pushing object; the content interaction record in the target content interaction record set comprises an interaction behavior type of a content interaction subject to a target pushing content, and the target pushing content is a pushing content related to the target pushing object; obtaining an object recognition stage sequence corresponding to the target pushing object, the object recognition stage sequence comprising a plurality of object recognition stages arranged according to the object recognition degree, and the object recognition stage sequence is obtained according to the interaction behavior type corresponding to the object conversion link of the target pushing object; obtaining a standard interaction behavior type corresponding to the object recognition stage in the object recognition stage sequence; determining a content interaction subject with a matching interaction behavior type and the standard interaction behavior type according to the target content interaction record set, and obtaining a target content interaction subject corresponding to the object recognition stage based on the content interaction subject with the matching interaction behavior type and the standard interaction behavior type.

[0006] The content pushing device comprises: a target content interaction record set obtaining module, configured to obtain a target content interaction record set corresponding to a target pushing object; a content interaction record in the target content interaction record set comprises an interaction behavior type of a content interaction subject to target pushing content, and the target pushing content is pushing content related to the target pushing object; an object recognition stage sequence obtaining module, configured to obtain an object recognition stage sequence corresponding to the target pushing object, the object recognition stage sequence comprising a plurality of object recognition stages arranged according to an object recognition degree, and the object recognition stage sequence being obtained according to an interaction behavior type corresponding to an object conversion link of the target pushing object; a standard interaction behavior type obtaining module, configured to obtain a standard interaction behavior type corresponding to an object recognition stage in the object recognition stage sequence; and a target content interaction subject obtaining module, configured to determine a content interaction subject with a matched interaction behavior type and the standard interaction behavior type according to the target content interaction record set, and obtain a target content interaction subject corresponding to the object recognition stage based on the content interaction subject with the matched interaction behavior type and the standard interaction behavior type.

[0007] In some embodiments, the target content interaction subject obtaining module comprises: a sub-interaction record set obtaining unit, configured to obtain content interaction records corresponding to each content interaction subject from the target content interaction record set to form a sub-interaction record set corresponding to the content interaction subject; and a first target content interaction subject obtaining unit, configured to take the content interaction subject as a target content interaction subject corresponding to the object recognition stage when the interaction behavior type in the sub-interaction record set matches the standard interaction behavior type.

[0008] In some embodiments, the object recognition stage corresponds to a stage priority, the stage priority being in a positive correlation with an object recognition degree corresponding to the object recognition stage; the standard interaction behavior type obtaining module is further configured to sequentially obtain a current object recognition stage from the object recognition stage sequence according to a stage priority order, obtain a current standard interaction behavior type corresponding to the current object recognition stage, and take the content interaction subject as a target content interaction subject corresponding to the current object recognition stage when the interaction behavior type in the sub-interaction record set matches the current standard interaction behavior type; otherwise, sequentially obtain a current object recognition stage from the object recognition stage sequence according to a stage priority order to update the current object recognition stage until the interaction behavior type in the sub-interaction record set matches the current standard interaction behavior type.

[0009] In some embodiments, the standard interaction behavior type obtaining module is further configured to obtain a current standard interaction behavior type corresponding to a current object recognition stage and an interaction behavior valid duration corresponding to the current object recognition stage; the first target content interaction subject obtaining unit is further configured to obtain an interaction behavior valid time period corresponding to the current object recognition stage based on the interaction behavior valid duration corresponding to the current object recognition stage; and when there is a content interaction record in the sub-interaction record set, the interaction behavior type of which matches the current standard interaction behavior type and the interaction time belongs to the interaction behavior valid time period, the content interaction subject is taken as a target content interaction subject corresponding to the current object recognition stage.

[0010] In some embodiments, the target content interaction subject obtaining module includes an interaction record sub-set obtaining unit configured to divide the target content interaction record set according to interaction behavior types to obtain interaction record sub-sets respectively corresponding to the interaction behavior types; and a second target content interaction subject obtaining unit configured to obtain an interaction record sub-set corresponding to the standard interaction behavior type, and obtain the target content interaction subject corresponding to the object recognition stage based on the interaction record sub-set corresponding to the standard interaction behavior type.

[0011] In some embodiments, the object recognition stage corresponds to a stage priority, and the stage priority is in a positive correlation with an object recognition degree corresponding to the object recognition stage; the second target content interaction subject obtaining unit is further configured to obtain a current interaction record sub-set corresponding to a standard interaction behavior type corresponding to a current object recognition stage, and form a current content interaction subject set by content interaction subjects in the current interaction record sub-set; obtain a priority interaction record sub-set corresponding to a standard interaction behavior type corresponding to a priority object recognition stage, and form a priority content interaction subject set by content interaction subjects in the priority interaction record sub-set, the priority object recognition stage being an object recognition stage whose stage priority is higher than that of the current object recognition stage in the object recognition stage sequence; and filter out the priority content interaction subject set from the current content interaction subject set to obtain the target content interaction subject corresponding to the current object recognition stage.

[0012] In some embodiments, the apparatus further comprises an object recognition stage sequence obtaining module, which comprises: a target object conversion link acquisition unit, configured to acquire a target object conversion link, the target object conversion link comprising a plurality of interaction behavior types arranged according to conversion progress; wherein the interaction behavior types are obtained according to behavior purposes; a conversion possibility acquisition unit, configured to acquire a conversion possibility corresponding to each of the interaction behavior types; and an object recognition stage sequence obtaining unit, configured to divide the target object conversion link into stages based on the conversion possibility corresponding to each of the interaction behavior types, to obtain the object recognition stage sequence; wherein the standard interaction behavior type corresponding to the object recognition stage is the interaction behavior type divided into the object recognition stage.

[0013] In some embodiments, the object recognition stage sequence obtaining unit is further configured to calculate, based on the conversion possibility corresponding to each of the interaction behavior types, a possibility difference between the conversion possibilities of adjacent interaction behavior types in the target object conversion link; and divide adjacent interaction behavior types with a possibility difference less than a difference threshold into the same object recognition stage.

[0014] In some embodiments, the conversion possibility acquisition unit is further configured to acquire a reference content interaction record set corresponding to the target object conversion link, wherein a content interaction record in the reference content interaction record set comprises a reference interaction subject's interaction behavior type with reference push content, and the reference push content is push content related to a reference push object corresponding to the target object conversion link; count the number of reference interaction subjects of the same interaction behavior type in the reference content interaction record set to obtain a reference subject number corresponding to the interaction behavior type; and obtain the conversion possibility corresponding to the interaction behavior type based on the reference subject number corresponding to the interaction behavior type.

[0015] In some embodiments, the conversion possibility acquisition unit is further configured to acquire a current reference subject number corresponding to a current interaction behavior type; acquire a conversion reference subject number corresponding to a conversion interaction behavior type; calculate a quantity ratio between the conversion reference subject number and the current reference subject number; and take the quantity ratio as the conversion possibility corresponding to the current interaction behavior type.

[0016] In some embodiments, the apparatus further comprises: a content interaction subject quantity obtaining module, configured to obtain a first object approval stage from the sequence of object approval stages, and obtain a content interaction subject quantity corresponding to each target subject feature in a plurality of target content interaction subjects corresponding to the first object approval stage; a representative subject feature selecting module, configured to select a representative subject feature corresponding to the first object approval stage based on the content interaction subject quantity corresponding to the target subject feature; a to-be-pushed subject selecting module, configured to select a target content interaction subject having the representative subject feature from target content interaction subjects corresponding to a second object approval stage as a to-be-pushed subject, the object approval degree of the second object approval stage being lower than the object approval degree of the first object approval stage; and a content pushing module, configured to push a pushing content corresponding to the target pushing object to a terminal of the to-be-pushed subject.

[0017] A computer device comprises a memory and a processor, the memory stores a computer program, and the processor implements the steps in the content pushing method when executing the computer program.

[0018] A computer readable storage medium stores a computer program, and the computer program implements the steps in the content pushing method when executed by a processor.

[0019] In some embodiments, a computer program product or computer program is provided, which comprises 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 the processor executes the computer instructions, so that the computer device executes the steps in each method embodiment.

[0020] The content pushing method, device, computer device and storage medium described above, obtain a target content interaction record set corresponding to a target pushing object and an object approval stage sequence corresponding to the target pushing object, obtain a standard interaction behavior type corresponding to an object approval stage in the object approval stage sequence, since the content interaction record in the target content interaction record set includes an interaction behavior type of a content interaction subject to the target pushing content, the target pushing content is pushing content related to the target pushing object, and the target content interaction record set records the interaction behavior type of the subject to the content related to the target pushing object, since the object approval stage sequence is obtained according to the interaction behavior type corresponding to the object conversion link of the target pushing object, and the object approval stages in the object approval stage sequence are arranged according to the object approval degree, each object approval stage in the object approval stage sequence corresponds to a different object approval degree, therefore, the content interaction subject with which the interaction behavior type matches the standard interaction behavior type is determined according to the target content interaction record set, the target content interaction subject corresponding to the object approval stage is obtained based on the matched content interaction subject, the subject with the interaction behavior to the pushing content of the target pushing object is divided according to the object approval stage, the target content interaction subject of each object approval stage is obtained, and therefore the approval degree of the subject to the object is determined, and therefore the subject can be pushed with the content according to the stage to which the subject belongs, that is, the approval degree of the subject to the object, and the content pushing accuracy is improved. BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 An application environment diagram of the content pushing method in some embodiments;

[0022] Figure 2 A flowchart of the content pushing method in some embodiments;

[0023] Figure 3 A diagram of consumer behavior of a content interaction subject to a brand in some embodiments;

[0024] Figure 4 A diagram of the principle of calculating the number of audiences in different object approval stages in some embodiments;

[0025] Figure 5 A diagram of the principle of using a table corresponding to different interaction behavior types in some embodiments;

[0026] Figure 6 A diagram of the coverage and flow rate corresponding to a target subject feature in some embodiments;

[0027] Figure 7 A flowchart of the content pushing method in some embodiments;

[0028] Figure 8Structure block diagram of content pushing device in some embodiments;

[0029] Figure 9 Structure block diagram of content pushing device in some embodiments;

[0030] Figure 10 Structure block diagram of content pushing device in some embodiments. DETAILED DESCRIPTION

[0031] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application is further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not to limit the present application.

[0032] The content pushing method provided by the present application can be applied in an application environment as shown in the figure. Figure 1 The terminal 102 communicates with the server 104 through a network. The server 104 can be a server for content pushing, for example, a server for advertisement pushing. The terminal 102 can be installed with a client corresponding to the server 104. The server 104 can push content to the terminal 102. The terminal 102 can display the pushed content in the client, for example, the client can display the pushed content in a pushed content display area, which can be an advertisement position for example. The client can include at least one of instant messaging software or shopping software.

[0033] Specifically, the server 104 can obtain a target content interaction record set corresponding to a target push object, a content interaction record in the target content interaction record set includes an interaction behavior type of a content interaction subject to target push content, the target push content is push content related to the target push object, the server 104 can obtain an object approval stage sequence corresponding to the target push object, the object approval stage sequence includes a plurality of object approval stages arranged according to an object approval degree, and the object approval stage sequence is obtained according to an interaction behavior type corresponding to an object conversion link of the target push object. The server 104 can also obtain a standard interaction behavior type corresponding to an object approval stage in the object approval stage sequence, determine a content interaction subject whose interaction behavior type matches the standard interaction behavior type according to the target content interaction record set, and obtain a target content interaction subject corresponding to the object approval stage based on the content interaction subject whose interaction behavior type matches the standard interaction behavior type. The server 104 can determine a push subject to be pushed according to an object approval stage to which the content interaction subject belongs, and push content related to the target push object to the push subject to be pushed.

[0034] As provided in the content push method, the server 104 can store the target content interaction subject corresponding to the object approval stage into the blockchain, for example, the subjects belonging to different object approval stages can generate corresponding query codes, the object approval stages are different, the generated query codes are different, for example, the query codes can be generated by using the object approval stages. The server 104 can store the target content interaction subject and the corresponding query code into the blockchain, so as to improve the security of the target content interaction subject belonging to the object approval stage, and reduce the case of malicious modification or theft of the target content interaction subject belonging to each object approval stage. When it is necessary to obtain the target content interaction subject belonging to the object approval stage, the server 104 can query the corresponding target content interaction subject from the blockchain by using the query code.

[0035] The terminal 102 can be, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers and portable wearable devices, and the server 104 can be implemented by an independent server or a server cluster composed of multiple servers.

[0036] In some embodiments, as shown in Figure 2 A content push method is provided, which can be executed by a terminal or a server, or jointly executed by a terminal and a server. In the embodiments of the present application, the method is applied to a terminal and a server. Figure 1The server 104 in the server 104 is taken as an example for illustration, including the following steps:

[0037] S202, obtaining a target content interaction record set corresponding to a target push object; a content interaction record in the target content interaction record set includes an interaction behavior type of a content interaction subject to target push content, and the target push content is push content related to the target push object.

[0038] The object can represent a single specific thing, for example, it can be a computer of a certain model, and the object can also represent a series of things, for example, it can represent a brand, and the push object refers to an object for pushing. The target push object can be any push object.

[0039] The content can be embodied in the form of text, voice, picture or video, for example, the content can be a song or a video. The push content refers to the content for pushing, for example, the push content can be content that can be shared through the network, and the push content can be an advertisement. The process of pushing the content can be referred to as exposing the content.

[0040] The push object can correspond to one or more push contents, and the push content is used to push the thing represented by the push object, and the push content can include at least one of the things represented by the push object. The target push content is the push content related to the target push object, for example, when the target push object is a brand of cosmetics, the target push content can be content for pushing a cosmetic product in the cosmetic brand, for example, the target push content can be an advertisement of the cosmetic product.

[0041] The subject can be a person, and the subject can also be an organization, for example, it can be an enterprise. The content interaction subject refers to the receiver of the target push content, which can also be referred to as the audience, for example, it can be a consumer.

[0042] The interaction behavior refers to the behavior between the subject and the push content, for example, it can include the behavior of the subject on the Internet to the push content, and the interaction behavior can include at least one of exposure, click, browsing, adding to the shopping cart, collecting, purchasing, downloading, submitting a form, evaluation, liking or forwarding, etc. When the content interaction subject is a consumer, the interaction behavior can also be referred to as consumer behavior. With the promotion of mobile Internet and cloud computing technologies, network services gradually penetrate into real life, and users meet the needs of entertainment, shopping or travel through the network, and also accumulate consumer behavior data for the network platform.

[0043] The interaction behavior type refers to the type of the interaction behavior. The type of the interaction behavior can be classified according to the behavior purpose, the interaction behaviors of one behavior purpose are classified into one interaction behavior type, for example, the purposes of the behaviors of exposure, click, and adding to the shopping cart are different, the purpose of the exposure content is to show the content, which is a passive behavior of the consumer, the purpose of the click content is to view the content, which is an active behavior of the consumer, and the two represent different consumer intentions. Therefore, the exposure, click, and adding to the shopping cart can be classified into different types of interaction behaviors.

[0044] The content interaction record is used to record the interaction behavior, and the content interaction record can include the interaction behavior type of the content interaction subject to the pushed content. Each interaction behavior of the content interaction subject to the pushed content can generate one content interaction record, for example, one record is generated for exposure, and one record is generated for click. The content interaction record can also include the interaction time, which refers to the time when the interaction behavior occurs, for example, the time when the click occurs, and the interaction time can be represented by a time stamp. The content interaction record can include the content interaction subject identifier and the pushed content identifier. The content interaction subject identifier is used to uniquely identify the content interaction subject. When the subject is a consumer, the subject identifier can be a consumer identifier, which is used to uniquely identify the consumer and identify the identity of the consumer. The pushed content identifier is used to uniquely identify the pushed content. The target content interaction record set includes a plurality of target content interaction records. The target content interaction record is the content interaction record corresponding to the target pushed object, and the target content interaction record includes the interaction behavior type of the content interaction subject to the target pushed content. The target content interaction record set can be pre-stored in the server, or it can be a set composed of each target content interaction record obtained by the server from other devices.

[0045] Specifically, the server can store each target content interaction record corresponding to the target pushed object, and the target content interaction record of the target time period can be selected from the stored target content interaction records to form the target content interaction record set. The server can divide the content interaction subject based on the target content interaction record set to obtain the content interaction subject belonging to each object recognition stage. The object recognition stage can be multiple, which means at least two, and the object recognition stage is used to reflect the object recognition degree. Different object recognition stages correspond to different object recognition degrees. The target time period can be pre-set or set according to needs, for example, it can be determined according to the interaction behavior effective time period corresponding to the object recognition stage, for example, the target time period can be a time period including the interaction behavior effective time period, for example, the target time period can include the interaction behavior effective time period of the object recognition stage corresponding to the maximum object recognition degree.

[0046] The interaction behavior effective duration refers to a duration of the interaction behavior of the interaction behavior type as a basis for determining the object recognition stage to which the content interaction subject belongs. The interaction behavior effective duration can be set as needed, for example, and can include at least one of 15 days, 30 days, or 1 year. When the time interval between the current time and the interaction time is less than the interaction behavior effective duration, the interaction behavior of the interaction behavior type can be used as a basis for determining the object recognition stage to which the content interaction subject belongs. For example, the current time is February 1, the standard behavior type corresponding to the "cognition stage" includes clicking, the interaction behavior effective duration corresponding to the "cognition stage" is 15 days, and it is assumed that the subject generated a clicking behavior on January 20. Since the time interval between January 20 and February 1 is less than 15 days, the clicking behavior generated on January 20 can be used to determine the stage to which the content interaction subject belongs. It is assumed that the subject generated a clicking behavior on January 1. Since the time interval between January 1 and February 1 is greater than 15 days, the clicking behavior generated on January 1 is not used as a basis for determining the stage to which the content interaction subject belongs.

[0047] In some embodiments, the server can determine an interaction behavior effective time period corresponding to the object recognition stage based on the interaction behavior effective duration of the object recognition stage, for example, the current time can be used as the end time of the interaction behavior effective time period, and a time interval of 15 days from the current time can be used as the start time of the interaction behavior effective time period, thereby obtaining the interaction behavior effective time period. For example, when the interaction behavior effective duration is 15 days, the interaction behavior effective time period is the last 15 days.

[0048] In some embodiments, when the server obtains the stage determination instruction for the target push object, the server obtains a plurality of target content interaction records of the target push object to form a target content interaction record set. The stage determination instruction is used to indicate the content interaction subjects corresponding to each object recognition stage. The stage determination instruction can be periodically triggered automatically, or triggered by the server according to the progress of the process.

[0049] S204, obtaining an object recognition stage sequence corresponding to the target push object, the object recognition stage sequence including a plurality of object recognition stages arranged according to object recognition degrees, and the object recognition stage sequence being obtained according to an interaction behavior type corresponding to an object conversion link of the target push object.

[0050] The object recognition stage sequence includes a plurality of object recognition stages, and each object recognition stage corresponds to an object recognition degree. The object recognition degree reflects the degree of recognition of the content interaction subject to the push object. The greater the object recognition degree, the greater the possibility of the content interaction subject purchasing or downloading the related product of the push object. The object recognition stages in the object recognition stage sequence can be arranged in order of increasing object recognition degree, and the smaller the object recognition degree, the higher the ranking in the object recognition stage sequence.

[0051] The object conversion link is a sequence of interaction behavior types corresponding to a plurality of key interaction behaviors in the process of the content interaction subject becoming the desired subject of the push object. For example, the object conversion link can be a sequence of behaviors such as the audience being exposed to an advertisement, clicking on the advertisement, purchasing a product pushed by the advertisement, or downloading software pushed by the advertisement. The interaction behavior with the greatest object recognition degree in the object conversion link is used to indicate that the subject has been converted into the desired subject. The desired subject can be set as needed. For example, the desired subject can be a subject who purchases a product of the push object or a subject who downloads software corresponding to the push object. For example, for a commodity advertisement, the purpose of pushing the advertisement is to hope that the user purchases the commodity in the advertisement. Therefore, the interaction behavior with the greatest object recognition degree in the object conversion link is a behavior used to indicate the purchase of the commodity, such as “transaction”. The key interaction behavior can include behaviors that are essential in the process of becoming the desired subject. The key interaction behavior can include at least one of exposure, clicking, or downloading. The interaction behavior types in the object conversion link can be arranged in order of increasing conversion progress. The greater the conversion progress, the higher the ranking of the interaction behavior type in the object conversion link. When the subject produces a behavior of the interaction behavior type ranked last in the object conversion link, it indicates that the subject has been converted into the desired subject. The conversion progress reflects the possibility of the content interaction subject becoming the desired subject after producing the behavior of the interaction behavior type. The conversion progress can be expressed as a percentage, such as 30%. The greater the conversion progress, the greater the possibility of becoming the desired subject. In the object conversion link, the production of the interaction behavior corresponding to the interaction behavior type after the other can depend on the interaction behavior corresponding to the interaction behavior type before the other. For example, the interaction behavior corresponding to the interaction behavior type after the other can only be produced on the premise that the interaction behavior corresponding to the interaction behavior type before the other is produced. The object conversion link can be a key conversion link in a marketing conversion link. The marketing conversion link includes each link in the process of marketing a brand. The key conversion link can be a link composed of key links in the marketing conversion link.

[0052] Since the types of interaction behaviors in the object conversion link are arranged according to the conversion progress, the larger the conversion progress is, the later the sorting in the object conversion link is, and therefore the subject of the behavior of the interaction behavior type arranged at the later in the object conversion link also produces the behavior of the interaction behavior type arranged at the earlier in the object conversion link. For example, the object conversion link is “advertisement exposure→advertisement click→game software download→game software activation→consumer registration→consumer payment”, if the subject produces the behavior of “advertisement click”, the subject necessarily also produces the behavior of “advertisement exposure”.

[0053] Specifically, the server can store the object conversion link corresponding to each target push object, for example, the server can store the marketing conversion link corresponding to each brand. The server can obtain the object conversion link corresponding to the target push object from the stored object conversion links as the target object conversion link according to the target push object, and determine the object approval stage sequence corresponding to the target push object based on the target object conversion link, for example, the server can obtain the object approval stage sequence by dividing the object conversion link.

[0054] In some embodiments, the server can classify the push objects to obtain the push objects corresponding to each push object type, the push object type refers to the type to which the push object belongs, for example, the push objects of the same industry can be classified into one type, and the push objects of different industries can be classified into different types. The industry can include at least one of the automobile industry or the game industry. The server can determine the object conversion link according to the push object type, for example, different push object types can correspond to different object conversion links, for example, different brands in different industries correspond to different marketing conversion links. For example, when the push object is a brand belonging to the automobile industry, the object conversion link can be the key conversion link of the automobile industry “advertisement exposure→advertisement click→consumer leaves a sales clue→sales consultant telephone confirmation→sales consultant invites consumer to the store→consumer test drive→transaction”, when the push object is a brand belonging to the game industry, the object conversion link can be the key conversion link of the game industry “advertisement exposure→advertisement click→game software download→game software activation→consumer registration→consumer payment”. The server can store multiple object conversion links corresponding to different push object types, the server can determine the push object type to which the target push object belongs as the target push object type, obtain the object conversion link corresponding to the target push object type from the stored multiple object conversion links corresponding to different push object types as the target object conversion link, and determine the object approval stage sequence corresponding to the target push object based on the target object conversion link.

[0055] In some embodiments, the server can store a sequence of object recognition stages corresponding to each object conversion link respectively, and the server can obtain the sequence of object recognition stages corresponding to the target conversion link as the sequence of object recognition stages corresponding to the target push object, from the stored sequences of object recognition stages corresponding to each object conversion link, according to the target conversion link.

[0056] In S206, the standard interaction behavior type corresponding to the object recognition stage in the sequence of object recognition stages is obtained.

[0057] The standard interaction behavior type corresponding to the object recognition stage is the interaction behavior type divided into the object recognition stage. The standard interaction behavior type can be one or more, and the more refers to at least two.

[0058] Specifically, the server can divide the target object conversion link into stages to obtain a sequence of object recognition stages. Each object recognition stage in the sequence of object recognition stages can correspond to a non-overlapping sub-link in the target object conversion link respectively. Each interaction behavior type in the sub-link is the interaction behavior type divided into the object recognition stage. The object recognition stages in the sequence of object recognition stages can be arranged according to the order of the sub-link in the target object conversion link. The earlier the order of the sub-link in the target object conversion link, the earlier the order of the object recognition stage corresponding to the sub-link in the sequence of object recognition stages. For example, the target object conversion link is “ad exposure→ad click→game software download→game software activation→consumer registration→consumer payment”, “ad exposure” is divided into “cognition stage”, “ad click” is divided into “interest stage”, “game software download→game software activation→consumer registration” is divided into “potential customer stage”, and “consumer payment” is divided into “player stage”. The sequence of object recognition stages can be represented as “cognition stage→interest stage→potential customer stage→player stage”. The standard interaction behavior type corresponding to the object recognition stage is the interaction behavior type divided into the object recognition stage. For example, the standard interaction behavior type corresponding to the “cognition stage” is “ad exposure”, and the standard interaction behavior type corresponding to the “potential customer stage” includes “game software download”, “game software activation”, and “consumer registration”. The stage can also be referred to as the state, and the “cognition stage” can also be referred to as the “cognition state”.

[0059] In some embodiments, the server can obtain the object recognition stage from the sequence of object recognition stages in sequence, for example, can obtain one object recognition stage from the sequence of object recognition stages at a time according to the order of the object recognition degree from large to small, and obtain the standard interaction behavior type corresponding to the object recognition stage.

[0060] S208, determine the content interaction subject whose interaction behavior type matches the standard interaction behavior type according to the target content interaction record set, and obtain the target content interaction subject corresponding to the object recognition stage based on the content interaction subject whose interaction behavior type matches the standard interaction behavior type.

[0061] The target content interaction subject corresponding to the object recognition stage refers to the content interaction subject in the object recognition stage among the content interaction subjects included in the target content interaction record set. A content interaction subject belongs to only one object recognition stage at the same time, for example, a consumer is in only one state at the same time. The object recognition stage to which a content interaction subject belongs can be transformed from a low stage priority to a high stage priority, for example, a content interaction subject in the “cognition stage” can be transformed to the “interest stage”, and the object recognition stage to which a content interaction subject belongs can also disappear, that is, the content interaction subject does not belong to any object recognition stage, for example, a content interaction subject in the “interest stage” can be transformed to a subject not belonging to any object recognition stage. When a content interaction subject in the object recognition stage with a high stage priority generates an interaction behavior of the standard interaction behavior type corresponding to the object recognition stage with a low stage priority, the content interaction subject does not change the object recognition stage, for example, a content interaction subject in the “interest stage” generates a click behavior, the click behavior belongs to the “cognition stage”, and the content interaction subject is still in the “interest stage” and will not be transformed to the “cognition stage”.

[0062] Specifically, the server can compare the interaction behavior type in the target content interaction record with the standard interaction behavior type, determine that the interaction behavior type matches the standard interaction behavior type when the comparison is consistent, obtain the target content interaction subject corresponding to the object recognition stage based on the content interaction subject in the target content interaction record corresponding to the interaction behavior type with consistent comparison, for example, the content interaction subject in the target content interaction record corresponding to the interaction behavior type with consistent comparison can be used as the target content interaction subject corresponding to the object recognition stage, or the target content interaction record corresponding to the interaction behavior type with consistent comparison can be further screened to obtain the target content interaction subject corresponding to the object recognition stage, for example, the object recognition stage can be screened by using the interaction behavior effective time length corresponding to the object recognition stage, for example, the server can determine the interaction behavior effective time period corresponding to the object recognition stage according to the interaction behavior effective time length corresponding to the object recognition stage, obtain the target content interaction record with the interaction time belonging to the interaction behavior effective time period from the target content interaction record corresponding to the interaction behavior type with consistent comparison, and use the content interaction subject corresponding to the target content interaction record with the interaction time belonging to the interaction behavior effective time period as the target content interaction subject corresponding to the object recognition stage.

[0063] The interaction behavior effective duration refers to a duration of the interaction behavior of the interaction behavior type as a basis for determining the object recognition stage to which the content interaction subject belongs. The interaction behavior effective duration can be set as needed, for example, and can include at least one of 15 days, 30 days, or 1 year. When the time interval between the current time and the interaction time is less than the interaction behavior effective duration, the interaction behavior of the interaction behavior type can be used as a basis for determining the object recognition stage to which the content interaction subject belongs. For example, the current time is February 1, the standard behavior type corresponding to the "cognition stage" includes clicking, and the interaction behavior effective duration corresponding to the "cognition stage" is the last 15 days. Assuming that the subject generated a clicking behavior on January 20, since the time interval between January 20 and February 1 is less than 15 days, the clicking behavior generated on January 20 can be used to determine the stage to which the content interaction subject belongs. Assuming that the subject generated a clicking behavior on January 1, since the time interval between January 1 and February 1 is greater than 15 days, the clicking behavior generated on January 1 is not used as a basis for determining the stage to which the content interaction subject belongs.

[0064] The start time of the interaction behavior effective time period is the current time, and the end time of the interaction behavior effective time period is a time at which the time interval between the current time is equal to the interaction behavior effective duration. For example, when the interaction behavior effective duration is 15 days, the interaction behavior effective time period is the last 15 days.

[0065] In some embodiments, the server can obtain, based on the content interaction subject whose interaction behavior type matches the standard interaction behavior type of the object recognition stage, the target content interaction subject corresponding to the object recognition stage

[0066] In some embodiments, the server can obtain, from the target content interaction record set, a target content interaction record in which the interaction time belongs to the interaction behavior effective time period, as an effective content interaction record, compare the interaction behavior type in the effective content interaction record with the standard interaction behavior type of the object recognition stage, and obtain, based on the content interaction subject in the effective content interaction record corresponding to the interaction behavior type that matches the comparison, the target content interaction subject corresponding to the object recognition stage. For example, the content interaction subject in the effective content interaction record corresponding to the interaction behavior type that matches the comparison can be used as the target content interaction subject corresponding to the object recognition stage, or the effective content interaction record corresponding to the interaction behavior type that matches the comparison can be further filtered, and the content interaction subject corresponding to the filtered effective content interaction record can be used as the target content interaction subject corresponding to the object recognition stage.

[0067] In some embodiments, the server can screen the frequency of interaction behaviors to obtain the target content interaction subject corresponding to the object recognition stage. The frequency of interaction behaviors can be the number of the same type of interaction behaviors in a period of time. The frequency of interaction behaviors can reflect the attitude of the consumer to the brand in the period of time. For example, when the number of exposures in a period of time exceeds a set number, but the consumer does not purchase the brand product, it can reflect that the consumer's recognition degree of the brand is low. Specifically, the server can obtain each target content interaction record corresponding to the same content interaction subject from the target content interaction record set, and form a sub-interaction record set corresponding to the content interaction subject. The server compares the interaction behavior type corresponding to the content interaction record in the sub-interaction record set with the standard interaction behavior type corresponding to the object recognition stage, classifies each content interaction record in the sub-interaction record set according to the consistent comparison of the interaction behavior type, obtains a classified interaction set corresponding to each interaction behavior type, and counts the number of content interaction records in the classified interaction set to obtain the number of interaction records corresponding to the classified interaction set. The server compares the number of interaction records with the record number threshold value, and when each number of interaction records is less than the record number threshold value, the content interaction subject is regarded as the target content interaction subject corresponding to the object recognition stage, otherwise, it is confirmed that the content interaction subject does not belong to the object recognition stage.

[0068] In the above content pushing method, the target content interaction record set corresponding to the target pushing object and the object recognition stage sequence corresponding to the target pushing object are obtained, and the standard interaction behavior type corresponding to the object recognition stage in the object recognition stage sequence is obtained. Since the content interaction record in the target content interaction record set includes the interaction behavior type of the content interaction subject to the target pushing content, the target pushing content is the pushing content related to the target pushing object, and the target content interaction record set records the interaction behavior type of the subject to the content related to the target pushing object. Since the object recognition stage sequence is obtained according to the interaction behavior type corresponding to the object conversion link of the target pushing object, and the object recognition stages in the object recognition stage sequence are arranged according to the object recognition degree, each object recognition stage in the object recognition stage sequence corresponds to a different object recognition degree. Therefore, according to the target content interaction record set, the content interaction subject with the matching interaction behavior type and the standard interaction behavior type is determined, the target content interaction subject corresponding to the object recognition stage is obtained based on the matched content interaction subject, the subject with the interaction behavior to the pushing content of the target pushing object is divided according to the object recognition stage, the target content interaction subject of each object recognition stage is obtained, and thus the recognition degree of the subject to the object is determined. Therefore, the content can be pushed to the subject according to the stage to which the subject belongs, i.e., the recognition degree of the subject to the object, and the content pushing accuracy is improved.

[0069] In some embodiments, determining the content interaction subject whose interaction behavior type matches the standard interaction behavior type according to the target content interaction record set, and obtaining the target content interaction subject corresponding to the object recognition stage based on the content interaction subject whose interaction behavior type matches the standard interaction behavior type, comprises: obtaining the content interaction record corresponding to each content interaction subject from the target content interaction record set, to form a sub-interaction record set corresponding to the content interaction subject; and when the interaction behavior type in the sub-interaction record set matches the standard interaction behavior type, regarding the content interaction subject as the target content interaction subject corresponding to the object recognition stage.

[0070] The sub-interaction record set is a set of content interaction records corresponding to the same content interaction subject in the target content interaction record set. Each content interaction subject can correspond to a sub-interaction record set.

[0071] Specifically, the server can compare the interaction behavior type of the content interaction record in the sub-interaction record set with the standard interaction behavior type, and when the comparison is consistent, regard the content interaction subject as the target content interaction subject corresponding to the object recognition stage.

[0072] In some embodiments, the server can obtain the interaction behavior effective time length corresponding to the object recognition stage, determine the interaction behavior effective time period corresponding to the object recognition stage according to the interaction behavior effective time length, obtain the content interaction record whose interaction time belongs to the interaction behavior effective time period from the sub-interaction record set, compare the interaction behavior type of the content interaction record whose interaction time belongs to the interaction behavior effective time period with the standard interaction behavior type, and when the comparison is consistent, regard the content interaction subject as the target content interaction subject corresponding to the object recognition stage. For example, the server can arrange the content interaction records in the sub-interaction record set in the order of interaction time from early to late to obtain a sub-interaction record sequence, and the earlier the interaction time, the higher the ranking of the content interaction record in the sub-interaction record sequence. The server obtains the content interaction record whose interaction time belongs to the interaction behavior effective time period from the sub-interaction record sequence, compares the interaction behavior type of the content interaction record whose interaction time belongs to the interaction behavior effective time period with the standard interaction behavior type, and when the comparison is consistent, regards the content interaction subject as the target content interaction subject corresponding to the object recognition stage.

[0073] In some embodiments, since the object recognition stages are multiple, the server can sequentially obtain an object recognition stage from the sequence of object recognition stages, compare the interaction behavior type in the sub-interaction record set with the standard interaction behavior type of the object recognition stage, and take the object recognition stage with the largest object recognition degree in the object recognition stage corresponding to the standard interaction behavior type that is consistent with the comparison as the object recognition stage to which the content interaction subject belongs. The server can obtain the object recognition stage from the object recognition stage in descending order of the object recognition degree, compare the interaction behavior type in the sub-interaction record set with the standard interaction behavior type of the object recognition stage, and when the comparison is consistent, stop obtaining the object recognition stage from the sequence of object recognition stages, and take the content interaction subject as the target content interaction subject of the object recognition stage corresponding to the standard interaction behavior type that is consistent with the comparison.

[0074] In the present embodiment, when the interaction behavior type in the sub-interaction record set matches the standard interaction behavior type, it can be determined that the content interaction subject has generated the interaction behavior of the standard interaction behavior type for the target push content, and the content interaction subject can be taken as the target content interaction subject corresponding to the object recognition stage, so that it can be determined that the target content interaction subject corresponding to the object recognition stage has generated the corresponding standard interaction behavior type for the target push content.

[0075] In some embodiments, the object recognition stage corresponds to a stage priority, and the stage priority is in a positive correlation with the object recognition degree corresponding to the object recognition stage. The standard interaction behavior type corresponding to the object recognition stage in the sequence of object recognition stages includes: sequentially obtaining the current object recognition stage from the sequence of object recognition stages in the order of the stage priority, and obtaining the current standard interaction behavior type corresponding to the current object recognition stage. When the interaction behavior type in the sub-interaction record set matches the standard interaction behavior type, taking the content interaction subject as the target content interaction subject corresponding to the object recognition stage includes: when the interaction behavior type in the sub-interaction record set matches the current standard interaction behavior type, taking the content interaction subject as the target content interaction subject corresponding to the current object recognition stage; otherwise, entering the step of sequentially obtaining the current object recognition stage from the sequence of object recognition stages in the order of the stage priority, to update the current object recognition stage, until the interaction behavior type in the sub-interaction record set matches the current standard interaction behavior type.

[0076] The object recognition stage sequence includes a plurality of object recognition stages, each object recognition stage can correspond to a stage priority, the stage priority is used to reflect the size of the object recognition degree, the stage priority and the object recognition degree are in a positive correlation relationship, the higher the stage priority, the greater the object recognition degree, for example, the object recognition stage sequence is "cognition stage → interest stage → conversion stage", since the object recognition degree of the interest stage is greater than that of the cognition stage, the stage priority of the interest stage is higher than that of the cognition stage, and since the object recognition degree of the conversion stage is greater than that of the interest stage, the stage priority of the conversion stage is higher than that of the interest stage. The stage priority can also be referred to as the state priority.

[0077] The positive correlation relationship refers to: under the condition that other conditions remain unchanged, the variation direction of two variables is the same, when one variable changes from large to small, the other variable also changes from large to small. It can be understood that the positive correlation relationship here refers to the consistent direction of change, but it does not require that when one variable changes a little, the other variable must also change. For example, it can be set that when variable a is 10 to 20, variable b is 100, when variable a is 20 to 30, variable b is 120. In this way, the variation direction of a and b is that when a becomes larger, b also becomes larger. But within the range of a from 10 to 20, b can be unchanged.

[0078] Each object recognition stage in the object recognition stage sequence can be arranged from low to high according to the stage priority, the lower the stage priority, the earlier the object recognition stage is arranged in the object recognition stage sequence. The current object recognition stage can be any object recognition stage in the object recognition stage sequence. The current standard interaction behavior type refers to the standard interaction behavior type corresponding to the current object recognition stage.

[0079] Specifically, when the current object recognition stage is obtained from the object recognition stage sequence, it can be obtained according to the stage priority, for example, the object recognition stages are obtained in the order from high to low according to the stage priority, as the current object recognition stage.

[0080] In some embodiments, the server obtains an object recognition stage from the sequence of object recognition stages in order of high to low stage priority as a current object recognition stage, obtains a standard interaction behavior type corresponding to the current object recognition stage as a current standard interaction behavior type, compares the interaction behavior type in the sub-interaction record set with the current standard interaction behavior type, when the comparison is consistent, takes the content interaction subject as a target content interaction subject corresponding to the current object recognition stage, and when the comparison is inconsistent, obtains an object recognition stage with a lower stage priority than the current object recognition stage from the sequence of object recognition stages in order of high to low stage priority as a current object recognition stage in the next round, i.e., when the comparison is inconsistent, returns the step of sequentially obtaining the current object recognition stage from the sequence of object recognition stages in order of stage priority.

[0081] In some embodiments, the interaction behavior type in the sub-interaction record set is compared with the current standard interaction behavior type, when the comparison is inconsistent and the current object recognition stage is an object recognition stage corresponding to the lowest stage priority in the sequence of object recognition stages, the content interaction subject is determined to be a subject in a preset stage, the preset stage being different from each object recognition stage in the sequence of object recognition stages, i.e., the content interaction subject is determined not to belong to each object recognition stage in the sequence of object recognition stages.

[0082] In the present embodiment, the current object recognition stage is sequentially obtained from the sequence of object recognition stages in order of stage priority, the current standard interaction behavior type corresponding to the current object recognition stage is obtained, when the interaction behavior type in the sub-interaction record set matches the current standard interaction behavior type, the content interaction subject is taken as a target content interaction subject corresponding to the current object recognition stage, otherwise, the step of sequentially obtaining the current object recognition stage from the sequence of object recognition stages in order of stage priority is entered to update the current object recognition stage until the interaction behavior type in the sub-interaction record set matches the current standard interaction behavior type, so that the content interaction subject can be quickly determined to belong to which object recognition stage.

[0083] In some embodiments, obtaining the current standard interaction behavior type corresponding to the current object recognition stage includes: obtaining the current standard interaction behavior type corresponding to the current object recognition stage and an interaction behavior valid duration corresponding to the current object recognition stage; and when the interaction behavior type in the sub-interaction record set matches the standard interaction behavior type, taking the content interaction subject as the target content interaction subject corresponding to the object recognition stage includes: obtaining an interaction behavior valid time period corresponding to the current object recognition stage based on the interaction behavior valid duration corresponding to the current object recognition stage; and when there is a content interaction record in the sub-interaction record set, the interaction behavior type of which matches the current standard interaction behavior type and the interaction time of which belongs to the interaction behavior valid time period, taking the content interaction subject as the target content interaction subject corresponding to the current object recognition stage.

[0084] Specifically, the server can take the current time as the start time of the interaction behavior valid time period, take the time interval between the current time and the time equal to the interaction behavior valid duration as the end time of the interaction behavior valid time period, and obtain the interaction behavior valid time period.

[0085] In some embodiments, the server can compare the interaction behavior type corresponding to the content interaction record in the sub-interaction record set with the current standard interaction behavior type, compare the interaction time of the content interaction record consistent with the comparison with the interaction behavior valid time period, and when it is determined that the interaction time belongs to the interaction behavior valid time period, take the content interaction subject as the target content interaction subject corresponding to the current object recognition stage.

[0086] In some embodiments, the server can obtain, from the sub-interaction record set, each content interaction record with the interaction time belonging to the interaction behavior valid time period of the current object recognition stage, as an effective content interaction record, and when there is a content interaction record with the interaction behavior type matching the current standard interaction behavior type in each effective content interaction record, take the content interaction subject as the target content interaction subject corresponding to the current object recognition stage.

[0087] For example, as shown in FIG. 6, the server can obtain the interaction behavior valid time period of the current object recognition stage based on the current time and the interaction behavior valid duration corresponding to the current object recognition stage, and compare the interaction time of each content interaction record in the sub-interaction record set with the interaction behavior valid time period. Figure 3As shown, the figure shows the consumer behavior caused by the brand's advertisement and the content in the advertisement of the content interaction subject in a period of time. The consumer behavior in the figure can be the consumer behavior of the audience in a marketing (i.e., S marketing in the figure) to the brand's advertisement and the content in the advertisement. It is assumed that the object recognition stage sequence corresponding to the brand is "cognition stage → interest stage → conversion stage", "exposure" belongs to the cognition stage, "click" belongs to the interest stage, and "purchase" belongs to the conversion stage. The effective time length of the interaction behavior corresponding to the cognition stage is 15 days, the effective time length of the interaction behavior corresponding to the interest stage is 15 days, and the effective time length of the interaction behavior corresponding to the conversion stage is 30 days. For audience A, there is no purchase behavior in the last 30 days, so audience A does not belong to the conversion stage. There is no click behavior in the last 15 days, so audience A does not belong to the interest stage. There is exposure in the last 15 days, so audience A belongs to the cognition stage. For audience B, there is no purchase behavior in the last 30 days, there is no click behavior in the last 15 days, and there is no exposure in the last 15 days. Therefore, audience B does not belong to any stage in the object recognition stage sequence. For audience C, there is no purchase behavior in the last 30 days, and there is click behavior in the last 15 days. Therefore, audience C belongs to the interest stage.

[0088] In some embodiments, the server can count the number of target content interaction subjects corresponding to each object recognition stage respectively, and send the counted number to the terminal. The terminal can display the number of subjects in each object recognition stage. The number of subjects can be different due to different effective time lengths of interaction behavior. As shown, Figure 4 As shown, the figure shows the number of audiences in different object recognition stages from the perspective of the enterprise. The enterprise carried out two advertisement marketing activities in January and February. The first marketing activity was from January 1st to January 10th, and the second marketing activity was from February 15th to February 20th. Figure 4 In (a) in the figure, the time effectiveness of the cognition state and the interest state is set to the last 15 days, and the time effectiveness of the conversion state is set to the last 30 days, Figure 4In (b), the recency of the cognitive state and the interest state is set as the last 15 days, and the recency of the conversion state is set as the last 1 year. As can be seen from (a) and (b), the number of audiences in the cognitive state in (a) and (b) is consistent, the number of audiences in the interest state is consistent, and the number of audiences in the conversion state is inconsistent. Since the recency of the conversion state in (a) is the last 30 days, the state of the audiences in the conversion stage generated by the first wave of delivery disappears at the second wave of delivery, that is, the audiences in the conversion state generated by the first wave of delivery are no longer in the conversion state at the second wave of delivery. Therefore, at the second wave of delivery, the number of audiences in the conversion state is not affected by the number of audiences in the conversion stage generated by the first wave of delivery. Since the recency of the conversion state in (b) is the last 1 year, the audiences in the conversion stage generated by the first wave of delivery are still in the conversion state at the second wave of delivery. Therefore, at the second wave of delivery, the number of audiences in the conversion state is increased based on the number of audiences in the conversion stage generated by the first wave of delivery. Therefore, the number of audiences in the conversion state in (a) and (b) is inconsistent. The recency of the last 30 days means that the interaction behavior is valid for 30 days, that is, the interaction behavior is valid for the last 30 days. The recency of the last 15 days means that the interaction behavior is valid for 15 days, that is, the interaction behavior is valid for the last 15 days.

[0089] In this embodiment, the interaction behavior effective time period corresponding to the current object recognition stage is obtained based on the interaction behavior effective time length corresponding to the current object recognition stage. When there is a content interaction record in the sub-interaction record set, the interaction behavior type matches the current standard interaction behavior type, and the interaction time belongs to the interaction behavior effective time period, the content interaction subject is taken as the target content interaction subject corresponding to the current object recognition stage, thereby improving the accuracy of determining the subject of each stage.

[0090] In some embodiments, the content interaction subject whose interaction behavior type matches the standard interaction behavior type is determined according to the target content interaction record set, and the target content interaction subject corresponding to the object recognition stage is obtained based on the content interaction subject whose interaction behavior type matches the standard interaction behavior type, which includes: dividing the target content interaction record set according to the interaction behavior type to obtain an interaction record sub-set corresponding to each interaction behavior type; obtaining the interaction record sub-set corresponding to the standard interaction behavior type, and obtaining the target content interaction subject corresponding to the object recognition stage based on the interaction record sub-set corresponding to the standard interaction behavior type.

[0091] In some embodiments, the content interaction record in the same interaction record sub-set corresponds to the same interaction behavior type.

[0092] Specifically, the server can divide the target content interaction record set according to interaction behavior types, obtain interaction record sub-sets respectively corresponding to each interaction behavior type, select an interaction record sub-set corresponding to a standard interaction behavior type from each interaction record sub-set as a standard interaction record sub-set, and obtain target content interaction subjects corresponding to the object approval stage according to content interaction subjects corresponding to content interaction records in the standard interaction record sub-set. For example, the server can take content interaction subjects corresponding to content interaction records in the standard interaction record sub-set as target content interaction subjects corresponding to the object approval stage, or the server can filter the standard interaction record sub-set, determine target content interaction subjects corresponding to the object approval stage according to content interaction records remaining after the filtering.

[0093] In some embodiments, the server can store content interaction records of the same interaction behavior type in the same table. For example, the server can obtain content interaction records with an interaction behavior type of “click”, and store the content interaction records with the interaction behavior type of “click” in the same table. The server can first obtain a target content interaction record set, and then divide the target content interaction record set to obtain interaction record sub-sets respectively corresponding to each interaction behavior type, and store the same interaction record sub-set in the same table. The server can also store content interaction records in corresponding tables according to interaction behavior types corresponding to the content interaction records when obtaining the content interaction records. The target content interaction record set can include each content interaction record stored in the table corresponding to each interaction behavior type. When storing content interaction records of the same interaction behavior type in the table, content interaction records corresponding to the same content interaction subject can be obtained, the content interaction records corresponding to the same content interaction subject can be merged into one record to obtain a merged interaction record, and the merged interaction record can be stored, thereby reducing the utilization rate of the storage space.

[0094] For example, Figure 5As shown, a table corresponding to different interaction behavior types is shown, wherein the "exposure temporary table" is a table corresponding to the interaction behavior type "exposure", the "click temporary table" is a table corresponding to the interaction behavior type "click", and the "conversion temporary table" is a table corresponding to the interaction behavior type "conversion". The subject that produces the "exposure" behavior to the target push content and the interaction time are stored in the "exposure temporary table". The subject that produces the "click" behavior to the target push content and the interaction time are stored in the "click temporary table". The subject that produces the "conversion" behavior to the target push content and the interaction time are stored in the "conversion temporary table". In the case where the number of content interaction subjects is large, the temporary intermediate table is used to divide the whole task into multiple temporary tables for parallel processing, thereby improving the task execution efficiency, i.e., improving the efficiency of determining the stage to which the content interaction subject belongs, i.e., improving the efficiency of determining the subject in each stage.

[0095] In this embodiment, the target content interaction subject corresponding to the object recognition stage is obtained based on the interaction record sub-set corresponding to the standard interaction behavior type, and multiple target content interaction subjects can be obtained at the same time, thereby improving the efficiency of obtaining the subject corresponding to the stage.

[0096] In some embodiments, the object recognition stage corresponds to a stage priority, and the stage priority is in a positive correlation with the object recognition degree corresponding to the object recognition stage. Obtaining the target content interaction subject corresponding to the object recognition stage based on the interaction record sub-set corresponding to the standard interaction behavior type includes: obtaining a current interaction record sub-set corresponding to the standard interaction behavior type corresponding to the current object recognition stage, and grouping the content interaction subjects in the current interaction record sub-set to form a current content interaction subject set; obtaining a priority interaction record sub-set corresponding to the standard interaction behavior type corresponding to the priority object recognition stage, and grouping the content interaction subjects in the priority interaction record sub-set to form a priority content interaction subject set, the priority object recognition stage being an object recognition stage in the object recognition stage sequence and having a stage priority higher than that of the current object recognition stage; and filtering out the priority content interaction subject set from the current content interaction subject set to obtain the target content interaction subject corresponding to the current object recognition stage.

[0097] The current object recognition stage and the priority object recognition stage are different object recognition stages in the object recognition stage sequence, and the stage priority of the priority object recognition stage is higher than that of the current object recognition stage. The priority object recognition stage corresponding to the current object recognition stage can be one or more, and the more means at least two, for example, the object recognition stage sequence is "cognition stage→interest stage→conversion stage", if the current object recognition stage is "cognition stage", the priority object recognition stage includes "interest stage" and "conversion stage", and if the current object recognition stage is "interest stage", the priority object recognition stage is "conversion stage".

[0098] The interaction behavior type of the content interaction record in the current interaction record sub-set is consistent with the standard interaction behavior type corresponding to the current object recognition stage. The interaction behavior type of the content interaction record in the priority interaction record sub-set is consistent with the standard interaction behavior type corresponding to the priority object recognition stage.

[0099] The current content interaction subject set is a set composed of each content interaction subject in the current interaction record sub-set, and the priority content interaction subject set is a set composed of each content interaction subject in the priority interaction record sub-set.

[0100] Specifically, the server can compare the content interaction subject in the priority content interaction subject set with the content interaction subject in the current content interaction subject set, filter out the content interaction subject consistent in comparison from the current content interaction subject set when the comparison is consistent, and take the content interaction subject left after filtering in the current content interaction subject set as the target content interaction subject corresponding to the current object recognition stage.

[0101] In the embodiment, since the subject with the behavior of high stage priority also has the behavior of low stage priority, the subject in the current content interaction subject set can exist in the priority content interaction subject set, that is, the same subject can exist in both the current content interaction subject set and the priority content interaction subject set. The target content interaction subject corresponding to the current object recognition stage is obtained by filtering out the priority content interaction subject set from the current content interaction subject set, which can filter out the subject belonging to the priority object recognition stage from the current content interaction subject set, and improve the accuracy of obtaining the subject corresponding to the stage.

[0102] In some embodiments, the step of dividing to obtain the object recognition stage sequence comprises: obtaining a target object conversion link, the target object conversion link comprising a plurality of interaction behavior types arranged according to conversion progress; wherein the interaction behavior type is divided according to the behavior purpose; obtaining the conversion possibility corresponding to each interaction behavior type; and performing stage division on the target object conversion link based on the conversion possibility corresponding to each interaction behavior type to obtain the object recognition stage sequence; wherein the standard interaction behavior type corresponding to the object recognition stage is the interaction behavior type divided to the object recognition stage.

[0103] The target object conversion link refers to the object conversion link corresponding to the target push object. The conversion possibility is used to reflect the possibility of the content interaction subject performing the interaction behavior type to become the expected subject, and the greater the conversion possibility, the greater the possibility of becoming the expected subject.

[0104] The object recognition stages in the sequence of object recognition stages can correspond to non-overlapping sub-chains in the target object conversion chain respectively, and each interaction behavior type in the sub-chain is the interaction behavior type divided into the object recognition stage. The object recognition stages in the sequence of object recognition stages can be arranged according to the order of the sub-chains in the target object conversion chain. The earlier the order of the sub-chain in the target object conversion chain, the earlier the order of the object recognition stage corresponding to the sub-chain in the sequence of object recognition stages. For example, the target object conversion chain is “ad exposure→ad click→game software download→game software activation→consumer registration→consumer payment”, “ad exposure” is divided into the “cognition stage”, “ad click” is divided into the “interest stage”, “game software download→game software activation→consumer registration” is divided into the “potential customer stage”, and “consumer payment” is divided into the “player stage”. The sequence of object recognition stages can be represented as “cognition stage→interest stage→potential customer stage→player stage”.

[0105] The standard interaction behavior type corresponding to the object recognition stage is the interaction behavior type divided into the object recognition stage. For example, the standard interaction behavior type corresponding to the “cognition stage” is “ad exposure”, and the standard interaction behavior type corresponding to the “potential customer stage” includes “game software download”, “game software activation”, and “consumer registration”.

[0106] Specifically, the server can calculate the difference between the conversion probabilities corresponding to the interaction behavior types, perform stage division on the target object conversion chain according to the calculated probability difference, and obtain the sequence of object recognition stages. For example, the server can calculate the probability difference between the conversion probabilities of adjacent interaction behavior types in the target object conversion chain, where the probability difference refers to the difference between the conversion probabilities. The target object conversion chain is divided into stages according to each probability difference, and the sequence of object recognition stages is obtained.

[0107] In this embodiment, the target object conversion chain is divided into stages based on the conversion probabilities corresponding to each interaction behavior type, and the sequence of object recognition stages is obtained. Therefore, interaction behavior types with smaller conversion probability differences can be divided into one stage, and interaction behavior types with larger conversion probability differences can be divided into different stages. Thus, the interaction behavior types corresponding to different object recognition stages have large differences in conversion probability, so that the likelihood of content interaction subjects belonging to different object recognition stages being converted into desired subjects is different, and the accuracy of stage division is improved.

[0108] In some embodiments, the target object conversion link is divided into stages based on the conversion likelihoods of the respective interaction behavior types, and the object approval stage sequence is obtained by: based on the conversion likelihoods of the respective interaction behavior types, calculating the likelihood difference between the conversion likelihoods of adjacent interaction behavior types in the target object conversion link; and dividing the adjacent interaction behavior types with a likelihood difference less than a difference threshold into the same object approval stage.

[0109] The difference threshold can be preset or calculated according to the respective likelihood differences. For example, the server can perform mean operation on the respective likelihood differences to obtain a difference average, and obtain the difference threshold based on the difference average. For example, the difference average can be taken as the difference threshold, or a difference coefficient can be obtained and multiplied with the difference average to obtain the difference threshold. The difference coefficient can be a positive number greater than 1, for example, 1.2.

[0110] Specifically, the server can obtain the conversion likelihood of the first interaction behavior type as the first conversion likelihood, obtain the conversion likelihood of the second interaction behavior type as the second conversion likelihood, perform difference operation on the first conversion likelihood and the second conversion likelihood, take the result of the difference operation as the likelihood difference, compare the likelihood difference with the difference threshold, and when the likelihood difference is less than the difference threshold, divide the first interaction behavior type and the second interaction behavior type into the same object approval stage, otherwise, divide the first interaction behavior type and the second interaction behavior type into different object approval stages. The first interaction behavior type and the second interaction behavior type are adjacent interaction behavior types in the target object conversion link.

[0111] In some embodiments, the server can obtain, as a current interaction behavior type, an interaction behavior type ranked first in the target object conversion link, obtain a conversion probability corresponding to the current interaction behavior type as a current conversion probability, obtain, as a neighboring interaction behavior type corresponding to the current interaction behavior type, an interaction behavior type backward adjacent to the current interaction behavior type in the target object conversion link, obtain a conversion probability corresponding to the neighboring interaction behavior type as a neighboring conversion probability, perform difference operation on the current conversion probability and the neighboring conversion probability to obtain a current probability difference, compare the current probability difference with a difference threshold, when the current probability difference is less than the difference threshold, divide the current interaction behavior type and the corresponding neighboring interaction behavior type into the same object approval stage, otherwise, divide the current interaction behavior type and the corresponding neighboring interaction behavior type into different object approval stages, take the neighboring interaction behavior type corresponding to the current interaction behavior type as an updated current conversion probability, return to the step of obtaining the conversion probability corresponding to the current interaction behavior type, until the neighboring interaction behavior type corresponding to the current interaction behavior type is an interaction behavior type ranked last.

[0112] For example, assuming that the object conversion link is "ad exposure (0.1) → ad click (0.3) → consumer leaving a sales clue (0.45) → sales consultant telephone confirmation (0.5) → sales consultant inviting consumers to the store (0.7) → consumer test drive (0.8) → deal (1)", with the deal as the final purpose of conversion, the numbers in the brackets are the conversion probabilities corresponding to the interaction behavior types, the conversion probability corresponding to "ad exposure" is 0.1, the conversion probability corresponding to "ad click" is 0.3, the conversion probability corresponding to "consumer leaving a sales clue" is 0.45, the conversion probability corresponding to "sales consultant telephone confirmation" is 0.5, the conversion probability corresponding to "sales consultant inviting consumers to the store" is 0.7, the conversion probability corresponding to "consumer test drive" is 0.8, and the conversion probability corresponding to "deal" is 1, and assuming that the difference threshold is 0.12, since the energy difference between "ad exposure" and "ad click" is 0.2, which is greater than 0.12, "ad exposure" and "ad click" are divided into different object recognition stages, for example, "ad exposure" is divided into the "cognition stage", and "ad click" is divided into the "interest stage", since the energy difference between "consumer leaving a sales clue" and "ad click" is 0.15, which is greater than 0.12, "consumer leaving a sales clue" and "ad click" are divided into different object recognition stages, for example, "consumer leaving a sales clue" is divided into the "fan stage", since the energy difference between "sales consultant telephone confirmation" and "consumer leaving a sales clue" is 0.05, which is less than 0.12, "sales consultant telephone confirmation" is also divided into the "fan stage", and by analogy, "sales consultant inviting consumers to the store" and "consumer test drive" can be divided into the "potential customer stage", and "deal" is divided into the "car friend stage", so that the object recognition stage sequence can be represented as "cognition stage → interest stage → fan stage → potential customer stage → car friend stage".

[0113] In the embodiment, adjacent interaction behavior types with a probability difference less than the difference threshold are divided into the same object recognition stage, which realizes the division of interaction behavior types with a smaller difference in conversion probability into the same object recognition stage, so that the interaction behavior types corresponding to the same object recognition stage have a smaller difference in conversion probability, so that the difference in the probability of the content interaction subject belonging to the same object recognition stage being converted into the expected subject is smaller, and the accuracy of stage division is improved.

[0114] In some embodiments, obtaining the conversion possibility of each interaction behavior type comprises: obtaining a reference content interaction record set corresponding to the target object conversion link, the content interaction record in the reference content interaction record set comprising an interaction behavior type of a reference interaction subject on a reference push content, the reference push content being push content related to a reference push object corresponding to the target object conversion link; counting the number of reference interaction subjects of the same interaction behavior type in the reference content interaction record set to obtain the reference subject number corresponding to the interaction behavior type; and obtaining the conversion possibility corresponding to the interaction behavior type based on the reference subject number corresponding to the interaction behavior type.

[0115] The object conversion link can be divided according to the push object type, and one push object type can correspond to one object conversion link. The reference push object can be a push object belonging to a target push object type, and the target push object type can be a push object type corresponding to the target object conversion link. The reference push object can be a push object corresponding to the target object conversion link. The reference push object can be the same as the target push object, or can be different from the target push object. For example, the target push object type is a cosmetic product, the target push object is a cosmetic product of brand A, and the reference push object is a cosmetic product of brand B. The target push object and the reference push object are different, but both belong to the same push object type, i.e., they both belong to the cosmetic product category.

[0116] The reference push content is push content related to the reference push object, such as an advertisement for a product related to the reference push object. The reference interaction subject refers to the receiver of the reference push content. The reference interaction subject can be the same as or different from the content interaction subject.

[0117] The reference content interaction record set comprises a plurality of content interaction records, and the content interaction record included in the reference content interaction record set is a reference content interaction record. The reference content interaction record can comprise an interaction behavior type of a reference interaction subject on a reference push content.

[0118] Specifically, the server can push the reference push content corresponding to the reference push object to the terminal corresponding to each reference interaction subject in the reference interaction subject set, so that the reference interaction subject interacts with the reference push content through the terminal. The server can record the interaction between the reference interaction subject and the reference push content, generate a reference content interaction record, and form a reference content interaction record set. The reference interaction subject set comprises a plurality of reference interaction subjects.

[0119] In some embodiments, the server can divide the reference content interaction record set according to interaction behavior types, divide reference content interaction records of the same interaction behavior type into the same reference record sub-set, divide reference content interaction records of different interaction behavior types into different reference record sub-sets, obtain reference record sub-sets corresponding to each interaction behavior type respectively, and count the number of reference interaction subjects in the reference record sub-set, and take the counted number as the reference subject number corresponding to the interaction behavior type corresponding to the reference record sub-set. For example, assuming that the reference record sub-set includes 100 records, the 100 records are records generated by 2 reference interaction subjects, for example, 50 records are generated by reference interaction subject A, and 50 records are generated by reference interaction subject B, and the reference subject number is 2.

[0120] In some embodiments, the server can obtain the reference subject number corresponding to the conversion interaction behavior type as the conversion reference subject number, the conversion interaction behavior type is the interaction behavior type corresponding to the largest conversion progress in the target object conversion link, and the server can calculate the conversion probability corresponding to the interaction behavior type based on the reference subject number corresponding to the interaction behavior type and the conversion reference subject number.

[0121] In this embodiment, the conversion probability corresponding to the interaction behavior type is obtained based on the reference subject number corresponding to the interaction behavior type, and the accuracy of the conversion probability is improved.

[0122] In some embodiments, obtaining the conversion probability corresponding to the interaction behavior type based on the reference subject number corresponding to the interaction behavior type includes: obtaining a current reference subject number corresponding to a current interaction behavior type; obtaining a conversion reference subject number corresponding to a conversion interaction behavior type; calculating a number ratio between the conversion reference subject number and the current reference subject number, and taking the number ratio as the conversion probability corresponding to the current interaction behavior type.

[0123] The current interaction behavior type can be any interaction behavior type. The current reference subject number refers to the reference subject number corresponding to the current interaction behavior type. The conversion interaction behavior type refers to the last interaction behavior type in the target object conversion link, that is, the interaction behavior type with the largest conversion progress in the target object conversion link. The subject of the interaction behavior with the conversion interaction behavior type is the expected subject.

[0124] Since the interaction behavior types in the object conversion link are arranged according to the conversion progress, the larger the conversion progress is, the later the arrangement in the object conversion link is, and thus the subject of the behavior of the interaction behavior type arranged later in the object conversion link also produces the behavior of the interaction behavior type arranged earlier in the object conversion link. Therefore, the subject of the conversion interaction behavior type also produces the behavior of the current interaction behavior type, and the reference interaction subject corresponding to the current interaction behavior type includes the reference interaction subject corresponding to the conversion interaction behavior type, that is, the reference interaction subject of the conversion interaction behavior type is derived from the reference interaction subject of the current interaction behavior type, and the conversion reference subject quantity is less than or equal to the current reference subject quantity.

[0125] Specifically, the server can obtain the reference subject quantity corresponding to each interaction behavior type respectively, divide the conversion reference subject quantity by the reference subject quantity corresponding to the interaction behavior type, and take the result obtained by the division as the conversion possibility corresponding to the interaction behavior type. Taking a marketing activity as an example, after the activity ends, the audience has a total of 100, of which 90 have exposure, 80 have clicks, and 10 have purchases. The reference subject quantities are 90, 80, and 10 respectively, the conversion reference subject quantity is 10, the conversion possibility corresponding to exposure is equal to 10÷90, and the conversion possibility corresponding to clicks is equal to 10÷80.

[0126] In this embodiment, the quantity ratio between the conversion reference subject quantity and the current reference subject quantity is calculated, and the quantity ratio is taken as the conversion possibility corresponding to the current interaction behavior type, so that the conversion possibility reflects the proportion of the subject producing the conversion interaction behavior type among the subjects producing the same interaction behavior type. The larger the proportion is, the larger the conversion possibility is, and the accuracy of the conversion possibility is improved.

[0127] In some embodiments, the method further includes: obtaining a first object recognition stage from the object recognition stage sequence, obtaining the content interaction subject quantity corresponding to each target subject feature from the plurality of target content interaction subjects corresponding to the first object recognition stage; selecting a representative subject feature corresponding to the first object recognition stage based on the content interaction subject quantity corresponding to the target subject feature; selecting a target content interaction subject having the representative subject feature from the target content interaction subjects corresponding to a second object recognition stage as a to-be-pushed subject, the object recognition degree of the second object recognition stage being lower than the object recognition degree of the first object recognition stage; and pushing the push content corresponding to the target push object to the terminal of the to-be-pushed subject.

[0128] The first object recognition stage can be any one of the object recognition stages in the sequence of object recognition stages, and the first object recognition stage can be obtained from the sequence of object recognition stages as needed, for example, can be determined according to the purpose of the marketing activity, for example, when the purpose of the marketing activity is to obtain more consumers in the fan stage, the fan stage can be taken as the first object recognition stage.

[0129] The subject feature refers to the feature possessed by the subject, and the subject feature can include at least one of an age feature, an education state feature, or a profession feature. The age feature can be multiple, and the age feature can be used to identify an age range, for example, the age range can include at least one of 0-14 years old, 18-24 years old, 25-29 years old, or 30-34 years old. The age feature can include a feature corresponding to the age range of 0-14 years old, for example, different age ranges correspond to different age features. The education state can include at least one of primary school, junior high school, high school, or undergraduate. The education state feature can be used to identify the education state, and different education states correspond to different education state features. The target subject feature can be selected as needed, and the target subject feature can be multiple, for example, can include at least one of an age feature or an education state feature, for example, can include at least one of an age feature corresponding to 0-14 years old, an age feature corresponding to 18-24 years old, or an education state feature corresponding to primary school.

[0130] The content interaction subject quantity corresponding to the target subject feature refers to the number of subjects with the target subject feature in the target content interaction subject corresponding to the first object recognition stage. For example, when the target subject feature represents 0-14 years old, the content interaction subject quantity is the number of subjects aged 0-14 years old in the target content interaction subject of the first object recognition stage.

[0131] The object recognition degree of the second object recognition stage is lower than that of the first object recognition stage, for example, the first object recognition stage is the "interest stage", and the second object recognition stage can be the "cognition stage". The to-be-pushed subject refers to the subject to which the target push object corresponding to the push content is to be pushed. The to-be-pushed subject is a target content interaction subject with a representative subject feature in the target content interaction subject of the second object recognition stage. The terminal of the to-be-pushed subject can be, for example, Figure 1 The terminal 102.

[0132] Specifically, the server can select the content interaction subject quantity greater than the quantity threshold value from the content interaction subject quantity corresponding to each target subject feature as the target content interaction subject quantity, and take the target subject feature corresponding to the target content interaction subject quantity as the representative subject feature. The quantity threshold value can be preset or set as needed.

[0133] In some embodiments, the server can arrange the content interaction subject numbers corresponding to the respective target subject features in descending order to obtain a content interaction subject number sequence, obtain the content interaction subject number ranked before a ranking threshold value from the content interaction subject number sequence as a target content interaction subject number, and take the target subject feature corresponding to the target content interaction subject number as the representative subject feature. The ranking threshold value can be preset or set according to requirements.

[0134] In some embodiments, the server can count the number of target content interaction subjects corresponding to the first object recognition stage to obtain a target subject number, and select a representative subject feature corresponding to the first object recognition stage based on the content interaction subject number corresponding to the target subject feature and the target subject number. For example, the server can perform ratio operation on the content interaction subject number and the target subject number to obtain a number ratio corresponding to the content interaction subject number, and select a representative subject feature corresponding to the first object recognition stage based on the number ratio. For example, the server can take the content interaction subject number with a number ratio greater than a ratio threshold value as a target content interaction subject number, and take the target subject feature corresponding to the target content interaction subject number as the representative subject feature. The ratio threshold value can be preset or set according to requirements. The number ratio can also be referred to as coverage. The coverage refers to the proportion of subjects with the target subject feature in the target content interaction subjects corresponding to the object recognition stage. For example, assuming that there are 1000 consumers in the first object recognition stage, and there are 5 consumers under the age of 14 among the 1000 consumers, the coverage of the consumers under the age of 14 is 0.5%. The server can select a target subject feature with a coverage greater than a coverage threshold value from the respective target subject features corresponding to the first object recognition stage as a representative subject feature corresponding to the first object recognition stage. The coverage threshold value can be preset or set according to requirements.

[0135] In some embodiments, the first object recognition stage and the second object recognition stage are adjacent object recognition stages in the sequence of object recognition stages, and the server can determine a flow rate corresponding to the target subject feature based on the number of content interaction subjects corresponding to the target subject feature, the flow rate corresponding to the target subject feature being a proportion of subjects having the target subject feature in the subjects included in the second object recognition stage among the subjects included in the first object recognition stage. For example, the server can count the number of target content interaction subjects corresponding to the second object recognition stage to obtain a second subject number, and calculate the flow rate corresponding to the target subject feature based on the number of content interaction subjects and the second subject number. For example, the target subject feature represents junior high school, the second subject number is 100, the number of subjects included in the first object recognition stage is 50, i.e., 50 subjects in the 100 subjects of the second object recognition stage are converted to the first object recognition stage, and it is assumed that there are 10 subjects of junior high school in the 50 subjects, then the flow rate corresponding to junior high school is 10 / 100. The server can select a target subject feature with a flow rate greater than a flow rate threshold from the target subject features corresponding to the first object recognition stage as a representative subject feature corresponding to the first object recognition stage. The flow rate threshold can be preset or set as needed.

[0136] In some embodiments, the server can determine the coverage rate and the flow rate corresponding to each of the target subject features based on the number of content interaction subjects corresponding to the target subject features, and determine the representative subject feature corresponding to the first object recognition stage according to at least one of the coverage rate or the flow rate, for example, the server can select a target subject feature with a flow rate greater than a flow rate threshold from the target subject features corresponding to the first object recognition stage as a representative subject feature corresponding to the first object recognition stage, or the server can select a target subject feature with a coverage rate greater than a coverage rate threshold from the target subject features corresponding to the first object recognition stage as a representative subject feature corresponding to the first object recognition stage, or determine the representative subject feature corresponding to the first object recognition stage according to the coverage rate and the flow rate corresponding to the target subject feature, for example, the server can obtain a target subject feature with a coverage rate greater than a coverage rate threshold and a flow rate greater than a flow rate threshold from the target subject features as a representative subject feature corresponding to the first object recognition stage.

[0137] For example, before the marketing activity, the consumers targeted by the marketing activity can be determined according to the purpose of the marketing activity. For example, for a marketing activity in the automobile industry, when the purpose of the marketing activity is to obtain more consumers in the fan stage, the server can obtain consumers in the fan stage, determine the number of consumers belonging to multiple age groups from the consumers in the fan stage, and also determine the number of consumers belonging to multiple education states from the consumers in the fan stage. According to the number of consumers in the age groups and the number of consumers in the education states, the representative subject characteristics, which can also be referred to as high-quality characteristics, are determined. The server can put advertisements to consumers meeting the high-quality characteristics in the marketing activity, so as to obtain more consumers in the fan stage.

[0138] As shown in Figure 6 , the coverage and the flow rate corresponding to multiple target subject characteristics are shown. As can be seen from the figure, the coverage of consumers below 14 years old and above 65 years old is less than 10%, the flow rate of consumers with a high school education state is greater than 1.5%, and the characteristics with a coverage greater than 10.0% and a flow rate greater than 1.5% are taken as high-quality characteristics.

[0139] In some embodiments, for characteristics with high flow rate and low coverage, more consumers with the characteristics can be obtained, so as to increase the number of consumers with the characteristics, and thus the accuracy of exposure can be improved when putting advertisements.

[0140] In this embodiment, the target content interaction subject with the representative subject characteristics is selected from the target content interaction subjects corresponding to the second object recognition stage, and is taken as the to-be-pushed subject. Since the object recognition degree of the second object recognition stage is lower than that of the first object recognition stage, the push content corresponding to the target push object is pushed to the terminal of the to-be-pushed subject, so that the target content interaction subject in the second object recognition stage can be converted to the first object recognition stage, thereby improving the conversion efficiency and improving the content pushing accuracy.

[0141] In some embodiments, as shown in Figure 7 , a content pushing method is provided, including the following steps:

[0142] S702, determining a target industry corresponding to a target brand, and obtaining a key conversion link in a marketing conversion link corresponding to the target brand as a target brand conversion link according to the target industry.

[0143] The target industry refers to the industry to which the target brand belongs.

[0144] S704, stage division is performed on the target brand conversion link to obtain a brand recognition stage sequence corresponding to the target brand, and each brand recognition stage in the brand recognition stage sequence corresponds to a different stage priority.

[0145] The brand recognition stage sequence includes a plurality of brand recognition stages arranged according to brand recognition degrees.

[0146] S706, obtain an interaction record sub-set of each consumer behavior type corresponding to the target brand.

[0147] The same interaction record sub-set includes target advertisement interaction records of the same consumer behavior type, and the target advertisement interaction record includes a type of consumer behavior generated between the consumer and the advertisement of the target brand, i.e., a consumer behavior type.

[0148] S708, obtain a current brand recognition stage from the brand recognition stage sequence in order according to the stage priority, obtain a current standard consumer behavior type corresponding to the current brand recognition stage and a consumer behavior effective time length corresponding to the current brand recognition stage, and obtain a consumer behavior effective time period corresponding to the current brand recognition stage based on the consumer behavior effective time length corresponding to the current brand recognition stage.

[0149] S710, obtain target advertisement interaction records of a consumer being a current consumer and an interaction time belonging to a consumer behavior effective time period corresponding to the current brand recognition stage from each interaction record sub-set to form a current interaction record set corresponding to the current consumer.

[0150] The current consumer can be any consumer corresponding to the target brand.

[0151] S712, determine whether there is a target advertisement interaction record in the current interaction record set with a consumer behavior type matching the current standard consumer behavior type, if not, return to step S708, and if yes, execute S714.

[0152] S714, take the current consumer as a target consumer corresponding to the current brand recognition stage.

[0153] The server can take the target consumer corresponding to each brand recognition stage as brand data assets of the target brand.

[0154] S716, obtain a representative feature of a target consumer of a first brand recognition stage, and obtain a consumer with the representative feature from target consumers corresponding to a second brand recognition stage as a consumer to be pushed corresponding to the target brand.

[0155] The stage priority corresponding to the first brand recognition stage is higher than the stage priority corresponding to the second brand recognition stage.

[0156] S718, pushing the advertisement corresponding to the target brand to the consumer to be pushed.

[0157] For brands in different industries, the degree of brand recognition expressed by the same consumer behavior is different, for example, the same advertisement exposure has different influences on stimulating consumers to download game software and purchase cars, and different consumer behaviors express different attitudes, for example, the behavior of actively searching can reflect that the consumer has a strong intention to the brand, the behavior of actively liking or forwarding can reflect that the consumer recognizes the brand, and passive exposure cannot reflect the real attitude of the consumer. The content pushing method provided in the application determines the target brand conversion link corresponding to the target brand through the target industry corresponding to the target brand, divides the target brand conversion link into stages, obtains the brand recognition stage sequence corresponding to the target brand, so that the brand recognition stage sequence presents the differences between different industries, and the brand recognition stage sequence obtained by different industries can be different, so that the definition of brand data assets is close to industry business, the brand data assets can fully measure the marketing effect, and the specified marketing strategy can be assisted. The user's preference for the product is mined from the consumer behavior, and the degree of recognition of the user to the product can be accurately obtained through the consumer behavior data, and the user with high recognition degree can be taken as the consumer group of the product.

[0158] The content pushing method provided in the application doubles the contract advertisement click rate in the actual marketing activity of a certain automobile industry, and the click rate in the effect advertisement is increased by 27%, and the conversion rate is increased by 45%. The delivery effect of the contract advertisement and the bidding advertisement exceeds the expectation of the brand advertiser, and good results are achieved, proving the feasibility and effectiveness of the content pushing method provided in the application.

[0159] It should be understood that, although Figures 2-7 The steps in the flowchart are displayed in sequence according to the direction of the arrow, but these steps are not necessarily executed in sequence according to the direction of the arrow. Unless otherwise specified in this article, the execution of these steps has no strict sequence limitation, and these steps can be executed in other orders. Moreover, Figures 2-7 At least part of the steps in the flowchart can include multiple steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily sequential, but can be executed in rotation or alternation with other steps or steps or stages in other steps.

[0160] In some embodiments, as Figure 8As shown, a content pushing device is provided, which can be a software module or a hardware module, or a combination of the two as part of a computer device, and specifically includes: a target content interaction record set obtaining module 802, an object recognition stage sequence obtaining module 804, a standard interaction behavior type obtaining module 806, and a target content interaction subject obtaining module 808, wherein:

[0161] The target content interaction record set obtaining module 802 is configured to obtain a target content interaction record set corresponding to a target pushing object; a content interaction record in the target content interaction record set includes an interaction behavior type of a content interaction subject to a target pushing content, and the target pushing content is pushing content related to the target pushing object;

[0162] The object recognition stage sequence obtaining module 804 is configured to obtain an object recognition stage sequence corresponding to the target pushing object, the object recognition stage sequence including a plurality of object recognition stages arranged according to an object recognition degree, and the object recognition stage sequence being obtained according to an interaction behavior type corresponding to an object conversion link of the target pushing object;

[0163] The standard interaction behavior type obtaining module 806 is configured to obtain a standard interaction behavior type corresponding to an object recognition stage in the object recognition stage sequence;

[0164] The target content interaction subject obtaining module 808 is configured to determine a content interaction subject with a matching interaction behavior type and standard interaction behavior type according to the target content interaction record set, and obtain a target content interaction subject corresponding to the object recognition stage based on the content interaction subject with the matching interaction behavior type and standard interaction behavior type.

[0165] In some embodiments, the target content interaction subject obtaining module 808 includes: a sub-interaction record set obtaining unit configured to obtain content interaction records corresponding to each content interaction subject from the target content interaction record set to form a sub-interaction record set corresponding to the content interaction subject; and a first target content interaction subject obtaining unit configured to take the content interaction subject as a target content interaction subject corresponding to the object recognition stage when the interaction behavior type in the sub-interaction record set matches the standard interaction behavior type.

[0166] In some embodiments, the object recognition stage corresponds to a stage priority, the stage priority is in a positive correlation with an object recognition degree corresponding to the object recognition stage; the standard interaction behavior type obtaining module 806 is further configured to sequentially obtain, according to the stage priority, a current object recognition stage from the sequence of object recognition stages, and obtain a current standard interaction behavior type corresponding to the current object recognition stage; the first target content interaction subject obtaining unit is further configured to, when the interaction behavior type in the sub-interaction record set matches the current standard interaction behavior type, take the content interaction subject as a target content interaction subject corresponding to the current object recognition stage; otherwise, sequentially obtain, according to the stage priority, a current object recognition stage from the sequence of object recognition stages, and update the current object recognition stage until the interaction behavior type in the sub-interaction record set matches the current standard interaction behavior type.

[0167] In some embodiments, the standard interaction behavior type obtaining module 806 is further configured to obtain a current standard interaction behavior type corresponding to the current object recognition stage and an interaction behavior valid time length corresponding to the current object recognition stage; the first target content interaction subject obtaining unit is further configured to obtain, based on the interaction behavior valid time length corresponding to the current object recognition stage, an interaction behavior valid time period corresponding to the current object recognition stage; when there is a content interaction record in the sub-interaction record set, the interaction behavior type of which matches the current standard interaction behavior type and the interaction time belongs to the interaction behavior valid time period, take the content interaction subject as a target content interaction subject corresponding to the current object recognition stage.

[0168] In some embodiments, the target content interaction subject obtaining module 808 comprises:

[0169] The interaction record sub-set obtaining unit is configured to divide the target content interaction record set according to the interaction behavior type, and obtain an interaction record sub-set corresponding to each interaction behavior type respectively;

[0170] The second target content interaction subject obtaining unit is configured to obtain an interaction record sub-set corresponding to the standard interaction behavior type, and obtain a target content interaction subject corresponding to the object recognition stage based on the interaction record sub-set corresponding to the standard interaction behavior type.

[0171] In some embodiments, the object recognition stage corresponds to a stage priority, the stage priority is in a positive correlation with an object recognition degree corresponding to the object recognition stage; the second target content interaction subject obtaining unit is further configured to: obtain a current interaction record sub-set corresponding to a standard interaction behavior type corresponding to a current object recognition stage, and form a current content interaction subject set by using content interaction subjects in the current interaction record sub-set; obtain a priority interaction record sub-set corresponding to a standard interaction behavior type corresponding to a priority object recognition stage, and form a priority content interaction subject set by using content interaction subjects in the priority interaction record sub-set, the priority object recognition stage being an object recognition stage in the object recognition stage sequence and having a stage priority higher than that of the current object recognition stage; and filter out the priority content interaction subject set from the current content interaction subject set to obtain a target content interaction subject corresponding to the current object recognition stage.

[0172] In some embodiments, the apparatus further includes an object recognition stage sequence obtaining module, the object recognition stage sequence obtaining module including:

[0173] a target object conversion link obtaining unit configured to obtain a target object conversion link, the target object conversion link including a plurality of interaction behavior types arranged according to conversion progress; wherein the interaction behavior types are obtained according to behavior purposes;

[0174] a conversion possibility obtaining unit configured to obtain a conversion possibility corresponding to each interaction behavior type;

[0175] an object recognition stage sequence obtaining unit configured to perform stage division on the target object conversion link based on the conversion possibility corresponding to each interaction behavior type to obtain an object recognition stage sequence; wherein a standard interaction behavior type corresponding to an object recognition stage is an interaction behavior type divided into the object recognition stage.

[0176] In some embodiments, the object recognition stage sequence obtaining unit is further configured to: based on the conversion possibility corresponding to each interaction behavior type, calculate a possibility difference between conversion possibilities of adjacent interaction behavior types in the target object conversion link; and divide the adjacent interaction behavior types with a possibility difference less than a difference threshold into the same object recognition stage.

[0177] In some embodiments, the conversion possibility acquisition unit is further configured to acquire a set of reference content interaction records corresponding to the target object conversion link, wherein each content interaction record in the set of reference content interaction records comprises a reference interaction subject and an interaction behavior type of the reference interaction subject with reference push content, and the reference push content is push content related to a reference push object corresponding to the target object conversion link; count a number of reference interaction subjects with the same interaction behavior type in the set of reference content interaction records to obtain a reference subject number corresponding to the interaction behavior type; and obtain the conversion possibility corresponding to the interaction behavior type based on the reference subject number corresponding to the interaction behavior type.

[0178] In some embodiments, the conversion possibility acquisition unit is further configured to acquire a current reference subject number corresponding to a current interaction behavior type; acquire a conversion reference subject number corresponding to a conversion interaction behavior type; and calculate a number ratio between the conversion reference subject number and the current reference subject number, and use the number ratio as the conversion possibility corresponding to the current interaction behavior type.

[0179] In some embodiments, the apparatus further comprises:

[0180] The content interaction subject number obtaining module is configured to obtain a first object recognition stage from the sequence of object recognition stages, and obtain a number of content interaction subjects corresponding to each target subject feature in a plurality of target content interaction subjects corresponding to the first object recognition stage.

[0181] The representative subject feature selection module is configured to select a representative subject feature corresponding to the first object recognition stage based on the number of content interaction subjects corresponding to the target subject feature.

[0182] The to-be-pushed subject selection module is configured to select a target content interaction subject having the representative subject feature from target content interaction subjects corresponding to a second object recognition stage as a to-be-pushed subject, wherein the object recognition degree of the second object recognition stage is lower than the object recognition degree of the first object recognition stage.

[0183] The content push module is configured to push push content corresponding to the target push object to a terminal of the to-be-pushed subject.

[0184] For specific limitations of the content push apparatus, refer to the limitations of the content push method described above, which will not be repeated here. Each module in the content push apparatus described above can be realized by software, hardware, and combinations thereof, in whole or in part. Each module described above can be embedded in or independent of a processor in a computer device in hardware form, or can be stored in a memory in a computer device in software form, so as to be called and executed by a processor to perform the operations corresponding to each module.

[0185] In some embodiments, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 9 As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a content delivery method. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0186] In some embodiments, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 10 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The database stores data involved in the content delivery method. The network interface communicates with external terminals via a network connection. When the computer program is executed by the processor, it implements a content delivery method.

[0187] Those skilled in the art will understand that Figure 9 and Figure 10 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0188] In some embodiments, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0189] In some embodiments, a computer readable storage medium storing a computer program is provided, the computer program, when executed by a processor, implements the steps in the above method embodiments.

[0190] In some embodiments, a computer program product or computer program is provided, the computer program product or computer program comprising 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 the processor executes the computer instructions, so that the computer device performs the steps in the above method embodiments.

[0191] A person of ordinary skill in the art can understand that all or part of the processes in the above method embodiments can be completed by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above method embodiments. Any reference to memory, storage, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0192] Any combination of the technical features of the above embodiments can be made. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combination of the technical features does not exist, it should be considered as the scope of the present application.

[0193] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, some modifications and improvements can be made, which are all within the scope of the present application. Therefore, the scope of protection of the patent of the present application should be subject to the appended claims.

Claims

1. A content push method characterized by, The method comprises: obtaining a target content interaction record set corresponding to a target push object; a content interaction record in the target content interaction record set comprises an interaction behavior type of a content interaction subject to target push content, and the target push content is push content related to the target push object; obtaining an object approval stage sequence corresponding to the target push object, the object approval stage sequence comprising a plurality of object approval stages arranged according to an object approval degree, the object approval stage sequence being obtained according to an interaction behavior type corresponding to an object conversion link of the target push object; the object conversion link comprising a plurality of interaction behavior types arranged according to a conversion progress; wherein the interaction behavior type is obtained according to a behavior purpose; obtaining a standard interaction behavior type corresponding to an object approval stage in the object approval stage sequence; determining a content interaction subject whose interaction behavior type matches the standard interaction behavior type according to the target content interaction record set, and obtaining a target content interaction subject corresponding to the object approval stage based on the content interaction subject whose interaction behavior type matches the standard interaction behavior type; obtaining a first object approval stage from the object approval stage sequence, and obtaining a number of content interaction subjects corresponding to each target subject feature from a plurality of target content interaction subjects corresponding to the first object approval stage; selecting a representative subject feature corresponding to the first object approval stage based on the number of content interaction subjects corresponding to the target subject feature; selecting a target content interaction subject having the representative subject feature from target content interaction subjects corresponding to a second object approval stage as a push subject, the object approval degree of the second object approval stage being lower than the object approval degree of the first object approval stage; pushing push content corresponding to the target push object to a terminal of the push subject.

2. The method of claim 1, wherein, The method comprises: obtaining a target content interaction record set corresponding to a target push object; a content interaction record in the target content interaction record set comprises an interaction behavior type of a content interaction subject to target push content, and the target push content is push content related to the target push object; obtaining an object approval stage sequence corresponding to the target push object, the object approval stage sequence comprising a plurality of object approval stages arranged according to an object approval degree, the object approval stage sequence being obtained according to an interaction behavior type corresponding to an object conversion link of the target push object; the object conversion link comprising a plurality of interaction behavior types arranged according to a conversion progress; wherein the interaction behavior type is obtained according to a behavior purpose; 3. The method of claim 2, wherein, obtaining a standard interaction behavior type corresponding to an object approval stage in the object approval stage sequence; determining a content interaction subject whose interaction behavior type matches the standard interaction behavior type according to the target content interaction record set, and obtaining a target content interaction subject corresponding to the object approval stage based on the content interaction subject whose interaction behavior type matches the standard interaction behavior type; obtaining a first object approval stage from the object approval stage sequence, and obtaining a number of content interaction subjects corresponding to each target subject feature from a plurality of target content interaction subjects corresponding to the first object approval stage; selecting a representative subject feature corresponding to the first object approval stage based on the number of content interaction subjects corresponding to the target subject feature; selecting a target content interaction subject having the representative subject feature from target content interaction subjects corresponding to a second object approval stage as a push subject, the object approval degree of the second object approval stage being lower than the object approval degree of the first object approval stage; pushing push content corresponding to the target push object to a terminal of the push subject. The method comprises: obtaining a target content interaction record set corresponding to a target push object; a content interaction record in the target content interaction record set comprises an interaction behavior type of a content interaction subject to target push content, and the target push content is push content related to the target push object; obtaining an object approval stage sequence corresponding to the target push object, the object approval stage sequence comprising a plurality of object approval stages arranged according to an object approval degree, the object approval stage sequence being obtained according to an interaction behavior type corresponding to an object conversion link of the target push object; the object conversion link comprising a plurality of interaction behavior types arranged according to a conversion progress; wherein the interaction behavior type is obtained according to a behavior purpose; obtaining a standard interaction behavior type corresponding to an object approval stage in the object approval stage sequence; determining a content interaction subject whose interaction behavior type matches the standard interaction behavior type according to the target content interaction record set, and obtaining a target content interaction subject corresponding to the object approval stage based on the content interaction subject whose interaction behavior type matches the standard interaction behavior type; obtaining a first object approval stage from the object approval stage sequence, and obtaining a number of content interaction subjects corresponding to each target subject feature from a plurality of target content interaction subjects corresponding to the first object approval stage; selecting a representative subject feature corresponding to the first object approval stage based on the number of content interaction subjects corresponding to the target subject feature; selecting a target content interaction subject having the representative subject feature from target content interaction subjects corresponding to a second object approval stage as a push subject, the object approval degree of the second object approval stage being lower than the object approval degree of the first object approval stage; pushing push content corresponding to the target push object to a terminal of the push subject. The content interaction subject is determined as the target content interaction subject corresponding to the object recognition stage when the interaction behavior type in the sub-interaction record set matches the standard interaction behavior type. The content interaction subject is determined as the target content interaction subject corresponding to the current object recognition stage when the interaction behavior type in the sub-interaction record set matches the current standard interaction behavior type. Otherwise, the current object recognition stage is updated by sequentially obtaining the current object recognition stage from the sequence of object recognition stages according to the stage priority order until the interaction behavior type in the sub-interaction record set matches the current standard interaction behavior type.

4. The method of claim 3, wherein, The current standard interaction behavior type corresponding to the current object recognition stage is obtained. The current standard interaction behavior type corresponding to the current object recognition stage and the effective interaction behavior duration corresponding to the current object recognition stage are obtained. The content interaction subject is determined as the target content interaction subject corresponding to the object recognition stage when the interaction behavior type in the sub-interaction record set matches the standard interaction behavior type. The effective interaction behavior time period corresponding to the current object recognition stage is obtained based on the effective interaction behavior duration corresponding to the current object recognition stage. The content interaction subject is determined as the target content interaction subject corresponding to the current object recognition stage when the interaction behavior type in the sub-interaction record set matches the current standard interaction behavior type and the interaction time belongs to the effective interaction behavior time period.

5. The method of claim 1, wherein, The target content interaction subject corresponding to the object recognition stage is obtained based on the content interaction subject whose interaction behavior type matches the standard interaction behavior type. The target content interaction record set is divided according to the interaction behavior type to obtain an interaction record sub-set corresponding to each interaction behavior type. The target content interaction subject corresponding to the object recognition stage is obtained based on the interaction record sub-set corresponding to the standard interaction behavior type.

6. The method of claim 5, wherein, The object recognition stage corresponds to a stage priority, and the stage priority is positively correlated with the object recognition degree corresponding to the object recognition stage. The target content interaction subject corresponding to the object recognition stage is obtained based on the interaction record sub-set corresponding to the standard interaction behavior type. The current interaction record sub-set corresponding to the standard interaction behavior type corresponding to the current object recognition stage is obtained, and the content interaction subjects in the current interaction record sub-set are grouped into a current content interaction subject set. obtaining a priority interaction record sub-set corresponding to a standard interaction behavior type of a priority object approval stage, and grouping content interaction subjects in the priority interaction record sub-set into a priority content interaction subject set, the priority object approval stage being an object approval stage in the object approval stage sequence and having a stage priority higher than that of the current object approval stage; filtering the priority content interaction subject set from the current content interaction subject set to obtain a target content interaction subject of the current object approval stage.

7. The method of claim 1, wherein, The step of dividing the object approval stage sequence includes: obtaining a target object conversion link and obtaining a conversion possibility degree corresponding to each interaction behavior type; dividing the target object conversion link into stages based on the conversion possibility degrees corresponding to the interaction behavior types to obtain the object approval stage sequence, wherein the standard interaction behavior type corresponding to the object approval stage is the interaction behavior type divided into the object approval stage.

8. The method of claim 7, wherein, The step of dividing the target object conversion link into stages based on the conversion possibility degrees corresponding to the interaction behavior types to obtain the object approval stage sequence includes: calculating a possibility difference between conversion possibility degrees of adjacent interaction behavior types in the target object conversion link based on the conversion possibility degrees corresponding to the interaction behavior types; dividing adjacent interaction behavior types with a possibility difference less than a difference threshold into the same object approval stage.

9. The method of claim 7, wherein, The step of obtaining the conversion possibility degree corresponding to each interaction behavior type includes: obtaining a reference content interaction record set corresponding to the target object conversion link, a content interaction record in the reference content interaction record set including an interaction behavior type of a reference interaction subject to a reference push content, the reference push content being push content related to a reference push object corresponding to the target object conversion link; counting a number of reference interaction subjects of the same interaction behavior type in the reference content interaction record set to obtain a reference subject number corresponding to the interaction behavior type; obtaining the conversion possibility degree corresponding to the interaction behavior type based on the reference subject number corresponding to the interaction behavior type.

10. The method of claim 9, wherein, The step of obtaining the conversion possibility degree corresponding to the interaction behavior type based on the reference subject number corresponding to the interaction behavior type includes: obtaining a current reference subject number corresponding to a current interaction behavior type; obtaining a conversion reference subject number corresponding to a conversion interaction behavior type; calculating a number ratio between the conversion reference subject number and the current reference subject number, and taking the number ratio as the conversion possibility degree corresponding to the current interaction behavior type.

11. A content pusher device, characterized by The apparatus includes: a target content interaction record set obtaining module configured to obtain a target content interaction record set corresponding to a target push object, a content interaction record in the target content interaction record set including an interaction behavior type of a content interaction subject to a target push content, the target push content being push content related to the target push object; An object recognition stage sequence obtaining module is configured to obtain an object recognition stage sequence corresponding to the target push object, the object recognition stage sequence including a plurality of object recognition stages arranged according to object recognition degrees, and the object recognition stage sequence being divided according to an interaction behavior type corresponding to an object conversion link of the target push object; the object conversion link including a plurality of interaction behavior types arranged according to conversion progress; wherein the interaction behavior type is divided according to a behavior purpose; A standard interaction behavior type obtaining module is configured to obtain a standard interaction behavior type corresponding to an object recognition stage in the object recognition stage sequence; A target content interaction subject obtaining module is configured to determine a content interaction subject whose interaction behavior type matches the standard interaction behavior type according to the target content interaction record set, and obtain a target content interaction subject corresponding to the object recognition stage based on the content interaction subject whose interaction behavior type matches the standard interaction behavior type; A content interaction subject quantity obtaining module is configured to obtain a first object recognition stage from the object recognition stage sequence, and obtain a quantity of content interaction subjects corresponding to each target subject feature in a plurality of target content interaction subjects corresponding to the first object recognition stage; A representative subject feature selecting module is configured to select a representative subject feature corresponding to the first object recognition stage based on the quantity of content interaction subjects corresponding to the target subject feature; A to-be-pushed subject selecting module is configured to select a target content interaction subject having the representative subject feature from target content interaction subjects corresponding to a second object recognition stage as a to-be-pushed subject, the object recognition degree of the second object recognition stage being lower than the object recognition degree of the first object recognition stage; A content push module is configured to push a push content corresponding to the target push object to a terminal of the to-be-pushed subject.

12. The content push apparatus according to claim 11, wherein The target content interaction subject obtaining module includes: A sub-interaction record set obtaining unit is configured to obtain content interaction records corresponding to each content interaction subject from the target content interaction record set to form a sub-interaction record set corresponding to the content interaction subject; A first target content interaction subject obtaining unit is configured to take the content interaction subject as a target content interaction subject corresponding to the object recognition stage when an interaction behavior type in the sub-interaction record set matches the standard interaction behavior type.

13. The content push apparatus according to claim 12, characterized by The object recognition stage corresponds to a stage priority, and the stage priority is in a positive correlation with an object recognition degree corresponding to the object recognition stage; The standard interaction behavior type obtaining module is further configured to sequentially obtain a current object recognition stage from the object recognition stage sequence according to a stage priority order, and obtain a current standard interaction behavior type corresponding to the current object recognition stage. The first target content interaction subject obtaining unit is further configured to, when the interaction behavior type in the sub-interaction record set matches the current standard interaction behavior type, recognize the content interaction subject as a target content interaction subject corresponding to a current object recognition stage; otherwise, update the current object recognition stage by sequentially obtaining a current object recognition stage from the object recognition stage sequence according to a stage priority order until the interaction behavior type in the sub-interaction record set matches the current standard interaction behavior type.

14. The content push apparatus according to claim 13, wherein The standard interaction behavior type obtaining module is further configured to obtain a current standard interaction behavior type corresponding to the current object recognition stage and an interaction behavior valid time length corresponding to the current object recognition stage. The first target content interaction subject obtaining unit is further configured to obtain an interaction behavior valid time period corresponding to the current object recognition stage based on the interaction behavior valid time length corresponding to the current object recognition stage; and when there is a content interaction record in the sub-interaction record set, the content interaction record has an interaction behavior type matching the current standard interaction behavior type and an interaction time belonging to the interaction behavior valid time period, recognize the content interaction subject as a target content interaction subject corresponding to the current object recognition stage.

15. The content push apparatus of claim 11, wherein, The target content interaction subject obtaining module includes: An interaction record sub-set obtaining unit configured to divide the target content interaction record set according to interaction behavior types to obtain interaction record sub-sets respectively corresponding to the interaction behavior types; A second target content interaction subject obtaining unit configured to obtain an interaction record sub-set corresponding to a standard interaction behavior type, and obtain a target content interaction subject corresponding to the object recognition stage based on the interaction record sub-set corresponding to the standard interaction behavior type.

16. The content push apparatus according to claim 15, wherein The object recognition stages correspond to stage priorities, and the stage priorities are in a positive correlation with object recognition degrees corresponding to the object recognition stages; The second target content interaction subject obtaining unit is further configured to obtain a current interaction record sub-set corresponding to a standard interaction behavior type corresponding to the current object recognition stage, and form a current content interaction subject set by using content interaction subjects in the current interaction record sub-set; Obtain a priority interaction record sub-set corresponding to a standard interaction behavior type corresponding to a priority object recognition stage, form a priority content interaction subject set by using content interaction subjects in the priority interaction record sub-set, and the priority object recognition stage is an object recognition stage in the object recognition stage sequence and has a stage priority higher than that of the current object recognition stage; and filter out the priority content interaction subject set from the current content interaction subject set to obtain a target content interaction subject corresponding to the current object recognition stage.

17. The content push apparatus of claim 11, wherein The apparatus further includes an object recognition stage sequence obtaining module, and the object recognition stage sequence obtaining module includes: A target object conversion link obtaining unit configured to obtain a target object conversion link; A conversion possibility obtaining unit configured to obtain a conversion possibility corresponding to each interaction behavior type; A target content interaction subject obtaining unit configured to obtain a target content interaction subject corresponding to the current object recognition stage based on the conversion possibility corresponding to the current object recognition stage. The object recognition stage sequence obtaining unit is configured to divide the target object conversion link into stages based on the conversion possibility degrees corresponding to the respective interaction behavior types, to obtain the object recognition stage sequence; wherein the standard interaction behavior type corresponding to the object recognition stage is the interaction behavior type that is divided into the object recognition stage.

18. The content push apparatus of claim 17, wherein, The object recognition stage sequence obtaining unit is further configured to calculate the possibility difference between the conversion possibility degrees of adjacent interaction behavior types in the target object conversion link based on the conversion possibility degrees corresponding to the respective interaction behavior types; and divide adjacent interaction behavior types with a possibility difference less than a difference threshold into the same object recognition stage.

19. The content push apparatus of claim 17, wherein The conversion possibility degree obtaining unit is further configured to obtain a reference content interaction record set corresponding to the target object conversion link, wherein the content interaction records in the reference content interaction record set include the interaction behavior types of reference interaction subjects with reference push content, and the reference push content is push content related to a reference push object corresponding to the target object conversion link. The number of reference interaction subjects of the same interaction behavior type in the reference content interaction record set is counted to obtain the reference subject number corresponding to the interaction behavior type; and the conversion possibility degree corresponding to the interaction behavior type is obtained based on the reference subject number corresponding to the interaction behavior type.

20. The content push apparatus according to claim 19, wherein The conversion possibility degree obtaining unit is further configured to obtain a current reference subject number corresponding to a current interaction behavior type; obtain a conversion reference subject number corresponding to a conversion interaction behavior type; calculate the number ratio between the conversion reference subject number and the current reference subject number; and use the number ratio as the conversion possibility degree corresponding to the current interaction behavior type. 21.A computer device, comprising a memory and a processor, wherein the memory stores a computer program. The processor executes the computer program to implement the method in any one of claims 1 to 10.

22. A computer readable storage medium storing a computer program, wherein the computer program comprises instructions that, when executed by a computer, cause the computer to perform the method according to any one of claims 1 to 21. The computer program is executed by the processor to implement the method in any one of claims 1 to 10.

23. A computer program product comprising computer instructions, characterized in that, The computer program is executed by the processor to implement the method in any one of claims 1 to 10. The computer program is executed by the processor to implement the method in any one of claims 1 to 10.

Images