CN111210258A - Advertisement putting method and device, electronic equipment and readable storage medium - Google Patents
Advertisement putting method and device, electronic equipment and readable storage medium Download PDFInfo
- Publication number
- CN111210258A CN111210258A CN201911341189.XA CN201911341189A CN111210258A CN 111210258 A CN111210258 A CN 111210258A CN 201911341189 A CN201911341189 A CN 201911341189A CN 111210258 A CN111210258 A CN 111210258A
- Authority
- CN
- China
- Prior art keywords
- advertisement
- user
- candidate
- recall
- candidate advertisement
- 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
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
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present disclosure provides an advertisement delivery method, an advertisement delivery device, an electronic device and a readable storage medium, wherein the method comprises the following steps: acquiring a user tag corresponding to a target user from a user tag set; acquiring a candidate advertisement set of a target merchant, wherein the candidate advertisement set comprises: the method comprises the steps that candidate advertisements and advertisement label sets of the candidate advertisements are generated through a recall template and recall materials corresponding to each pit site in the recall template, the recall template is obtained from a template data set, and the recall materials are obtained from a material data set; for each candidate advertisement, determining the matching degree of the target user and the candidate advertisement according to the user label and the advertisement label set of the candidate advertisement; and obtaining the candidate advertisement with the highest matching degree with the target user from the candidate advertisement set of the target merchant so as to deliver the candidate advertisement to the target user. The present disclosure may improve the diversity of the candidate advertisement sets.
Description
Technical Field
The present disclosure relates to the field of personalized recommendation technologies, and in particular, to an advertisement delivery method, an advertisement delivery device, an electronic device, and a readable storage medium.
Background
In the technical field of personalized recommendation, objects such as businesses, commodities and the like can be generally recommended to a browsing user in the form of advertisements. Because the recommended object is generally generated for the user, the system considers that the user may be interested in the recommended object, so that the user can conveniently and directly place an order for the merchant or the commodity, and the order placing rate is improved.
In the prior art, a commonly used advertisement delivery method includes: firstly, labeling user behaviors based on big data analysis; then, acquiring advertisement position code information loaded when a user accesses a website page, and extracting user characteristic information to return the advertisement position code information and the user characteristic information to an online advertisement accurate delivery system; and finally, the online accurate advertisement putting system selects target advertisements from the artificially pre-made candidate advertisement set according to the real-time bidding and interest weight of the advertisements so as to put the advertisements.
After the inventor researches the scheme, the inventor finds that the candidate advertisement set adopted by the advertisement putting method is artificially preset and lacks diversity.
Disclosure of Invention
The disclosure provides an advertisement delivery method, an advertisement delivery device, an electronic device and a readable storage medium, which can dynamically generate a candidate advertisement set according to a large number of recall templates and recall materials, and are beneficial to improving the diversity of the candidate advertisement set.
According to a first aspect of the present disclosure, there is provided an advertisement delivery method, the method comprising:
acquiring a user tag corresponding to a target user from a user tag set;
acquiring a candidate advertisement set of a target merchant, wherein the candidate advertisement set comprises: the method comprises the steps that candidate advertisements and advertisement label sets of the candidate advertisements are generated through a recall template and recall materials corresponding to each pit site in the recall template, the recall template is obtained from a preset template data set, and the recall materials are obtained from a preset material data set;
for each candidate advertisement, determining the matching degree of the target user and the candidate advertisement according to the user label and the advertisement label set of the candidate advertisement;
and obtaining the candidate advertisement with the highest matching degree with the target user from the candidate advertisement set of the target merchant so as to deliver the candidate advertisement to the target user.
According to a second aspect of the present disclosure, there is provided an advertisement delivery apparatus, the apparatus comprising:
the user tag acquisition module is used for acquiring a user tag corresponding to a target user from the user tag set;
a candidate advertisement set obtaining module, configured to obtain a candidate advertisement set of a target merchant, where the candidate advertisement set includes: the method comprises the steps that candidate advertisements and advertisement label sets of the candidate advertisements are generated through a recall template and recall materials corresponding to each pit site in the recall template, the recall template is obtained from a preset template data set, and the recall materials are obtained from a preset material data set;
a matching degree determination module, configured to determine, for each candidate advertisement, a matching degree between the target user and the candidate advertisement according to the user tag and the advertisement tag set of the candidate advertisement;
and the advertisement delivery module is used for acquiring the candidate advertisement with the highest matching degree with the target user from the candidate advertisement set of the target merchant so as to deliver the candidate advertisement to the target user.
According to a third aspect of the present disclosure, there is provided an electronic device comprising:
a processor, a memory and a computer program stored on the memory and executable on the processor, the processor implementing the aforementioned advertisement delivery method when executing the program.
According to a fourth aspect of the present disclosure, there is provided a readable storage medium, wherein instructions, when executed by a processor of an electronic device, enable the electronic device to perform the aforementioned advertisement delivery method.
The disclosure provides an advertisement delivery method, an advertisement delivery device, an electronic device and a readable storage medium, which can firstly obtain a user tag corresponding to a target user from a user tag set; then, a candidate advertisement set of the target merchant is obtained, wherein the candidate advertisement set comprises: the method comprises the steps that candidate advertisements and advertisement label sets of the candidate advertisements are generated through a recall template and recall materials corresponding to each pit site in the recall template, the recall template is obtained from a preset template data set, and the recall materials are obtained from a preset material data set; then aiming at each candidate advertisement, determining the matching degree of the target user and the candidate advertisement according to the user label and the advertisement label set of the candidate advertisement; and finally, acquiring the candidate advertisement with the highest matching degree with the target user from the candidate advertisement set of the target merchant so as to deliver the candidate advertisement to the target user. The method and the device can dynamically generate the candidate advertisement set according to a large number of recall templates and recall materials, and are favorable for improving the diversity of the candidate advertisement set.
Drawings
In order to more clearly illustrate the technical solutions of the present disclosure, the drawings needed to be used in the description of the present disclosure will be briefly introduced below, and it is apparent that the drawings in the following description are only some embodiments of the present disclosure, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without inventive labor.
FIG. 1 shows a flow diagram of steps of a method of advertisement delivery in one embodiment of the present disclosure;
FIG. 2 shows a relationship diagram of a backend server and various databases, clients of the present disclosure;
FIG. 3 illustrates a flowchart of the steps of the present disclosure to generate a user tag set;
FIG. 4 illustrates a flowchart of the steps of generating a recall template of the present disclosure;
FIG. 5 shows a flowchart of the steps of the present disclosure to generate recall material;
FIG. 6 is a flowchart illustrating the steps of the present disclosure for determining a degree of match of a target user with a candidate advertisement;
FIG. 7 illustrates a flow chart of steps of the present disclosure for delivering advertisements to targeted users;
FIG. 8 shows a block diagram of an advertisement delivery device in an embodiment of the present disclosure;
FIG. 9 illustrates a block diagram of the module for generating a user tag set of the present disclosure;
FIG. 10 illustrates a block diagram of a candidate advertisement set acquisition module for generating a recall template of the present disclosure;
FIG. 11 illustrates a block diagram of a candidate advertisement set acquisition module for generating recall material of the present disclosure;
FIG. 12 is a block diagram illustrating a match determination module of the present disclosure;
FIG. 13 illustrates a block diagram of an advertisement delivery module of the present disclosure;
fig. 14 shows a block diagram of an electronic device of the present disclosure.
Detailed Description
The technical solutions in the present disclosure will be described clearly and completely with reference to the accompanying drawings in the present disclosure, and it is obvious that the described embodiments are some, not all embodiments of the present disclosure. All other embodiments, which can be derived by a person skilled in the art from the embodiments disclosed herein without making any creative effort, shall fall within the protection scope of the present disclosure.
Referring to fig. 1, a flowchart illustrating steps of an advertisement delivery method in an embodiment of the present disclosure is shown, specifically as follows:
The user tag set includes user tags of a large number of users, and the user tags may be generated by a machine or added manually. Specifically, the user tag set may be stored in any device or cache having a storage function, for example, the user tag set may be stored in a user tag database in the form of a data table, and the user identifier is used as the index information, and the user tag is used as the storage information, so that the corresponding user tag may be directly obtained from the user tag database according to the user identifier of the target user.
Wherein the target users are active users on the network sales platform, including but not limited to: a user browsing a web page, a user placing an order, a user logging in. For example, for a user USERA accessing a network sales platform, a user identifier is firstly extracted from user information of the USERA; and then, acquiring the user label corresponding to the user identification from the user label set.
Wherein the advertisement tag set is a number of tags for the candidate advertisement. It is to be understood that since the candidate advertisement in the present disclosure is generated by the recall template and the recall material, the template tag of the recall template and the material tag of the recall material can be taken as the tag set of the candidate advertisement. The template tags and the material tags can be manually configured through an operation module, and can also be extracted from the recall templates and the recall materials through a machine. For example, if the template tags of the recall template include: LBL1, LBL2, LBL3, LBL4, the material tag of the recall material includes: LBL5, LBL6, LBL7, LBL8, LBL9, LBL10, such that the recall template and the set of advertisement tags of candidate advertisements generated by the recall material include: LBL1, LBL2, LBL3, LBL4, LBL5, LBL6, LBL7, LBL8, LBL9, LBL 10.
The template data set comprises a large number of templates, each template comprises a plurality of pit positions, each pit position can be filled with materials, and the materials are used as candidate advertisements after being filled with the materials. The template may be pre-configured by the operation module, and when configuring the template, the merchant condition using the template may be set for the template, for example, which type of merchant or which merchant the template is applicable to is specified. Therefore, whether the target merchant meets the merchant condition can be judged, and if so, the template can be used as a recall template of the target merchant. In addition, when the template is configured, a material recall condition for filling each pit can be set, for example, the material type of one pit is designated as characters, and the number of the characters is 8 at most.
The material data set contains a large number of materials which are used for filling pit positions of the template. The material may be pre-configured by the operation module, and when the material is configured, the attribute of the material may be set, for example, the type of the material is text, picture, video, or the like, and when the type of the material is text, the number of the text may also be set. The attributes of the materials are used for matching the material recall conditions of the pit positions, so that the materials with the attributes meeting the material recall conditions of the pit positions are obtained from the material data set and serve as the recall materials of the pit positions.
After the recall template and the recall material are obtained, the recall material can be filled into the corresponding pit positions in the recall template to obtain a candidate advertisement set. For example, for a target merchant, first, a template that satisfies the merchant conditions of the target merchant is obtained: template TEP1 and template TEP2, template TEP1 includes two pit positions PIP1 and PIP2, template TEP2 includes three pit positions PIP3, PIP4 and PIP 5; then, the material that satisfies the material recall condition of the pit position PIP1 is recalled: SOM1, SOM2, recall the material that satisfies the material recall condition of pit position PIP 2: SOM3, recalling the material satisfying the material recall condition of pit PIP 3: SOM4, recalling the material satisfying the material recall condition of pit PIP 4: SOM5, SOM6, recall the material that satisfies the material recall condition of pit position PIP 5: SOM 7; and finally, filling the recall material into corresponding pits in the recall template to obtain a candidate advertisement set of the target merchant: ADV1, ADV2, ADV3, ADV 4.
The ADV1 is obtained by filling the material SOM1 and the material SOM3 with pit positions PIP1 and PIP2 in a template TEP1, the ADV2 is obtained by filling the material SOM2 and the material SOM3 with pit positions PIP1 and PIP2 in a template TEP1, the ADV3 is obtained by filling the material SOM4, the material SOM5 and the material SOM7 with pit positions PIP3, PIP4 and PIP5 in a template TEP2, and the ADV4 is obtained by filling the material SOM4, the material SOM6 and the material SOM7 with pit positions PIP3, PIP4 and PIP5 in a template TEP 2.
It should be noted that the generation processes of the recall template, the recall material and the candidate advertisement may be completed offline, that is, completed in advance before the advertisement is delivered, and the obtained candidate advertisement set is stored in any device or cache with a storage function, so that the step 102 may directly query the candidate advertisement set of the target merchant. For example, the candidate advertisement set may be stored in the advertisement database in the form of a data table, with the merchant identifier as the index information and the candidate advertisement set as the storage information, so that the corresponding candidate advertisement set may be directly obtained from the advertisement database according to the merchant identifier of the target merchant.
It can be appreciated that generating the set of candidate advertisements offline helps to improve advertisement placement efficiency relative to generating the set of candidate advertisements in real-time.
Step 103, for each candidate advertisement, determining the matching degree between the target user and the candidate advertisement according to the user tag and the advertisement tag set of the candidate advertisement.
Specifically, the matching degree of the user tag and the advertisement tag in the advertisement tag set can be predicted first; then, the matching degree of the user label and the advertisement label is directly used as the matching degree of the target user and the candidate advertisement, and linear or nonlinear transformation can also be carried out on the matching degree of the user label and the candidate advertisement to obtain the matching degree of the target user and the candidate advertisement.
And 104, acquiring the candidate advertisement with the highest matching degree with the target user from the candidate advertisement set of the target merchant so as to deliver the candidate advertisement to the target user.
The method can be applied to a background server of the network sales application, the network sales application is installed on electronic equipment as a client, when the client detects a user who browses, logs in and places an order, the client takes the user as a target user, generates an advertisement putting request carrying a user identifier of the target user and sends the advertisement putting request to the background server, the background server puts advertisements to the target user through steps 101 to 104, and the client can display the put advertisements on an interface of the client.
Referring to fig. 2, a relationship diagram of the background server and each database and client of the present disclosure is shown. As shown in fig. 2, the background server of the present disclosure may be connected to the client, the user tag database storing the user tag set, the template database storing the template database, the material database storing the material database, and the advertisement database storing the candidate advertisement set, where the background server may access the user tag database to obtain the user tag according to the user identifier of the target user, and may also access the user template database to obtain the recall template; the background server can also access the material database to obtain the recall material; the background server can also access the advertisement database to obtain a candidate advertisement set of the target merchant according to the merchant identification of the target merchant.
In another embodiment of the present disclosure, reference is made to FIG. 3, which illustrates a flowchart of the steps of the present disclosure for generating a user tab set. As shown in fig. 3, the user tag set in step 101 is generated in advance through the following steps 105 to 107:
step 105, obtaining at least one online user and user characteristics of the online user, where the online user includes an active user in a current time period, and the active user includes the target user.
The online users may be periodically acquired active users, and the active users may include but are not limited to: users with click behavior, users with login behavior, and users with ordering behavior. For example, a user who has click action, login action, and order placing action on the current day may be acquired as an online user every day.
In addition, the online user may also be an active user detected in real time, for example, if it is detected that a user has click behavior, login behavior, and order placing behavior at a certain time in a day, the user is also regarded as the online user, so that the user tag may be analyzed online through the storm task.
After the online user is obtained, user characteristics of the online user also need to be obtained, which may include but are not limited to: behavior characteristics and attribute characteristics. The behavior characteristics can be formed by historical behavior sequences and represent the historical behavior preference of the user; the attribute features may include, but are not limited to: gender, average amount of money consumed, historical frequency of logging in, historical amount of units.
Step 106, inputting the user characteristics of the online users into a user label prediction model for each online user to obtain user labels of the online users; the user label prediction model is obtained by training at least one second sample, and the second sample comprises: a sample user characteristic, a second sample user label.
The user tag prediction model may be any machine learning model, such as a DNN model. The user label prediction model can adopt multiple iterations when being trained, in the first iteration, the user characteristics in each second sample can be input into the user label prediction model to obtain a trained user label, so that a loss value can be calculated according to the trained user label and the second sample user label, if the loss value does not meet a preset condition, the parameters of the user label prediction model are adjusted according to the loss value to carry out the second iteration, the multiple iterations are carried out until the loss value meets the preset condition, the obtained user label prediction model is a trained model, and the accuracy of the prediction result reaches the requirement.
The LOSS1 in the training process can be calculated by using any existing LOSS function, such as a square LOSS function, a cross entropy LOSS function, an absolute value LOSS function, and the like, and the specific form of the LOSS function is not limited by the present disclosure. How to calculate LOSS1 is illustrated below by taking the square LOSS function as an example:
where M is the number of second samples in each iteration, SULmFor second exemplar user labels in the m second exemplar, TULmAnd training user labels corresponding to the mth second sample.
It can be understood that the present disclosure can periodically acquire online users and store their user tags in a user tag set; in addition, online users can be detected in real time, and the user tags of the online users can be updated to the user tag set, so that the user tags in the user tag set cover more users.
The embodiment of the disclosure can adopt the user label prediction model to predict the user label of the active user in the current time period, thereby ensuring that the user label is continuously updated and being beneficial to improving the accuracy of the user label.
In another embodiment of the present disclosure, the target merchant in step 102 is predetermined by the following step 108:
and 108, determining the successful bidding merchants according to the bidding decision.
The bidding decision is used for selecting merchants who bid successfully from the bidding merchants aiming at one advertising slot, and the merchant with the highest price can be selected from the bidding merchants as the merchant who bid successfully based on the goal of interest maximization. For example, for a merchant with three bids: SLR1, SLR2, SLR3, wherein the bid of SLR1 is 30 yuan, the bid of SLR2 is 20 yuan, and the bid of SLR3 is 10 yuan, so that SLR1 is determined as a merchant with successful bidding.
The embodiment of the disclosure can be applied to a bidding scene of a merchant on an advertisement space, and for the merchant who successfully bids, the candidate advertisement with the highest matching degree with the target user is determined from the candidate advertisement set of the merchant so as to be displayed to the target user at the advertisement space. When the user clicks the advertisement in the advertisement position, the user can enter the interface of the merchant, so that the user can browse the commodities in the interface of the merchant and then place an order.
In another embodiment of the present disclosure, the candidate advertisement in step 102 is generated offline through a recall template and recall material corresponding to each pit site in the recall template, the recall template in step 102 is obtained offline from a preset template data set, the recall material in step 102 is obtained offline from a preset material data set, and step 102 in fig. 1 includes sub-step 1021:
and a substep 1021, obtaining a candidate advertisement set of the target merchant from a preset advertisement database, wherein the candidate advertisement set of the target merchant is stored into the advertisement database in an offline manner.
The advertisement database is used for storing candidate advertisement sets of a large number of users, and the large number of users comprise target users. The candidate advertisement sets for the plurality of users may be generated offline from recall templates obtained offline from a preset template dataset and recall materials obtained offline from the preset material dataset corresponding to each pit site in the recall templates for each user.
The embodiment of the disclosure can generate a large number of candidate advertisement sets of users in an off-line manner, and store the candidate advertisement sets into the advertisement database so as to directly inquire when the advertisements are delivered, thereby being beneficial to reducing the time consumed by the advertisement delivery and improving the efficiency of the advertisement delivery.
In another embodiment of the present disclosure, the candidate advertisement in step 102 is generated in real time through a recall template and recall material corresponding to each pit site in the recall template, the recall template in step 102 is obtained from a preset template data set in real time, the recall material in step 102 is obtained from a preset material data set in real time, step 102 in fig. 1 includes sub-steps 1022:
sub-step 1022, adding the candidate advertisement generated in real time to the set of candidate advertisements for the target merchant.
In a certain embodiment of the disclosure, if the candidate advertisement set of the target merchant does not exist in the advertisement database, the recall template and the recall material may be obtained in real time, and the candidate advertisement set of the target merchant may be generated in real time, so as to ensure that the advertisement of the target merchant is successfully delivered to the target user.
In another embodiment of the present disclosure, referring to FIG. 4, a flowchart illustrating the steps of the present disclosure to generate a recall template is shown. As shown in fig. 4, the recalled template in step 102 is obtained from a preset template data set through sub-steps 1023 to 1024 as follows:
and a substep 1023, obtaining the business type of the merchant according to a preset time period for each merchant in a preset merchant set, wherein the merchant set comprises the target merchant.
The preset merchant set may be all merchants registered in the network sales application, and each merchant may set its service type during registration. The business type is used to specify the type of goods or services offered by the merchant, including but not limited to: toys for children, snacks, desserts, men's clothing, women's clothing.
It can be understood that since there may be changes to the merchants registered on the network sales platform, periodic acquisition is required to ensure the latest business types of the merchants. For example, the business type of the merchant may be acquired once a week or once a day.
And a substep 1024, for each merchant, obtaining a template corresponding to the service type from the template data set, and using the template as a recall template of the merchant.
Each template in the template data set can be set with the corresponding service type during configuration, so that a template corresponding to the same service type as a merchant can be recalled and used as a recall template of the merchant. For example, for a merchant, the service type of the merchant is STP obtained through step D1, so that all templates with the service type of STP in the template dataset may be used as recall templates of the merchant, for example, if the template with the service type of STP in the template dataset includes: TEP1, TEP2, …, TEP20, whereby obtaining the merchant's recall template comprises: TEP1, TEP2, …, TEP 20.
The method and the device can determine the recall template for each merchant according to the service type, so that each merchant corresponds to a large number of recall templates, and the recall templates are suitable for the type of merchant, so that the variety of the candidate advertisements of the merchant can be ensured, and meanwhile, the applicability of the recall templates can be ensured.
In another embodiment of the present disclosure, referring to FIG. 5, a flowchart of the steps of the present disclosure to generate recall material is shown. As shown in fig. 5, the recalled material in step 102 is obtained through sub-steps 1025 to 1026 as follows:
and a substep 1025 of obtaining a material recall condition of each pit in the recall template for each recall template of each merchant.
The material recall condition is a condition that the material filled in the pit site should meet, and the material recall condition may be configured in advance. For example, for a pit, the material recall condition may be: pictures, size below 128 x 526.
And a substep 1026 of, for each pit position, obtaining a material meeting the material recall condition of the pit position from the material data set, and using the material as the recall material corresponding to the pit position.
According to the method and the device, a large number of materials can be recalled for each pit site according to the material recall condition, so that diversified candidate advertisements are generated, and the materials of each pit site are in line with the recall condition of the pit site, so that the quality of the candidate advertisements is improved.
In another embodiment of the present disclosure, the advertisement tag set of each candidate advertisement includes at least one advertisement tag, and referring to fig. 6, it shows a flowchart of the steps of the present disclosure for determining the matching degree between the target user and the candidate advertisement. As shown in fig. 6, step 103 in fig. 1 includes sub-steps 1031 to 1032:
The label matching degree prediction model may be any machine learning model, for example, a DNN (Deep neural networks) model. The label matching degree prediction model can carry out multiple rounds of iteration during training, for each first sample in each round of iteration, a first sample user label and a sample advertisement label in the first sample can be input into the label matching degree prediction model, the training matching degree corresponding to the first sample and the training weight of the sample advertisement label are obtained through prediction, after the first round of iteration, a loss value is calculated according to the training matching degree and the training weight, and if the loss value meets a preset condition, the training is ended; and if the loss value does not meet the preset condition, adjusting parameters of the label matching degree prediction model according to the loss value so as to carry out a second iteration, wherein the second iteration can be trained by adopting a new batch of first samples, and carrying out multiple iterations until the loss value meets the preset condition, wherein the obtained label matching degree prediction model is a trained model, and the accuracy of the prediction result meets the requirement.
The LOSS value LOSS4 in the training process can be calculated by using the following formula:
LOSS4＝a1·LOSS2+a2·LOSS3 (2)
wherein LOSS2 is the LOSS value of the sample matching degree and the training matching degree, LOSS3 is the matching degree of the sample weight and the training weight, a1And a2Weights of LOSS2 and LOSS3, respectively, so that the degree of influence of LOSS2 and LOSS3 on the LOSS value, a, can be adjusted1And a2Values greater than 0 may be taken.
It is understood that LOSS2 and LOSS3 in formula (1) can be calculated by using any LOSS function, such as a square LOSS function, a cross entropy LOSS function, an absolute value LOSS function, etc., and the specific form of the LOSS function is not limited by the present disclosure. The following describes how LOSS2 and LOSS3 are calculated using a squared LOSS function as an example:
where N is the number of first samples taken per iteration, SMDnFor the sample match degree, TMD, in the nth first samplenFor training match degree corresponding to the n-th first sample, SWnSample weights, TW, for sample advertisement tags in the nth first samplenAnd advertising training weights corresponding to the labels for the samples in the nth first sample.
And a substep 1032, for each candidate advertisement, performing a weighted operation according to the matching degree between the user tag and the advertisement tag of the candidate advertisement and the weight of the advertisement tag to obtain the matching degree between the target user and the candidate advertisement.
Specifically, the matching degree MAD2 of the target user and the ith candidate advertisementiCan be calculated according to the following formula:
where J is the number of ad tags for the ith candidate ad, MAD1i,jIs the matching degree, p, of the user label and the jth advertisement label of the ith candidate advertisementi,jThe weight of the jth advertisement tag of the ith candidate advertisement.
The embodiment of the disclosure can predict the matching degree of the user label and the advertisement label and the weight of each advertisement label through the machine learning model, and is beneficial to improving the accuracy of the matching degree.
In another embodiment of the present disclosure, reference is made to FIG. 7, which illustrates a flowchart of the steps of the present disclosure for delivering advertisements to targeted users. As shown in fig. 7, step 104 in fig. 1 includes sub-steps 1041 to 1043:
sub-step 1041, obtaining, for each candidate advertisement in the set of candidate advertisements for the targeted merchant, a quality score for the candidate advertisement, the quality score for the candidate advertisement being determined from at least one of a quality score for a recall template of the candidate advertisement and a quality score for a recall material of the candidate advertisement.
The quality score of the candidate advertisement may be a quality score of a recall template used for generating the candidate advertisement, a quality score of a recall material used for generating the candidate advertisement, or a weighted sum of the quality score of the recall template used for generating the candidate advertisement and the quality score of the recall material, that is: firstly, calculating the product of the quality score of the recall template and the weight of the recall template to obtain the weighted quality score of the recall template; then calculating the product of the quality score of the recalled material and the weight of the recalled material to obtain the weighted quality score of the recalled material; and finally, calculating the sum of the weighted quality score of the recall template and the weighted quality score of the recall material to obtain the quality score of the candidate advertisement. The quality score of the recall template can be the click rate of the user on the recall template, and the quality score of the recall material can be the click rate of the user on the recall material.
Specifically, firstly, periodically acquiring an advertisement use log; then, carrying out Spark distributed ETL (Extract-Transform-Load) processing on the advertisement use log to obtain effective advertisement use data; and finally, counting the click rate of each template in the template data set and the click rate of each material in the material data set from the effective advertisement use data, saving the click rate of each template as the quality score of each template into the template data set, and saving the click rate of each material as the quality score of each material into the material data set, so that the corresponding quality score can be directly obtained according to the template identifier of the recall template when the substep 1041 is carried out, and the corresponding quality score can be obtained according to the material identifier of the recall material.
Sub-step 1042, for each candidate advertisement in the candidate advertisement set of the target merchant, determining a composite score of the candidate advertisement according to the quality score of the candidate advertisement and the matching degree of the target user and the candidate advertisement.
Wherein, the comprehensive score of the candidate advertisement and the quality score of the candidate advertisement,The relationship between the degree of match of the target user with the candidate advertisement may be as follows: if the quality score of the candidate advertisement is larger, the matching degree of the target user and the candidate advertisement is larger, and the comprehensive score of the candidate advertisement is larger; if the quality score of the candidate advertisement is smaller, the matching degree of the target user and the candidate advertisement is smaller, and the comprehensive score of the candidate advertisement is smaller. Based on the above relationship, the comprehensive score TTS of the ith candidate advertisement can be calculated according to the following formulai：
TTSi＝c1·MAD2i+c2·QTSi(6)
Wherein, QTSiThe quality score, c, of the ith candidate advertisement determined for substep 10411And c2Weights for degree of match and quality score, respectively, c1And c2Are all values greater than 0.
Sub-step 1043, obtaining the candidate advertisement with the highest comprehensive score from the candidate advertisement set of the target merchant, so as to deliver to the target user.
The embodiment of the disclosure can put the advertisement to the target user by combining the quality score and the matching degree, so that the put advertisement has higher matching degree with the user and higher quality.
In summary, the present disclosure provides an advertisement delivery method, including: acquiring a user tag corresponding to a target user from a user tag set; acquiring a candidate advertisement set of a target merchant, wherein the candidate advertisement set comprises: the method comprises the steps that candidate advertisements and advertisement label sets of the candidate advertisements are generated through a recall template and recall materials corresponding to each pit site in the recall template, the recall template is obtained from a preset template data set, and the recall materials are obtained from a preset material data set; for each candidate advertisement, determining the matching degree of the target user and the candidate advertisement according to the user label and the advertisement label set of the candidate advertisement; and obtaining the candidate advertisement with the highest matching degree with the target user from the candidate advertisement set of the target merchant so as to deliver the candidate advertisement to the target user. The method and the device can dynamically generate the candidate advertisement set according to a large number of recall templates and recall materials, and are favorable for improving the diversity of the candidate advertisement set.
Referring to fig. 8, a block diagram of an advertisement delivery device in one embodiment of the present disclosure is shown. As shown in fig. 8, the advertisement delivery apparatus includes:
a user tag obtaining module 201, configured to obtain a user tag corresponding to a target user from a user tag set.
A candidate advertisement set obtaining module 202, configured to obtain a candidate advertisement set of a target merchant, where the candidate advertisement set includes: the method comprises the steps of candidate advertisements and advertisement label sets of the candidate advertisements, wherein the candidate advertisements are generated through recall templates and recall materials corresponding to all pit places in the recall templates, the recall templates are obtained from preset template data sets, and the recall materials are obtained from preset material data sets.
A matching degree determining module 203, configured to determine, for each candidate advertisement, a matching degree between the target user and the candidate advertisement according to the user tag and the advertisement tag set of the candidate advertisement.
An advertisement delivery module 204, configured to obtain, from the candidate advertisement set of the target merchant, a candidate advertisement with the highest matching degree with the target user, so as to deliver the candidate advertisement to the target user.
In another embodiment of the present disclosure, reference is made to FIG. 9, which illustrates a block diagram of the module for generating a user tag set of the present disclosure. As shown in fig. 9, the user tag set in the user tag obtaining module 201 is generated in advance by the following user feature obtaining module 205, user tag predicting module 206 and user tag set generating module 207:
a user characteristic obtaining module 205, configured to obtain at least one online user and a user characteristic of the online user, where the online user includes an active user in a current time period, and the active user includes the target user.
A user tag prediction module 206, configured to, for each online user, input the user characteristics of the online user into a user tag prediction model to obtain a user tag of the online user; the user label prediction model is obtained by training at least one second sample, and the second sample comprises: a sample user characteristic, a second sample user label.
A user tag set generating module 207, configured to store the online user and the user tag of the online user into the user tag set according to a corresponding relationship.
In another embodiment of the present disclosure, the target merchant in the candidate advertisement set obtaining module 202 is predetermined by the target merchant determining module 208 as follows:
and the target merchant determining module 208 is configured to determine, according to the bidding decision, a merchant with a successful bidding as a target merchant.
In another embodiment of the present disclosure, the candidate advertisement in the candidate advertisement set obtaining module 202 is generated offline through a recall template and a recall material corresponding to each pit site in the recall template, the recall template in the candidate advertisement set obtaining module 202 is obtained offline from a preset template data set, the recall material in the candidate advertisement set obtaining module 202 is obtained offline from a preset material data set, the candidate advertisement set obtaining module 202 in fig. 8 includes a first candidate advertisement set obtaining sub-module 2021:
the first candidate advertisement set obtaining sub-module 2021 is configured to obtain a candidate advertisement set of a target merchant from a preset advertisement database, where the candidate advertisement set of the target merchant is stored in the advertisement database offline.
In another embodiment of the present disclosure, the candidate advertisement in the candidate advertisement set obtaining module 202 is generated in real time through a recall template and recall materials corresponding to each pit site in the recall template, the recall template in the candidate advertisement set obtaining module 202 is obtained in real time from a preset template data set, and the candidate advertisement set obtaining module 202 in fig. 8 includes a second candidate advertisement set obtaining sub-module 2022:
the second candidate advertisement set obtaining sub-module 2022 is configured to add the candidate advertisement generated in real time to the candidate advertisement set of the target merchant.
In another embodiment of the present disclosure, referring to FIG. 10, a block diagram of a candidate advertisement set acquisition module for generating a recall template of the present disclosure is shown. As shown in fig. 10, the recall template in the candidate advertisement set obtaining module 202 is obtained from the preset template data set through the following service type obtaining sub-module 2023 and template recall sub-module 2024:
the business type obtaining sub-module 2023 is configured to obtain, according to a preset time period, a business type of each merchant in a preset merchant set, where the merchant set includes the target merchant.
The template recall sub-module 2024 is configured to, for each merchant, obtain a template corresponding to the service type from the template data set, and use the template as a recall template for the merchant.
In another embodiment of the present disclosure, reference is made to FIG. 11, which illustrates a block diagram of a candidate advertisement set acquisition module for generating recall material of the present disclosure. As shown in fig. 11, the recalled material in the candidate advertisement set acquiring module 202 is acquired from the preset material data set through the following material recall condition acquiring sub-module 2025 and the material recall sub-module 2026:
the material recall condition obtaining sub-module 2025 is configured to obtain, for each recall template of each merchant, a material recall condition of each pit in the recall template.
The material recall sub-module 2026 is configured to, for each pit site, obtain, from the material data set, a material that satisfies the material recall condition of the pit site, and use the material as the recall material corresponding to the pit site.
In another embodiment of the present disclosure, reference is made to fig. 12, which shows a structural diagram of a matching degree determination module of the present disclosure. As shown in fig. 12, the advertisement tag set of each candidate advertisement includes at least one advertisement tag, and the matching degree determining module 203 in fig. 8 includes a matching degree and weight predicting sub-module 2031 and a matching degree determining sub-module 2032:
a matching degree and weight predicting sub-module 2031, configured to predict, for each advertisement tag of each candidate advertisement, a matching degree between the user tag and the advertisement tag of the candidate advertisement and a weight of the advertisement tag by using a tag matching degree prediction model; the label matching degree prediction model is obtained by training a first sample, wherein the first sample comprises: the method comprises the steps of obtaining a first sample user label, a sample advertisement label, a sample matching degree and a sample weight of the sample advertisement label, wherein the sample matching degree is the matching degree of the first sample user label and the sample advertisement label.
The matching degree determining sub-module 2032 is configured to perform, for each candidate advertisement, a weighting operation according to the matching degree between the user tag and the advertisement tag of the candidate advertisement and the weight of the advertisement tag, so as to obtain the matching degree between the target user and the candidate advertisement.
In another embodiment of the present disclosure, referring to FIG. 13, a block diagram of an advertising module of the present disclosure is shown. As shown in fig. 13, the advertisement delivery module 204 in fig. 8 includes an advertisement quality prediction sub-module 2041, a comprehensive score determination sub-module 2042, and an advertisement delivery sub-module 2043:
the advertisement quality prediction sub-module 2041 is configured to, for each candidate advertisement in the candidate advertisement set of the target merchant, obtain a quality score of the candidate advertisement, where the quality score of the candidate advertisement is determined according to at least one of a quality score of a recall template of the candidate advertisement and a quality score of a recall material of the candidate advertisement.
A comprehensive score determining sub-module 2042, configured to determine, for each candidate advertisement in the candidate advertisement set of the target merchant, a comprehensive score of the candidate advertisement according to the quality score of the candidate advertisement and the matching degree between the target user and the candidate advertisement.
The advertisement delivery sub-module 2043 is configured to obtain the candidate advertisement with the highest comprehensive score from the candidate advertisement set of the target merchant, and deliver the candidate advertisement to the target user.
To sum up, the present disclosure provides an advertisement delivery device, the device includes: the user tag acquisition module is used for acquiring a user tag corresponding to a target user from the user tag set; a candidate advertisement set obtaining module, configured to obtain a candidate advertisement set of a target merchant, where the candidate advertisement set includes: the method comprises the steps that candidate advertisements and advertisement label sets of the candidate advertisements are generated through a recall template and recall materials corresponding to each pit site in the recall template, the recall template is obtained from a preset template data set, and the recall materials are obtained from a preset material data set; a matching degree determination module, configured to determine, for each candidate advertisement, a matching degree between the target user and the candidate advertisement according to the user tag and the advertisement tag set of the candidate advertisement; and the advertisement delivery module is used for acquiring the candidate advertisement with the highest matching degree with the target user from the candidate advertisement set of the target merchant so as to deliver the candidate advertisement to the target user. The method and the device can dynamically generate the candidate advertisement set according to a large number of recall templates and recall materials, and are favorable for improving the diversity of the candidate advertisement set.
The device embodiment of the present disclosure may refer to the details of the method embodiment, which are not described herein again.
The present disclosure also provides an electronic device, referring to fig. 14, which shows a structural diagram of the electronic device of the present disclosure. As shown in fig. 14, the electronic apparatus includes: a processor 301, a memory 302 and a computer program 3021 stored on and executable on the memory 302, the processor 301 implementing the advertisement delivery method of the foregoing embodiments when executing the programs.
The present disclosure also provides a readable storage medium, wherein when the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the advertisement delivery method of the foregoing embodiment.
For the device embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, refer to the partial description of the method embodiment.
The algorithms and displays presented herein are not inherently related to any particular computer, virtual machine, or other apparatus. Various general purpose systems may also be used with the teachings herein. The required structure for constructing such a system will be apparent from the description above. Moreover, this disclosure is not directed to any particular programming language. It is appreciated that a variety of programming languages may be used to implement the present disclosure as described herein, and any descriptions above of specific languages are provided for disclosure of enablement and best mode of the present disclosure.
In the description provided herein, numerous specific details are set forth. It can be appreciated, however, that the present disclosure may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.
Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the disclosure, various features of the disclosure are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. However, the disclosed method should not be interpreted as reflecting an intention that: that is, the claimed disclosure requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this disclosure.
Those skilled in the art will appreciate that the modules in the device in an embodiment may be adaptively changed and disposed in one or more devices different from the embodiment. The modules or units or components of the embodiments may be combined into one module or unit or component, and furthermore they may be divided into a plurality of sub-modules or sub-units or sub-components. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or elements of any method or apparatus so disclosed, may be combined in any combination, except combinations where at least some of such features and/or processes or elements are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.
Various component embodiments of the disclosure may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art will appreciate that a microprocessor or Digital Signal Processor (DSP) may be used in practice to implement some or all of the functions of some or all of the components in an advertising device according to the present disclosure. The present disclosure may also be embodied as an apparatus or device program for performing a portion or all of the methods described herein. Such programs implementing the present disclosure may be stored on a computer-readable medium or may be in the form of one or more signals. Such a signal may be downloaded from an internet website or provided on a carrier signal or in any other form.
It should be noted that the above-mentioned embodiments illustrate rather than limit the disclosure, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The disclosure may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The usage of the words first, second and third, etcetera do not indicate any ordering. These words may be interpreted as names.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
The above description is only exemplary of the present disclosure and should not be taken as limiting the disclosure, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present disclosure should be included in the scope of the present disclosure.
The above description is only for the specific embodiments of the present disclosure, but the scope of the present disclosure is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present disclosure, and all the changes or substitutions should be covered within the scope of the present disclosure. Therefore, the protection scope of the present disclosure shall be subject to the protection scope of the claims.
Claims (12)
1. An advertisement delivery method, the method comprising:
acquiring a user tag corresponding to a target user from a user tag set;
acquiring a candidate advertisement set of a target merchant, wherein the candidate advertisement set comprises: the method comprises the steps that candidate advertisements and advertisement label sets of the candidate advertisements are generated through a recall template and recall materials corresponding to each pit site in the recall template, the recall template is obtained from a preset template data set, and the recall materials are obtained from a preset material data set;
for each candidate advertisement, determining the matching degree of the target user and the candidate advertisement according to the user label and the advertisement label set of the candidate advertisement;
and obtaining the candidate advertisement with the highest matching degree with the target user from the candidate advertisement set of the target merchant so as to deliver the candidate advertisement to the target user.
2. The method of claim 1, wherein the user tag set is pre-generated by:
obtaining at least one online user and user characteristics of the online user, wherein the online user comprises an active user in a current time period, and the active user comprises the target user;
for each online user, inputting the user characteristics of the online user into a user label prediction model to obtain a user label of the online user; the user label prediction model is obtained by training at least one second sample, and the second sample comprises: a sample user characteristic, a second sample user label;
and storing the online user and the user tags of the online user into the user tag set according to the corresponding relation.
3. The method of claim 1, wherein the target merchant is predetermined by:
and determining the commercial tenant with successful bidding according to the bidding decision, and taking the commercial tenant as the target commercial tenant.
4. The method of claim 1, wherein the candidate advertisements are generated offline from recall templates obtained offline from a preset template data set and recall material obtained offline from a preset material data set corresponding to each pit in the recall template, and wherein the step of obtaining the set of candidate advertisements for the target merchant comprises:
acquiring a candidate advertisement set of a target merchant from a preset advertisement database, wherein the candidate advertisement set of the target merchant is stored in the advertisement database in an off-line manner.
5. The method of claim 1, wherein the candidate advertisements are generated in real-time from recall templates obtained in real-time from a preset template dataset and recall material obtained in real-time from a preset material dataset and corresponding to each pit in the recall templates, and wherein the step of obtaining the set of candidate advertisements for the target merchant comprises:
adding the candidate advertisement generated in real time to the candidate advertisement set of the target merchant.
6. The method of claim 1, wherein the recall template is obtained from a preset template dataset by:
acquiring the business type of each merchant in a preset merchant set according to a preset time period, wherein the merchant set comprises the target merchant;
and for each merchant, acquiring a template corresponding to the service type from the template data set to serve as a recall template of the merchant.
7. The method of claim 1, wherein the recalled material is obtained from a predetermined set of material data by:
acquiring a material recall condition of each pit position in the recall template aiming at each recall template of each merchant;
and for each pit position, acquiring a material meeting the material recall condition of the pit position from the material data set as the recall material corresponding to the pit position.
8. The method of claim 1, wherein the advertisement tag set of each candidate advertisement includes at least one advertisement tag, and wherein the step of determining, for each candidate advertisement, a matching degree between the target user and the candidate advertisement according to the user tag and the advertisement tag set of the candidate advertisement comprises:
for each advertisement label of each candidate advertisement, adopting a label matching degree prediction model to predict the matching degree of the user label and the advertisement label of the candidate advertisement and the weight of the advertisement label; the label matching degree prediction model is obtained by training a first sample, wherein the first sample comprises: the method comprises the steps of obtaining a first sample user label, a sample advertisement label, a sample matching degree and a sample weight of the sample advertisement label, wherein the sample matching degree is the matching degree of the first sample user label and the sample advertisement label;
and for each candidate advertisement, carrying out weighted operation according to the matching degree of the user label and the advertisement label of the candidate advertisement and the weight of the advertisement label to obtain the matching degree of the target user and the candidate advertisement.
9. The method according to claim 1, wherein the step of obtaining the candidate advertisement with the highest matching degree with the target user from the candidate advertisement set of the target merchant for delivery to the target user comprises:
for each candidate advertisement in the set of candidate advertisements for the target merchant, obtaining a quality score for the candidate advertisement, the quality score for the candidate advertisement being determined from at least one of a quality score for a recall template of the candidate advertisement and a quality score for recall material of the candidate advertisement;
for each of the candidate advertisements in the set of candidate advertisements for the target merchant, determining a composite score for the candidate advertisement based on the quality score for the candidate advertisement and the degree of match between the target user and the candidate advertisement;
and acquiring the candidate advertisement with the highest comprehensive score from the candidate advertisement set of the target merchant so as to deliver the candidate advertisement to the target user.
10. An advertising device, the device comprising:
the user tag acquisition module is used for acquiring a user tag corresponding to a target user from the user tag set;
a candidate advertisement set obtaining module, configured to obtain a candidate advertisement set of a target merchant, where the candidate advertisement set includes: the method comprises the steps that candidate advertisements and advertisement label sets of the candidate advertisements are generated through a recall template and recall materials corresponding to each pit site in the recall template, the recall template is obtained from a preset template data set, and the recall materials are obtained from a preset material data set;
a matching degree determination module, configured to determine, for each candidate advertisement, a matching degree between the target user and the candidate advertisement according to the user tag and the advertisement tag set of the candidate advertisement;
and the advertisement delivery module is used for acquiring the candidate advertisement with the highest matching degree with the target user from the candidate advertisement set of the target merchant so as to deliver the candidate advertisement to the target user.
11. An electronic device, comprising:
processor, memory and computer program stored on the memory and executable on the processor, characterized in that the processor, when executing the program, implements the advertisement delivery method according to any of claims 1-9.
12. A readable storage medium, wherein instructions in the storage medium, when executed by a processor of an electronic device, enable the electronic device to perform the method of advertisement delivery of any of method claims 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911341189.XA CN111210258A (en) | 2019-12-23 | 2019-12-23 | Advertisement putting method and device, electronic equipment and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911341189.XA CN111210258A (en) | 2019-12-23 | 2019-12-23 | Advertisement putting method and device, electronic equipment and readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111210258A true CN111210258A (en) | 2020-05-29 |
Family
ID=70786360
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911341189.XA Pending CN111210258A (en) | 2019-12-23 | 2019-12-23 | Advertisement putting method and device, electronic equipment and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111210258A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111626254A (en) * | 2020-06-02 | 2020-09-04 | 上海商汤智能科技有限公司 | Display animation triggering method and device |
| CN111833099A (en) * | 2020-06-24 | 2020-10-27 | 广州筷子信息科技有限公司 | Method and system for generating creative advertisement |
| CN111882362A (en) * | 2020-07-31 | 2020-11-03 | 苏州云开网络科技有限公司 | Artificial intelligence advertisement delivery system based on 5G communication network |
| CN112685516A (en) * | 2021-01-12 | 2021-04-20 | 上海微问家信息技术有限公司 | Multi-channel recall recommendation method and device, electronic equipment and medium |
| CN113139834A (en) * | 2021-04-29 | 2021-07-20 | 北京沃东天骏信息技术有限公司 | Information processing method, device, electronic equipment and storage medium |
| CN113570416A (en) * | 2021-07-30 | 2021-10-29 | 北京达佳互联信息技术有限公司 | Method and device for determining delivered content, electronic equipment and storage medium |
| CN114266592A (en) * | 2021-12-13 | 2022-04-01 | 北京蚂蜂窝网络科技有限公司 | Distribution method, distribution device, equipment and medium of advertisement materials |
| CN114630188A (en) * | 2022-02-18 | 2022-06-14 | 北京三快在线科技有限公司 | Short video advertisement generation method and device, electronic equipment and readable storage medium |
| CN117312658A (en) * | 2023-09-08 | 2023-12-29 | 广州风腾网络科技有限公司 | Popularization method and system based on big data analysis |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108256886A (en) * | 2016-12-29 | 2018-07-06 | 北京奇虎科技有限公司 | Advertisement placement method and device |
| CN108510309A (en) * | 2018-02-27 | 2018-09-07 | 阿里巴巴集团控股有限公司 | The method and device that advertisement is recalled |
| CN108776907A (en) * | 2018-05-31 | 2018-11-09 | 康键信息技术（深圳）有限公司 | Advertisement intelligent recommends method, server and storage medium |
| CN108960898A (en) * | 2018-06-11 | 2018-12-07 | 北京奇艺世纪科技有限公司 | A kind of advertisement placement method and device |
| CN109636479A (en) * | 2018-12-19 | 2019-04-16 | 武汉斗鱼鱼乐网络科技有限公司 | A kind of advertisement recommended method, device, electronic equipment and storage medium |
| CN109961308A (en) * | 2017-12-25 | 2019-07-02 | 北京京东尚科信息技术有限公司 | The method and apparatus of assessment tag data |
| CN110222894A (en) * | 2019-06-06 | 2019-09-10 | 阿里巴巴集团控股有限公司 | Advertisement placement method, device and equipment |
| CN110325986A (en) * | 2017-12-15 | 2019-10-11 | 腾讯科技（深圳）有限公司 | Article processing method, device, server and storage medium |
-
2019
- 2019-12-23 CN CN201911341189.XA patent/CN111210258A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108256886A (en) * | 2016-12-29 | 2018-07-06 | 北京奇虎科技有限公司 | Advertisement placement method and device |
| CN110325986A (en) * | 2017-12-15 | 2019-10-11 | 腾讯科技（深圳）有限公司 | Article processing method, device, server and storage medium |
| CN109961308A (en) * | 2017-12-25 | 2019-07-02 | 北京京东尚科信息技术有限公司 | The method and apparatus of assessment tag data |
| CN108510309A (en) * | 2018-02-27 | 2018-09-07 | 阿里巴巴集团控股有限公司 | The method and device that advertisement is recalled |
| CN108776907A (en) * | 2018-05-31 | 2018-11-09 | 康键信息技术（深圳）有限公司 | Advertisement intelligent recommends method, server and storage medium |
| CN108960898A (en) * | 2018-06-11 | 2018-12-07 | 北京奇艺世纪科技有限公司 | A kind of advertisement placement method and device |
| CN109636479A (en) * | 2018-12-19 | 2019-04-16 | 武汉斗鱼鱼乐网络科技有限公司 | A kind of advertisement recommended method, device, electronic equipment and storage medium |
| CN110222894A (en) * | 2019-06-06 | 2019-09-10 | 阿里巴巴集团控股有限公司 | Advertisement placement method, device and equipment |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111626254B (en) * | 2020-06-02 | 2024-04-16 | 上海商汤智能科技有限公司 | Method and device for triggering display animation |
| CN111626254A (en) * | 2020-06-02 | 2020-09-04 | 上海商汤智能科技有限公司 | Display animation triggering method and device |
| CN111833099A (en) * | 2020-06-24 | 2020-10-27 | 广州筷子信息科技有限公司 | Method and system for generating creative advertisement |
| CN111833099B (en) * | 2020-06-24 | 2021-08-31 | 广州筷子信息科技有限公司 | Method and system for generating creative advertisement |
| CN111882362A (en) * | 2020-07-31 | 2020-11-03 | 苏州云开网络科技有限公司 | Artificial intelligence advertisement delivery system based on 5G communication network |
| CN112685516A (en) * | 2021-01-12 | 2021-04-20 | 上海微问家信息技术有限公司 | Multi-channel recall recommendation method and device, electronic equipment and medium |
| CN113139834A (en) * | 2021-04-29 | 2021-07-20 | 北京沃东天骏信息技术有限公司 | Information processing method, device, electronic equipment and storage medium |
| CN113570416A (en) * | 2021-07-30 | 2021-10-29 | 北京达佳互联信息技术有限公司 | Method and device for determining delivered content, electronic equipment and storage medium |
| CN113570416B (en) * | 2021-07-30 | 2022-04-01 | 北京达佳互联信息技术有限公司 | Method and device for determining delivered content, electronic equipment and storage medium |
| CN114266592A (en) * | 2021-12-13 | 2022-04-01 | 北京蚂蜂窝网络科技有限公司 | Distribution method, distribution device, equipment and medium of advertisement materials |
| CN114630188A (en) * | 2022-02-18 | 2022-06-14 | 北京三快在线科技有限公司 | Short video advertisement generation method and device, electronic equipment and readable storage medium |
| CN114630188B (en) * | 2022-02-18 | 2024-03-15 | 北京三快在线科技有限公司 | Short video advertisement generation method and device, electronic equipment and readable storage medium |
| CN117312658A (en) * | 2023-09-08 | 2023-12-29 | 广州风腾网络科技有限公司 | Popularization method and system based on big data analysis |
| CN117312658B (en) * | 2023-09-08 | 2024-04-09 | 广州风腾网络科技有限公司 | Popularization method and system based on big data analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111210258A (en) | Advertisement putting method and device, electronic equipment and readable storage medium | |
| JP6427417B2 (en) | Multidimensional ad bidding | |
| US9460451B2 (en) | Quality scoring system for advertisements and content in an online system | |
| CN104919482B (en) | Dynamic content establishment of item | |
| CN106372249B (en) | A kind of clicking rate predictor method, device and electronic equipment | |
| US8788338B1 (en) | Unified marketplace for advertisements and content in an online system | |
| US8725559B1 (en) | Attribute based advertisement categorization | |
| CN108280749B (en) | Method and device for displaying service function entry | |
| EP2590128A1 (en) | Information processing device, information processing method, information processing program, and recording medium in which information processing program is recorded | |
| US20090248511A1 (en) | Advertisements on demand | |
| US11127032B2 (en) | Optimizing and predicting campaign attributes | |
| CN107590691B (en) | Information publishing method and device, storage medium and terminal | |
| CN111798280B (en) | Multimedia information recommendation method, device and equipment and storage medium | |
| CN111095330A (en) | Machine learning method and system for predicting online user interactions | |
| CN110766513A (en) | Information sorting method and device, electronic equipment and readable storage medium | |
| CN111259272B (en) | Search result ordering method and device | |
| CN111429214B (en) | Transaction data-based buyer and seller matching method and device | |
| WO2019052870A1 (en) | A method and system for intelligent adaptive bidding in an automated online exchange network | |
| CN113761348A (en) | Information recommendation method and device, electronic equipment and storage medium | |
| CN113393299A (en) | Recommendation model training method and device, electronic equipment and storage medium | |
| WO2022156589A1 (en) | Method and device for determining live broadcast click rate | |
| CN114862480A (en) | Advertisement putting orientation method and its device, equipment, medium and product | |
| CN110570271A (en) | information recommendation method and device, electronic equipment and readable storage medium | |
| JP2016062358A (en) | Extractor, and method and program for extraction | |
| CN116762070A (en) | System, method, computing platform, and storage medium for providing image recommendations |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200529 |