CN112767053A - Information processing method, information processing device, electronic equipment and storage medium - Google Patents

Information processing method, information processing device, electronic equipment and storage medium Download PDF

Info

Publication number
CN112767053A
CN112767053A CN202110129488.8A CN202110129488A CN112767053A CN 112767053 A CN112767053 A CN 112767053A CN 202110129488 A CN202110129488 A CN 202110129488A CN 112767053 A CN112767053 A CN 112767053A
Authority
CN
China
Prior art keywords
information
click probability
target
account
candidate information
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202110129488.8A
Other languages
Chinese (zh)
Inventor
赵惜墨
闫铭
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Dajia Internet Information Technology Co Ltd
Original Assignee
Beijing Dajia Internet Information Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing Dajia Internet Information Technology Co Ltd filed Critical Beijing Dajia Internet Information Technology Co Ltd
Priority to CN202110129488.8A priority Critical patent/CN112767053A/en
Publication of CN112767053A publication Critical patent/CN112767053A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0251—Targeted advertisements
    • G06Q30/0255—Targeted advertisements based on user history
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0251—Targeted advertisements
    • G06Q30/0254—Targeted advertisements based on statistics
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0251—Targeted advertisements
    • G06Q30/0269—Targeted advertisements based on user profile or attribute
    • G06Q30/0271—Personalized advertisement
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0277—Online advertisement

Landscapes

  • Business, Economics & Management (AREA)
  • Finance (AREA)
  • Strategic Management (AREA)
  • Engineering & Computer Science (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Physics & Mathematics (AREA)
  • Game Theory and Decision Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Probability & Statistics with Applications (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

The disclosure shows an information processing method, an information processing device, an electronic device and a storage medium, wherein when an information acquisition request sent by a client is received, a plurality of candidate information are acquired, and the information acquisition request comprises an account identifier and a request type; acquiring the estimated click probability of the account on the request type information; determining the number of first targets according to the estimated click probability, wherein the number of the first targets and the estimated click probability form a positive correlation; and screening a first target number of candidate information from the plurality of candidate information to obtain first target information. Because the number of the first targets is determined by the estimated click probability, and the number of the first targets and the estimated click probability form a positive correlation relationship, a larger number of candidate information can be screened for an account with a high estimated click probability, so that the probability of finally screening high-quality candidate information is improved; for the account with low estimated click probability, a smaller amount of candidate information can be screened out, so that the computing resources occupied by the post-screening are saved.

Description

Information processing method, information processing device, electronic equipment and storage medium
Technical Field
The present disclosure relates to the field of computer technologies, and in particular, to an information processing method and apparatus, an electronic device, and a storage medium.
Background
With the rapid development of internet technology, internet advertisements gradually become a mainstream advertisement medium. Compared with the traditional advertisement media, the internet advertisement has the advantages of wide coverage, strong initiative and enthusiasm, relatively low cost, high cost performance, strong interactivity and the like, so the internet advertisement is more and more favored by various companies and merchants.
The media platform can place advertisements in the set advertisement positions. When the media platform server receives an advertisement putting request sent by a client, the server acquires an advertisement to be put from an advertisement database, and then screens the advertisement to be put step by step according to the flow of orientation → recall → rough row → fine row and the calculation result of income acquired by displaying the advertisement for thousands of times.
In the related art, in order to improve the probability of screening out high-value advertisements, the number of advertisements with larger coarse rank and fine rank is generally increased according to the available computing resources, so that the problem of computing resource waste is caused.
Disclosure of Invention
The present disclosure provides an information processing method, an information processing apparatus, an electronic device, and a storage medium, to at least solve the problem of computing resource waste in the related art. The technical scheme of the disclosure is as follows:
according to a first aspect of the present disclosure, there is provided an information processing method, the method including:
when an information acquisition request sent by a client is received, acquiring a plurality of candidate information, wherein the information acquisition request comprises an account identifier and a request type, and the request type is the type of each candidate information;
obtaining the estimated click probability of the account corresponding to the account identification on the information of the request type;
determining the number of first targets according to the estimated click probability, wherein the number of the first targets and the estimated click probability form a positive correlation;
and screening the candidate information with the first target number from the candidate information to obtain first target information.
In an optional implementation manner, the step of obtaining the estimated click probability of the account corresponding to the account identifier for the information of the request type includes:
extracting first characteristic information of the account;
inputting the first characteristic information into a click rate estimation model obtained through pre-training to obtain estimated click probability of the account on the information of the request type, wherein the click rate estimation model is obtained through training according to the first characteristic information of a plurality of sample accounts and click labels of the sample accounts on the information of the request type, and the click labels are used for indicating whether the sample accounts click on the information of the request type.
In an optional implementation manner, the step of determining the first target number according to the estimated click probability includes:
when the estimated click probability is larger than a first preset threshold value, determining that the first target number is a preset upper limit value;
when the estimated click probability is smaller than a second preset threshold value, determining the number of the first targets as a preset lower limit value;
and when the estimated click probability is smaller than or equal to the first preset threshold and larger than or equal to the second preset threshold, determining the number of the first targets according to the preset upper limit, the preset lower limit, the estimated click probability, the first preset threshold and the second preset threshold.
In an optional implementation manner, when the estimated click probability is less than or equal to the first preset threshold and greater than or equal to the second preset threshold, the first target number is determined according to the following formula:
Figure BDA0002924654240000021
wherein the upper _ score represents the first preset threshold, the lower _ score represents the second preset threshold, the quota represents the preset upper limit, the lower _ num represents the preset lower limit, the qs represents the estimated click probability, and the first preset threshold is greater than the second preset threshold.
In an optional implementation manner, the step of filtering the candidate information of the first target number from the candidate information to obtain the first target information includes:
calculating a first estimation effect parameter of each candidate information by adopting a first neural network model, wherein the first estimation effect parameter is used for representing estimation of a delivery effect after the candidate information is delivered to the account;
and selecting a first target number of candidate information with larger first estimation effect parameters from the plurality of candidate information to obtain the first target information.
In an optional implementation manner, after the step of selecting a first target number of candidate information with a larger first pre-estimated effect parameter from the plurality of candidate information to obtain the first target information, the method further includes:
determining the number of second targets according to the estimated click probability;
calculating a second estimated effect parameter of each first target information by adopting a second neural network model, wherein the second estimated effect parameter is used for representing estimation of a delivery effect after the first target information is delivered to the account;
and selecting first target information with a second target number with a larger second estimation effect parameter from the first target information with the first target number to obtain second target information.
In an optional implementation manner, the step of obtaining a plurality of candidate information includes:
extracting second characteristic information of the account;
and when the second characteristic information is matched with the directional delivery information, acquiring candidate information corresponding to the directional delivery information, wherein the directional delivery information is information which is required by a delivery object of the candidate information.
According to a second aspect of the present disclosure, there is provided an information processing apparatus, the apparatus including:
the information processing method comprises a first module, a second module and a third module, wherein the first module is configured to obtain a plurality of candidate information when receiving an information obtaining request sent by a client, the information obtaining request comprises an account identifier and a request type, and the request type is a type to which each candidate information belongs;
the second module is configured to acquire the estimated click probability of the account corresponding to the account identifier on the information of the request type;
a third module, configured to determine a first target number according to the estimated click probability, wherein the first target number has a positive correlation with the estimated click probability;
a fourth module configured to filter the candidate information with the first target number from the plurality of candidate information to obtain first target information.
In an alternative implementation, the second module is specifically configured to:
extracting first characteristic information of the account;
inputting the first characteristic information into a click rate estimation model obtained through pre-training to obtain estimated click probability of the account on the information of the request type, wherein the click rate estimation model is obtained through training according to the first characteristic information of a plurality of sample accounts and click labels of the sample accounts on the information of the request type, and the click labels are used for indicating whether the sample accounts click on the information of the request type.
In an alternative implementation, the third module includes:
the first unit is configured to determine that the first target number is a preset upper limit value when the estimated click probability is larger than a first preset threshold value;
the second unit is configured to determine that the first target number is a preset lower limit value when the estimated click probability is smaller than a second preset threshold value;
a third unit, configured to determine the first number of targets according to the preset upper limit, the preset lower limit, the estimated click probability, the first preset threshold and the second preset threshold when the estimated click probability is less than or equal to the first preset threshold and greater than or equal to the second preset threshold.
In an alternative implementation, the third unit is specifically configured to: determining the first target quantity according to the following formula:
Figure BDA0002924654240000041
wherein the upper _ score represents the first preset threshold, the lower _ score represents the second preset threshold, the quota represents the preset upper limit, the lower _ num represents the preset lower limit, the qs represents the estimated click probability, and the first preset threshold is greater than the second preset threshold.
In an alternative implementation, the fourth module is specifically configured to:
calculating a first estimation effect parameter of each candidate information by adopting a first neural network model, wherein the first estimation effect parameter is used for representing estimation of a delivery effect after the candidate information is delivered to the account;
and selecting a first target number of candidate information with larger first estimation effect parameters from the plurality of candidate information to obtain the first target information.
In an optional implementation, the apparatus further includes:
a fifth module configured to determine a second number of targets according to the estimated click probability;
a sixth module, configured to calculate a second estimated effect parameter of each of the first target information by using a second neural network model, where the second estimated effect parameter is used to represent an estimation of a delivery effect after the first target information is delivered to the account;
and the seventh module is configured to select the first target information with the second estimated effect parameter and the larger second target number from the first target information with the first target number to obtain second target information.
In an optional implementation, the first module is specifically configured to:
extracting second characteristic information of the account;
and when the second characteristic information is matched with the directional delivery information, acquiring candidate information corresponding to the directional delivery information, wherein the directional delivery information is information which is required by a delivery object of the candidate information.
According to a third aspect of the present disclosure, there is provided an electronic apparatus comprising:
a processor;
a memory for storing the processor-executable instructions;
wherein the processor is configured to execute the instructions to implement the information processing method according to the first aspect.
According to a fourth aspect of the present disclosure, there is provided a computer-readable storage medium in which instructions, when executed by a processor of an electronic device, enable the electronic device to perform the information processing method according to the first aspect.
According to a fifth aspect of the present disclosure, there is provided a computer program product, wherein instructions of the computer program product, when executed by a processor of an electronic device, enable the electronic device to perform the information processing method according to the first aspect.
The technical scheme provided by the embodiment of the disclosure at least brings the following beneficial effects:
the technical scheme of the disclosure provides an information processing method, an information processing device, electronic equipment and a storage medium, wherein when an information acquisition request sent by a client is received, a plurality of candidate information are acquired, the information acquisition request comprises an account identifier and a request type, and the request type is the type of each candidate information; acquiring the estimated click probability of the account corresponding to the account identifier on the request type information; then, determining the number of the first targets according to the estimated click probability; and then screening a first target number of candidate information from the plurality of candidate information to obtain first target information. In the process of screening a plurality of candidate information, the number of the screened candidate information, namely the first target number, is determined according to the estimated click probability, the first target number and the estimated click probability form a positive correlation relationship, and for an account with high estimated click probability, a larger number of candidate information can be screened, so that the probability of finally screening high-quality (such as high eCPM) candidate information is improved; for the accounts with low estimated click probability, a smaller amount of candidate information can be screened out, so that the computing resources occupied by the post-screening are saved.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the disclosure and are not to be construed as limiting the disclosure.
Fig. 1 is a flow chart illustrating an information processing method according to an example embodiment.
FIG. 2 is a flow diagram illustrating a method of screening candidate information according to an example embodiment.
Fig. 3 is a block diagram illustrating a structure of an information processing apparatus according to an exemplary embodiment.
FIG. 4 is a block diagram illustrating an electronic device in accordance with an example embodiment.
FIG. 5 is a block diagram illustrating an electronic device in accordance with an example embodiment.
Detailed Description
In order to make the technical solutions of the present disclosure better understood by those of ordinary skill in the art, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings.
It should be noted that the terms "first," "second," and the like in the description and claims of the present disclosure and in the above-described drawings are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the disclosure described herein are capable of operation in sequences other than those illustrated or otherwise described herein. The implementations described in the exemplary embodiments below are not intended to represent all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.
The main task of the advertisement background system (such as a server) is to screen out the advertisement with the highest value in the current advertisement database, wherein the value of the advertisement can be represented by eCPM, and the eCPM is the expected income obtained by displaying the advertisement for thousands of times.
Generally, eCPM is pCTR × Bid, where pCTR is the probability that the advertisement is clicked by the user and Bid is the Bid of the advertiser, i.e., how much money can be auctioned by each advertisement.
In order to deliver the most valuable advertisement to the user, the server screens the advertisement in the advertisement database according to the flow of targeting → recalling → bold row → fine row.
The targeting is to find all advertisements meeting the conditions according to the attributes (male and female, region and the like) of the user and output the advertisements to a recall; the recalling is to adopt a lightweight model to carry out eCPM estimation on each advertisement and screen out the quota with the maximum valuerThe advertisements are output to the rough row; coarse ranking is to adopt a lightweight model to carry out eCPM estimation on each advertisement to screen out the quota with the maximum valuepAdvertising and outputting the advertising to the fine ranks; and fine ranking is to adopt a complex model to carry out eCPM estimation on each advertisement, screen out Q advertisements with the maximum value, return Q (1 or more) advertisements as output results to a client and display the output results to a user. The lightweight models adopted for recalling and coarse typesetting are simple in structure, and can quickly obtain estimated results to complete screening.
In practical application, the screened advertisement quantity quota is recalledrAnd the advertisement quantity quota screened out in a coarse rowpThe larger the screening effect, the better the screening effect, i.e. the larger the probability of finally screening out high-value advertisements, the same asThe more computing resources are required in the screening process. In the related art, quota is generally set according to available computing resourcesrAnd quotapThe values of these two parameters. The inventors found that the same quota is used for all usersrAnd quotapThere is a problem of wasting computing resources, for example, for a user who never clicks any advertisement, still according to the preset quotarAnd quotapValue for advertisement screening is not necessary. Therefore, how to scientifically adjust the information screening quantity in the recalling and rough ranking processes and balance the relationship between the screening effect and the computing resource is a problem to be solved urgently by the technical personnel in the field.
To scientifically set the information filtering amount, fig. 1 is a flowchart illustrating an information processing method according to an exemplary embodiment, which may include the following steps, as shown in fig. 1.
In step S11, when an information acquisition request sent by the client is received, a plurality of candidate information is acquired, the information acquisition request includes an account identifier and a request type, and the request type is a type to which each candidate information belongs.
The execution subject of the embodiment may be an electronic device, such as a server.
For the disclosed embodiments, the candidate information may include audio information, video information, image information, and text information, among others. The request type is the type of each candidate information, and the types of the candidate information can comprise advertisements, short videos, commodities, shops and the like. The present embodiment explains the technical solution by taking the type of the candidate information as an advertisement as an example.
In an optional implementation manner, the step S11 may specifically include: extracting second characteristic information of the account; and when the second characteristic information is matched with the directional delivery information, acquiring candidate information corresponding to the directional delivery information, wherein the directional delivery information is information which is required by a delivery object of the candidate information.
The second characteristic information of the account may include at least one of account attribute information (e.g., gender, age, etc.), account behavior information (e.g., interests, hobbies, etc.), account social relationship information (e.g., friend relationships, etc.), and the like.
When the type of the candidate information is an advertisement, the targeted delivery information is generally set by the advertiser, the advertiser can limit the delivery object of the advertisement by setting the targeted delivery information, and the targeted delivery information can be information that the advertiser selects the set advertisement delivery object according to the advertisement content. The targeted delivery information may include gender, age, etc. of the advertisement delivery subject.
In a specific implementation, when the second feature information matches with the directional delivery information of a certain candidate information, the candidate information is obtained. For example, when the second characteristic information includes "female", "30 +" and the like, and the targeted delivery information is "female", the second characteristic information may be considered to match the targeted delivery information.
In step S12, the estimated click probability of the account corresponding to the account identifier for the request type information is obtained.
In a specific implementation manner, the ratio of the number of clicks to the number of displays of the information of the request type by the account can be calculated according to the historical click data of the information of the request type by the account, and the ratio is used as the estimated click probability.
In another optional implementation manner, step S12 may specifically include: extracting first characteristic information of an account; inputting the first characteristic information into a click rate estimation model obtained through pre-training to obtain estimated click probability of the account for the information of the request type, wherein the click rate estimation model is obtained through training according to the first characteristic information of a plurality of sample accounts and click labels of the sample accounts for the information of the request type, and the click labels are used for indicating whether the sample accounts click the information of the request type.
In the specific implementation, when the request type, that is, the type to which the candidate information belongs, is an advertisement, the estimated click probability of the account on the advertisement is obtained through the step. In the process of training the click-through rate estimation model, the pair < account, whether to click on the displayed advertisement > may be used as a training sample, where click is used as a positive sample and no click is used as a negative sample. The model features retain only the first feature information of the account. The click rate estimation model is used for measuring the probability of clicking any advertisement by one account, if the click probability is high, the account (the flow) is good in quality, otherwise, the account (the flow) is poor in quality.
The first feature information may include at least one of account attribute information (e.g., gender, age, etc.), account behavior information (e.g., interests, hobbies, etc.), account social relationship information (e.g., friend relationships, etc.), and the like.
In step S13, a first number of objects is determined according to the estimated click probability, wherein the first number of objects has a positive correlation with the estimated click probability.
The first target quantity and the estimated click probability form a positive correlation relationship, and a larger quantity of candidate information can be screened out for an account with a high estimated click probability, so that the probability of finally screening high-quality (such as high eCPM) candidate information is improved; for the account with low estimated click probability, a smaller amount of candidate information can be screened out, so that the computing resources occupied by the post-screening are saved. In the embodiment, the first target number of each account is determined according to the estimated click probability, instead of screening the candidate information with the same number for all accounts, so that the delivery quality can be ensured and the calculation resources can be reduced.
The inventor establishes the following mathematical model according to the utility formula:
Figure BDA0002924654240000081
Figure BDA0002924654240000082
Figure BDA0002924654240000083
0≤xir≤1
0≤xip≤1
where i represents the ith flow, i.e., one access to the ith account, quotarA preset upper limit, quota, representing the number of candidate messages recalled to the roughpA preset upper limit value, x, representing the number of candidate messages for coarse ranking to fine rankingirRepresenting the ratio of the actual amount of the ith flow recalled to the coarse row to a preset upper limit value, xipRepresenting the ratio of the actual number of the i-th flow which is coarsely discharged to the fine discharge to the preset upper limit value, CrRepresenting a coarse ranking of the computing resources, CpRepresenting a fine-grained computational resource, alphaiAnd betaiIs a function related to the account (flow i).
In order to increase the probability of screening to high quality candidate information, the method needs to be implemented
Figure BDA0002924654240000084
Maximization, i.e. a preset upper limit value quota for a given recall to a coarse bankrAnd a preset upper limit value quotia for coarse and fine rowspIn this case, the larger the actual number recalled to the coarse-grained and the actual number recalled to the fine-grained, the better the final screening effect, that is, the higher the probability of finally screening the high-quality candidate information.
Solving the mathematical model based on dual conditions can obtain:
Figure BDA0002924654240000085
Figure BDA0002924654240000086
in practical applications, quotarAnd quotapThe numerical value of (a) may be dynamically adjusted according to the indexes such as actual computing resources and availability according to methods such as PID, and the specific numerical value is not limited in this embodiment.
According to the result of the solution,due to sigmaiαi，∑iβi，CrAnd CpAre all constant, so the actual number of i-th flow recalled to the gross is equal to quotar*xrα constant, actual amount of i flow coarsely discharged to fine line quotap*xpβ constant. Since α and β are functions related to accounts, the present embodiment is characterized by the estimated click probability of each account.
In an alternative implementation, the first target number may be determined by one of the following steps, based on the mathematical model described above:
and when the estimated click probability is larger than a first preset threshold value, determining that the number of the first targets is a preset upper limit value.
And when the estimated click probability is smaller than a second preset threshold value, determining the number of the first targets as a preset lower limit value.
And when the estimated click probability is less than or equal to a first preset threshold and greater than or equal to a second preset threshold, determining the number of the first targets according to a preset upper limit value, a preset lower limit value, the estimated click probability, the first preset threshold and the second preset threshold.
The first preset threshold is larger than the second preset threshold.
In a particular implementation, the first target number may be determined according to the following formula:
Figure BDA0002924654240000091
wherein upper _ score represents a first preset threshold, lower _ score represents a second preset threshold, quota represents a preset upper limit, lower _ num represents a preset lower limit, and qs represents the estimated click probability calculated in step S12. The preset upper limit value, the preset lower limit value, the first preset threshold value and the second preset threshold value may be set according to actual requirements, and the specific values are not limited in this embodiment.
By adopting the formula, the first target quantity can be determined in a segmented mode according to the estimated click probability qs, and therefore the candidate information is subjected to personalized screening according to the estimated click probability of each account. For accounts with the estimated click probability qs larger than a first preset threshold upper _ score, the first target number can be set as a preset upper limit value, and the preset upper limit value can be the maximum number of candidate information which can be processed by available computing resources, so that the probability of finally screening out high-quality (such as high eCPM) candidate information is improved; for accounts with the estimated click probability qs smaller than a second preset threshold value lower _ score, the first target number can be set as a preset lower limit value, and computing resources occupied by subsequent screening can be saved as far as possible while high-quality candidate information can be screened; for accounts with the estimated click probability qs being less than or equal to a first preset threshold upper _ score and greater than or equal to a second preset threshold lower _ score, the first target number is determined by a preset upper limit value quota, a preset lower limit value lower _ num, the estimated click probability qs, the first preset threshold upper _ score and the second preset threshold lower _ score, and the relationship between the screening effect and the computing resources can be balanced.
In the recalling and coarse ranking processes, the screening number of the candidate information can be determined by adopting the formula, namely the first target number can represent the actual number of the candidate information recalled to the coarse ranking or the actual number of the candidate information coarsely ranked to the fine ranking.
In the recall process, the quota value in the formula is recalled to the preset upper limit value quota of the rough rowrThe lower _ num value is recalled to the preset lower limit value of the rough rowrAnd the calculated first target number is the actual number of the candidate information recalled to the rough ranking. In the course of rough arrangement, the value of quota in the formula is roughly arranged to the preset upper limit value quota of fine arrangementpAnd the lower-limit value lower _ num is preset and is roughly arranged for fine arrangementpAnd the calculated first target quantity is the actual quantity of the candidate information which is coarsely arranged to be finely arranged.
In step S14, a first target number of candidate information is filtered from the plurality of candidate information, and first target information is obtained.
In a specific implementation, a first target number of candidate information with a larger estimated effect parameter may be screened from the multiple candidate information according to a first estimated effect parameter (e.g., eCPM) of the multiple candidate information, so as to obtain first target information. The screening process will be described in detail in the examples which follow.
According to the information processing method provided by the embodiment of the disclosure, in the process of screening a plurality of candidate information, the number of the screened candidate information, namely the first target number, is determined according to the estimated click probability of an account, and the first target number and the estimated click probability form a positive correlation relationship, for the account with high estimated click probability, a larger number of candidate information can be screened, so that the probability of finally screening high-quality (such as high eCPM) candidate information is improved; for the accounts with low estimated click probability, a smaller amount of candidate information can be screened out, so that the computing resources occupied by the post-screening are saved. The embodiment determines the first target quantity through a scientific method, and can balance the relationship between the screening effect and the computing resources.
In an alternative implementation manner, referring to fig. 2, step S14 may specifically include:
and step S21, calculating a first estimation effect parameter of each candidate message by adopting a first neural network model, wherein the first estimation effect parameter is used for representing estimation of the release effect of the candidate messages released to the account.
The first predicted effect parameter may be eCPM, which is the income that can be obtained by presenting the candidate information for the desired thousand times.
In specific implementation, the feature information of the account and the feature information of the candidate information may be input into the first neural network model, the click probability of the account on the candidate information is output, and the eCPM is calculated according to the click probability of the account on the candidate information. Wherein the first neural network model is a lightweight model.
Step S22, selecting a first target number of candidate information with a larger first estimation effect parameter from the plurality of candidate information to obtain first target information.
In a specific implementation, the candidate information may be sorted according to a descending order of the first pre-estimated effect parameter, and a first target number of candidate information arranged in front of the sequence may be used as the first target information.
In this implementation, the first neural network model is a model used in the recall process, and the first target number is an actual number of candidate information recalled to the rough ranking.
And step S23, determining the number of second targets according to the estimated click probability.
In a specific implementation, the implementation process of step S23 is similar to that of step S13, and is not described here again.
And step S24, calculating a second estimated effect parameter of each first target information by adopting a second neural network model, wherein the second estimated effect parameter is used for representing estimation of the delivery effect after the first target information is delivered to the account.
The second predicted effect parameter may be eCPM, which is expected to show the income obtainable by the first target information thousands of times.
In specific implementation, the feature information of the account and the feature information of the first target information may be input into a second neural network model, the click probability of the account on the first target information is output, and the eCPM is calculated according to the click probability of the account on the first target information. Wherein the second neural network model is a lightweight model.
Step S25, selecting a second number of first target information with a larger second estimated effect parameter from the first number of first target information to obtain second target information.
In a specific implementation, the plurality of first target information may be sorted according to a descending order of the second pre-estimated effect parameter, and the first target information with a second target number arranged in front of the sequence may be used as the second target information.
Fig. 3 is a block diagram illustrating an information processing apparatus according to an example embodiment. Referring to fig. 3, may include:
a first module 31, configured to, when receiving an information acquisition request sent by a client, acquire a plurality of candidate information, where the information acquisition request includes an account identifier and a request type, and the request type is a type to which each of the candidate information belongs;
a second module 32, configured to obtain an estimated click probability of the account corresponding to the account identifier for the information of the request type;
a third module 33, configured to determine a first number of targets according to the estimated click probability, where the first number of targets and the estimated click probability have a positive correlation;
a fourth module 34 configured to filter the candidate information of the first target number from the plurality of candidate information to obtain first target information.
In an alternative implementation, the second module 32 is specifically configured to:
extracting first characteristic information of the account;
inputting the first characteristic information into a click rate estimation model obtained through pre-training to obtain estimated click probability of the account on the information of the request type, wherein the click rate estimation model is obtained through training according to the first characteristic information of a plurality of sample accounts and click labels of the sample accounts on the information of the request type, and the click labels are used for indicating whether the sample accounts click on the information of the request type.
In an alternative implementation, the third module 33 includes:
the first unit is configured to determine that the first target number is a preset upper limit value when the estimated click probability is larger than a first preset threshold value;
the second unit is configured to determine that the first target number is a preset lower limit value when the estimated click probability is smaller than a second preset threshold value;
a third unit, configured to determine the first number of targets according to the preset upper limit, the preset lower limit, the estimated click probability, the first preset threshold and the second preset threshold when the estimated click probability is less than or equal to the first preset threshold and greater than or equal to the second preset threshold.
In an alternative implementation, the third unit is specifically configured to: determining the first target quantity according to the following formula:
Figure BDA0002924654240000121
wherein the upper _ score represents the first preset threshold, the lower _ score represents the second preset threshold, the quota represents the preset upper limit, the lower _ num represents the preset lower limit, the qs represents the estimated click probability, and the first preset threshold is greater than the second preset threshold.
In an alternative implementation, the fourth module 34 is specifically configured to:
calculating a first estimation effect parameter of each candidate information by adopting a first neural network model, wherein the first estimation effect parameter is used for representing estimation of a delivery effect after the candidate information is delivered to the account;
and selecting a first target number of candidate information with larger first estimation effect parameters from the plurality of candidate information to obtain the first target information.
In an optional implementation, the apparatus further includes:
a fifth module configured to determine a second number of targets according to the estimated click probability;
a sixth module, configured to calculate a second estimated effect parameter of each of the first target information by using a second neural network model, where the second estimated effect parameter is used to represent an estimation of a delivery effect after the first target information is delivered to the account;
and the seventh module is configured to select the first target information with the second estimated effect parameter and the larger second target number from the first target information with the first target number to obtain second target information.
In an alternative implementation, the first module 31 is specifically configured to:
extracting second characteristic information of the account;
and when the second characteristic information is matched with the directional delivery information, acquiring candidate information corresponding to the directional delivery information, wherein the directional delivery information is information which is required by a delivery object of the candidate information.
With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.
Fig. 4 is a block diagram of one type of electronic device 800 shown in the present disclosure. For example, the electronic device 800 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, an exercise device, a personal digital assistant, and the like.
Referring to fig. 4, electronic device 800 may include one or more of the following components: a processing component 802, a memory 804, a power component 806, a multimedia component 808, an audio component 810, an input/output (I/O) interface 812, a sensor component 814, and a communication component 816.
The processing component 802 generally controls overall operation of the electronic device 800, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations. The processing component 802 may include one or more processors 820 to execute instructions to perform all or a portion of the steps of any of the information processing methods described in any of the embodiments. Further, the processing component 802 can include one or more modules that facilitate interaction between the processing component 802 and other components. For example, the processing component 802 can include a multimedia module to facilitate interaction between the multimedia component 808 and the processing component 802.
The memory 804 is configured to store various types of data to support operation at the device 800. Examples of such data include instructions for any application or method operating on the electronic device 800, contact data, phonebook data, messages, pictures, videos, and so forth. The memory 804 may be implemented by any type or combination of volatile or non-volatile memory devices, such as Static Random Access Memory (SRAM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.
The power supply component 806 provides power to the various components of the electronic device 800. The power components 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the electronic device 800.
The multimedia component 808 includes a screen that provides an output interface between the electronic device 800 and a user. In some embodiments, the screen may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundary of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 808 includes a front facing camera and/or a rear facing camera. The front-facing camera and/or the rear-facing camera may receive external multimedia data when the device 800 is in an operating mode, such as a shooting mode or a video mode. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.
The audio component 810 is configured to output and/or input audio signals. For example, the audio component 810 includes a Microphone (MIC) configured to receive external audio signals when the electronic device 800 is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals may further be stored in the memory 804 or transmitted via the communication component 816. In some embodiments, audio component 810 also includes a speaker for outputting audio signals.
The I/O interface 812 provides an interface between the processing component 802 and peripheral interface modules, which may be keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to: a home button, a volume button, a start button, and a lock button.
The sensor assembly 814 includes one or more sensors for providing various aspects of state assessment for the electronic device 800. For example, the sensor assembly 814 may detect an open/closed state of the device 800, the relative positioning of components, such as a display and keypad of the electronic device 800, the sensor assembly 814 may also detect a change in the position of the electronic device 800 or a component of the electronic device 800, the presence or absence of user contact with the electronic device 800, orientation or acceleration/deceleration of the electronic device 800, and a change in the temperature of the electronic device 800. Sensor assembly 814 may include a proximity sensor configured to detect the presence of a nearby object without any physical contact. The sensor assembly 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 814 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.
The communication component 816 is configured to facilitate wired or wireless communication between the electronic device 800 and other devices. The electronic device 800 may access a wireless network based on a communication standard, such as WiFi, a carrier network (such as 2G, 3G, 4G, or 5G), or a combination thereof. In an exemplary embodiment, the communication component 816 receives a broadcast signal or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 816 further includes a Near Field Communication (NFC) module to facilitate short-range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.
In an exemplary embodiment, the electronic device 800 may be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, micro-controllers, microprocessors or other electronic components for performing the information processing methods described in any of the embodiments.
In an exemplary embodiment, a computer-readable storage medium comprising instructions, such as the memory 804 comprising instructions, executable by the processor 820 of the electronic device 800 to perform the information processing method of any of the embodiments is also provided. For example, the computer readable storage medium may be a ROM, a Random Access Memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.
In an exemplary embodiment, a computer program product is also provided, which comprises readable program code executable by the processor 820 of the device 800 to perform the information processing method according to any of the embodiments. Alternatively, the program code may be stored in a storage medium of the apparatus 800, which may be a ROM, a Random Access Memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.
Fig. 5 is a block diagram of one type of electronic device 1900 shown in the present disclosure. For example, the electronic device 1900 may be provided as a server.
Referring to fig. 5, electronic device 1900 includes a processing component 1922 further including one or more processors and memory resources, represented by memory 1932, for storing instructions, e.g., applications, executable by processing component 1922. The application programs stored in memory 1932 may include one or more modules that each correspond to a set of instructions. Further, the processing component 1922 is configured to execute instructions to perform the information processing method according to any of the embodiments.
The electronic device 1900 may also include a power component 1926 configured to perform power management of the electronic device 1900, a wired or wireless network interface 1950 configured to connect the electronic device 1900 to a network, and an input/output (I/O) interface 1958. The electronic device 1900 may operate based on an operating system, such as Windows Server, MacOSXTM, UnixTM, LinuxTM, FreeBSDTM, or the like, stored in memory 1932.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims (10)

1. An information processing method, characterized in that the method comprises:
when an information acquisition request sent by a client is received, acquiring a plurality of candidate information, wherein the information acquisition request comprises an account identifier and a request type, and the request type is the type of each candidate information;
obtaining the estimated click probability of the account corresponding to the account identification on the information of the request type;
determining the number of first targets according to the estimated click probability, wherein the number of the first targets and the estimated click probability form a positive correlation;
and screening the candidate information with the first target number from the candidate information to obtain first target information.
2. The information processing method according to claim 1, wherein the step of obtaining the estimated click probability of the account corresponding to the account identifier for the request type of information includes:
extracting first characteristic information of the account;
inputting the first characteristic information into a click rate estimation model obtained through pre-training to obtain estimated click probability of the account on the information of the request type, wherein the click rate estimation model is obtained through training according to the first characteristic information of a plurality of sample accounts and click labels of the sample accounts on the information of the request type, and the click labels are used for indicating whether the sample accounts click on the information of the request type.
3. The information processing method of claim 1, wherein the step of determining the first number of objects based on the estimated click probability comprises:
when the estimated click probability is larger than a first preset threshold value, determining that the first target number is a preset upper limit value;
when the estimated click probability is smaller than a second preset threshold value, determining the number of the first targets as a preset lower limit value;
and when the estimated click probability is smaller than or equal to the first preset threshold and larger than or equal to the second preset threshold, determining the number of the first targets according to the preset upper limit, the preset lower limit, the estimated click probability, the first preset threshold and the second preset threshold.
4. The information processing method according to claim 3, wherein when the estimated click probability is less than or equal to the first preset threshold and greater than or equal to the second preset threshold, the first target number is determined according to the following formula:
Figure FDA0002924654230000011
wherein the upper _ score represents the first preset threshold, the lower _ score represents the second preset threshold, the quota represents the preset upper limit, the lower _ num represents the preset lower limit, the qs represents the estimated click probability, and the first preset threshold is greater than the second preset threshold.
5. The information processing method according to claim 1, wherein the step of filtering the first target number of candidate information from the plurality of candidate information to obtain first target information comprises:
calculating a first estimation effect parameter of each candidate information by adopting a first neural network model, wherein the first estimation effect parameter is used for representing estimation of a delivery effect after the candidate information is delivered to the account;
and selecting a first target number of candidate information with larger first estimation effect parameters from the plurality of candidate information to obtain the first target information.
6. The information processing method according to claim 5, wherein after the step of selecting a first target number of candidate information with a larger first predicted effect parameter from the plurality of candidate information to obtain the first target information, the method further comprises:
determining the number of second targets according to the estimated click probability;
calculating a second estimated effect parameter of each first target information by adopting a second neural network model, wherein the second estimated effect parameter is used for representing estimation of a delivery effect after the first target information is delivered to the account;
and selecting first target information with a second target number with a larger second estimation effect parameter from the first target information with the first target number to obtain second target information.
7. An information processing apparatus characterized in that the apparatus comprises:
the information processing method comprises a first module, a second module and a third module, wherein the first module is configured to obtain a plurality of candidate information when receiving an information obtaining request sent by a client, the information obtaining request comprises an account identifier and a request type, and the request type is a type to which each candidate information belongs;
the second module is configured to acquire the estimated click probability of the account corresponding to the account identifier on the information of the request type;
a third module, configured to determine a first target number according to the estimated click probability, wherein the first target number has a positive correlation with the estimated click probability;
a fourth module configured to filter the candidate information with the first target number from the plurality of candidate information to obtain first target information.
8. An electronic device, characterized in that the electronic device comprises:
a processor;
a memory for storing the processor-executable instructions;
wherein the processor is configured to execute the instructions to implement the information processing method of any one of claims 1 to 6.
9. A computer-readable storage medium whose instructions, when executed by a processor of an electronic device, enable the electronic device to perform the information processing method of any one of claims 1 to 6.
10. A computer program product comprising a computer program, characterized in that the computer program realizes the information processing method according to any one of claims 1 to 6 when executed by a processor.
CN202110129488.8A 2021-01-29 2021-01-29 Information processing method, information processing device, electronic equipment and storage medium Pending CN112767053A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110129488.8A CN112767053A (en) 2021-01-29 2021-01-29 Information processing method, information processing device, electronic equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110129488.8A CN112767053A (en) 2021-01-29 2021-01-29 Information processing method, information processing device, electronic equipment and storage medium

Publications (1)

Publication Number Publication Date
CN112767053A true CN112767053A (en) 2021-05-07

Family

ID=75703978

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110129488.8A Pending CN112767053A (en) 2021-01-29 2021-01-29 Information processing method, information processing device, electronic equipment and storage medium

Country Status (1)

Country Link
CN (1) CN112767053A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113630334A (en) * 2021-06-30 2021-11-09 北京三快在线科技有限公司 Network object processing method and device, electronic equipment and readable storage medium
CN114896475A (en) * 2022-06-08 2022-08-12 北京达佳互联信息技术有限公司 Medium information processing method, medium information processing device, electronic equipment and storage medium

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104965890A (en) * 2015-06-17 2015-10-07 深圳市腾讯计算机***有限公司 Advertisement recommendation method and apparatus
CN106055617A (en) * 2016-05-26 2016-10-26 乐视控股（北京）有限公司 Data pushing method and device
CN106250499A (en) * 2016-08-02 2016-12-21 合网络技术（北京）有限公司 A kind of video is to method for digging and device
CN108228784A (en) * 2017-12-28 2018-06-29 暴风集团股份有限公司 Video recommendation method and device, electronic equipment, storage medium, program
CN110782286A (en) * 2019-10-25 2020-02-11 上海喜马拉雅科技有限公司 Advertisement pushing method and device, server and computer readable storage medium
CN111310040A (en) * 2020-02-11 2020-06-19 腾讯科技（北京）有限公司 Artificial intelligence based recommendation method and device, electronic equipment and storage medium
CN111369271A (en) * 2018-12-25 2020-07-03 北京达佳互联信息技术有限公司 Advertisement sorting method and device, electronic equipment and storage medium
CN111768244A (en) * 2020-06-30 2020-10-13 深圳前海微众银行股份有限公司 Advertisement delivery recommendation method and device

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104965890A (en) * 2015-06-17 2015-10-07 深圳市腾讯计算机***有限公司 Advertisement recommendation method and apparatus
CN106055617A (en) * 2016-05-26 2016-10-26 乐视控股（北京）有限公司 Data pushing method and device
CN106250499A (en) * 2016-08-02 2016-12-21 合网络技术（北京）有限公司 A kind of video is to method for digging and device
CN108228784A (en) * 2017-12-28 2018-06-29 暴风集团股份有限公司 Video recommendation method and device, electronic equipment, storage medium, program
CN111369271A (en) * 2018-12-25 2020-07-03 北京达佳互联信息技术有限公司 Advertisement sorting method and device, electronic equipment and storage medium
CN110782286A (en) * 2019-10-25 2020-02-11 上海喜马拉雅科技有限公司 Advertisement pushing method and device, server and computer readable storage medium
CN111310040A (en) * 2020-02-11 2020-06-19 腾讯科技（北京）有限公司 Artificial intelligence based recommendation method and device, electronic equipment and storage medium
CN111768244A (en) * 2020-06-30 2020-10-13 深圳前海微众银行股份有限公司 Advertisement delivery recommendation method and device

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113630334A (en) * 2021-06-30 2021-11-09 北京三快在线科技有限公司 Network object processing method and device, electronic equipment and readable storage medium
CN114896475A (en) * 2022-06-08 2022-08-12 北京达佳互联信息技术有限公司 Medium information processing method, medium information processing device, electronic equipment and storage medium

Similar Documents

Publication Publication Date Title
CN111431727B (en) Group display method, device, terminal, server and system
CN107992604B (en) Task item distribution method and related device
CN107463643B (en) Barrage data display method and device and storage medium
CN112131410A (en) Multimedia resource display method, device, system and storage medium
CN111369271B (en) Advertisement ordering method and device, electronic equipment and storage medium
CN111556366A (en) Multimedia resource display method, device, terminal, server and system
CN109714643B (en) Recommendation method, system, server and storage medium for video data
CN109670632B (en) Advertisement click rate estimation method, advertisement click rate estimation device, electronic device and storage medium
CN112508612B (en) Method for training advertisement creative generation model and generating advertisement creative and related device
CN111859020A (en) Recommendation method and device, electronic equipment and computer-readable storage medium
CN113727200A (en) Video abstract information determination method and device, electronic equipment and storage medium
CN112767053A (en) Information processing method, information processing device, electronic equipment and storage medium
CN112131466A (en) Group display method, device, system and storage medium
CN112148923A (en) Search result sorting method, sorting model generation method, device and equipment
CN112685641B (en) Information processing method and device
CN112784151A (en) Method and related device for determining recommendation information
CN113553509B (en) Content recommendation method and device, electronic equipment and storage medium
CN112308588A (en) Advertisement putting method and device and storage medium
CN114615516A (en) Data processing method and device, electronic equipment and storage medium
CN113709572A (en) Content promotion method and device, electronic equipment and storage medium
CN109165963B (en) Data prediction method and device and electronic equipment
CN113190725B (en) Object recommendation and model training method and device, equipment, medium and product
CN110674416A (en) Game recommendation method and device
CN115484471B (en) Method and device for recommending anchor
CN111898028B (en) Entity object recommendation method, device and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination