CN111127106B - Advertisement putting control method and storage medium - Google Patents
Advertisement putting control method and storage medium Download PDFInfo
- Publication number
- CN111127106B CN111127106B CN201911368188.4A CN201911368188A CN111127106B CN 111127106 B CN111127106 B CN 111127106B CN 201911368188 A CN201911368188 A CN 201911368188A CN 111127106 B CN111127106 B CN 111127106B
- Authority
- CN
- China
- Prior art keywords
- target
- advertisement
- vertex
- control parameter
- advertisement putting
- 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.)
- Active
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
- G06Q30/0242—Determining effectiveness of advertisements
- G06Q30/0244—Optimization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Economics (AREA)
- General Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses an advertisement putting control method and a storage medium, wherein the method acquires advertisement putting demand data of target advertisements, and the advertisement putting demand data comprises target advertisement putting topics, target advertisement putting funds and at least one target advertisement putting region; according to the target advertisement putting funds and the first control parameters corresponding to the target advertisement putting regions obtained from all the target advertisement putting regions; obtaining a second control parameter and a third control parameter which are adapted to each target advertisement putting region according to the first control parameter; for each target advertisement putting region, determining a target user according to the second control parameter, the third control parameter and the target advertisement putting theme; and delivering the target advertisement to the target user. The invention realizes the directional delivery of advertisements.
Description
Technical Field
The present disclosure relates to the field of advertisement data processing, and in particular, to an advertisement delivery control method and a storage medium.
Background
The accuracy of advertisement delivery has great influence on advertisement delivery effect, the click rate of new advertisements can be predicted based on the click rate expression of existing advertisements in the prior art, or the click rate prediction result of new advertisements can be obtained through analogy of the click rate expression of similar advertisements, so that a plurality of methods exist in the prior art for accurately controlling advertisement delivery, and the control accuracy is continuously increased. The embodiment of the invention combines the control research and development results of the advertisement putting of the embodiment and provides a brand-new advertisement putting control method, which aims to accurately control the advertisement putting of each advertisement putting region and achieve the most reasonable advertisement effect under the condition of limited resources.
Disclosure of Invention
In order to perform efficient and accurate advertisement delivery, the embodiment of the invention provides an advertisement delivery control method and a storage medium.
An advertisement placement control method, the method comprising:
acquiring advertisement putting demand data of a target advertisement, wherein the advertisement putting demand data comprises a target advertisement putting theme, target advertisement putting funds and at least one target advertisement putting region;
according to the target advertisement putting funds and first control parameters corresponding to the target advertisement putting regions obtained from all the target advertisement putting regions, the first control parameters are used for characterizing the advertisement putting funds configured for each target advertisement putting region;
obtaining a second control parameter and a third control parameter which are adapted to each target advertisement putting region according to the first control parameter, wherein the second control parameter is used for controlling the clustering granularity of the reference advertisements of each target advertisement putting region, and the third control parameter is used for controlling the clustering target selection granularity of the reference advertisements of each advertisement putting region;
for each target advertisement putting region, determining a target user according to the second control parameter, the third control parameter and the target advertisement putting theme;
and delivering the target advertisement to the target user.
Preferably, the second control parameter is positively correlated with the first control parameter; a third control parameter is positively correlated with the first control parameter.
Preferably, the determining the target user according to the second control parameter, the third control parameter and the target advertisement delivery theme includes:
extracting all reference advertisements of the target advertisement putting region, wherein each reference advertisement comprises a reference advertisement putting theme and a reference forward user attribute set;
constructing advertisement portraits for each reference advertisement putting theme based on a portrait extraction model;
clustering the reference advertisements according to the advertisement portraits of the reference advertisements and the second control parameters;
calculating the advertisement portrait pointed by the target advertisement putting theme based on a portrait extraction model, and determining a target cluster related to the advertisement portrait in the plurality of clusters according to a third control parameter;
the target user is determined based on a union of the reference forward user attribute sets of the respective reference advertisements in the target class cluster.
Preferably, the clustering of the reference advertisements according to the advertisement portraits of the reference advertisements and the second control parameters comprises:
calculating the difference degree of the advertisement portrait of each reference advertisement and the advertisement portrait of other reference advertisements according to the advertisement portrait of each reference advertisement;
constructing a difference map according to each difference and other differences, wherein each vertex in the difference map represents an advertisement portrait of a reference advertisement, each vertex and a related node are provided with a unique connecting line, the weight of the connecting line is the difference between the advertisement portrait of the reference advertisement represented by the vertex and the advertisement portrait of the reference advertisement represented by the related node, and the related node is other vertex adjacent to the vertex;
calculating a target difference according to the difference map and the second control parameter;
clustering the reference advertisements according to the target difference degree to obtain a plurality of clusters, so that the difference degree of advertisement figures of the reference advertisements in the same cluster is not larger than the target cluster difference degree, and the difference degree of advertisement figures of any two reference advertisements in different clusters is larger than the target cluster difference degree.
Preferably, the calculating the target difference according to the difference map and the second control parameter includes:
acquiring a first vertex set and a first connecting set according to the difference degree graph;
initializing a second vertex set and a second connecting line set, wherein the second vertex set has one element and only one element, and the second connecting line set is empty;
constructing a first attribute set, wherein elements in the first attribute set are used for recording first attributes of each vertex, and the first attributes represent minimum weights of connecting lines formed by all relevant elements in a second vertex set when the vertices are in a difference set of the first vertex set and the second vertex set, and the vertices corresponding to the relevant elements are vertices adjacent to the vertices of the difference graph;
constructing a second attribute set, wherein elements in the second attribute set are used for recording second attributes of each vertex, the second attributes represent another vertex which is different from the vertex in a connecting line with the minimum weight and hooked by each related element in the second vertex set when the vertex is in a difference set of the first vertex set and the second vertex set, and the vertex corresponding to the related element is a vertex adjacent to the vertex in the difference graph;
executing a preset operation, and updating the second vertex set, the second connection set, the first attribute set and the second attribute set until a preset requirement is met;
and arranging the elements in the first attribute set in descending order according to the numerical value, and determining the value of the N-1 element as the target difference degree, wherein N is a third control parameter.
Preferably, the executing the preset operation, updating the second vertex set, the second connection set, the first attribute set and the second attribute set until reaching a preset requirement, includes:
the executing the preset operation, updating the second vertex set, the second connection set, the first attribute set and the second attribute set until reaching the preset requirement, includes:
the following operations are performed until the second vertex set and the first vertex set have the same element number:
selecting a target connecting line with the smallest weight value from the first connecting line set, wherein a first vertex of the target connecting line is positioned in a second vertex set, and a second vertex of the target connecting line is positioned in a difference set between the first vertex set and the second vertex set;
adding the vertex of the difference set between the first vertex set and the second vertex set in the target connecting line into the second vertex set, and adding the target connecting line into the second connecting line set;
(3) Updating the first set of attributes and the second set of attributes.
Preferably, the determining the target user based on the union of the reference forward user attribute sets of the reference advertisements in the target class cluster includes:
the users in the reference forward user attribute set may be selected randomly or all selected as target users.
A computer storage medium may store a plurality of instructions that may be adapted to be loaded by a processor and to perform an advertising control method.
The embodiment of the invention provides an advertisement putting control method and a storage medium, which are used for obtaining experiences enough for advertisement putting control by carrying out full data analysis on reference advertisements, and determining target users according to the experiences, so that the front users of the past advertisement putting participate in the determination of the target users, and accurate advertisement directional putting is realized on the premise of limited advertisement putting resources.
Drawings
In order to more clearly illustrate the technical solutions and advantages of embodiments of the present application or of the prior art, the following description will briefly introduce the drawings that are required to be used in the embodiments or the prior art descriptions, it is obvious that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart of an advertisement delivery control method according to an embodiment of the present application;
FIG. 2 is a schematic flow chart of determining a target user according to the second control parameter, the third control parameter and the target advertisement delivery topic according to the embodiment of the present application;
FIG. 3 is a schematic flow chart of an image extraction model according to an embodiment of the present application;
FIG. 4 is a flowchart of a process for obtaining a joint vector sequence corresponding to each existing topic according to an embodiment of the present application;
FIG. 5 is a flowchart of clustering reference advertisements according to advertisement representations of respective reference advertisements and the second control parameter provided in an embodiment of the present application;
fig. 6 is a flowchart of calculating a target variability according to the variability map and the second control parameter according to the embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all, of the embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present application based on the embodiments herein.
It should be noted that the terms "first," "second," and the like in the description and claims of the present application and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that embodiments of the present application described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or server that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
In order to perform efficient and accurate advertisement delivery, an embodiment of the present invention provides an advertisement delivery control method, as shown in fig. 1, where the method includes:
s101, acquiring advertisement putting demand data of target advertisements, wherein the advertisement putting demand data comprises target advertisement putting topics, target advertisement putting funds and at least one target advertisement putting region.
S103, according to the target advertisement putting funds and first control parameters corresponding to the target advertisement putting regions obtained from all the target advertisement putting regions, the first control parameters are used for representing the advertisement putting funds configured for each target advertisement putting region.
S105, obtaining a second control parameter and a third control parameter which are adapted to each target advertisement putting region according to the first control parameter, wherein the second control parameter is used for controlling the clustering granularity of the reference advertisements of each target advertisement putting region, and the third control parameter is used for controlling the clustering target selection granularity of the reference advertisements of each advertisement putting region.
Specifically, the higher the first control parameter is, the more advertisement delivery funds in the target advertisement delivery region are represented, and the greater the delivery strength is. In order to achieve the advertisement objective to the maximum extent on the premise of limited funds in each target advertisement delivery region, the embodiment of the invention needs to calculate the second control parameter and the third control parameter according to the first control parameter, wherein the first control parameter is used for controlling the clustering granularity of the reference advertisements, the stronger the clustering granularity is, the higher the reference value of each cluster for the delivery of the target advertisements is, and therefore, the second control parameter is positively correlated with the first control parameter. The third control parameter is used for controlling the granularity of selecting the clustering targets, and each selected clustering target is directly used for determining a target user, so that the third control parameter is positively correlated with the first control parameter.
S107, for each target advertisement putting region, determining target users according to the second control parameters, the third control parameters and the target advertisement putting subjects.
Specifically, the determining the target user according to the second control parameter, the third control parameter and the target advertisement delivery theme, as shown in fig. 2, includes:
s1071, extracting all reference advertisements of the target advertisement putting region, wherein each reference advertisement comprises a reference advertisement putting theme and a reference forward user attribute set.
Specifically, each advertisement that has been placed in the targeted advertisement placement region may be used as a reference advertisement, and the reference forward user attribute set is a reference user set determined according to the user who has pushed the reference advertisement or who has pushed the reference advertisement and clicked on the user who has read the reference advertisement, and is determined according to the reference user combination. Specifically, the user attribute set is a union of attributes of each reference user in the reference user set. Of course, the user's attributes point to the user portraits or portions of the user portraits, the concept of which is well known in the art and will not be described in detail.
S1073, constructing advertisement images for each reference advertisement putting subject based on the image extraction model.
Specifically, the advertisement portraits may be characterized by an attribute set or a tag set of the subject matter of the reference advertisement, and the meaning of the portraits is known and known to those skilled in the art, and will not be described in detail herein.
Specifically, the portrait extraction model may be obtained by training the following method, as shown in fig. 3, including:
s1, acquiring a sample data set, wherein the sample data set comprises a plurality of existing topics and existing portraits corresponding to each existing topic.
S3, acquiring a joint vector sequence corresponding to each existing theme, and taking the joint vector sequence corresponding to each existing theme and the existing portrait of the existing theme as training elements to obtain a training data set.
Specifically, the obtaining the joint vector sequence corresponding to each existing topic, as shown in fig. 4, includes:
s31, performing word segmentation on the existing subject to obtain an initial word segmentation vector.
S33, inputting the initial word segmentation vector into a weight proportioning model to obtain a weight vector corresponding to each element in the initial word segmentation vector.
Specifically, the weight ratio model is used for determining a word element set corresponding to each element in the initial word segmentation vector according to the initial word segmentation vector, and further calculating a weight vector corresponding to each element. For example, the weight ratio model determines the word element x 3 The word elements of the associated word element set are x 1 ，x 2 ，x 4 And x 5 The weight vector set corresponding to the word element set comprises word element x 1 Corresponding weight vector a 3,1 The method comprises the steps of carrying out a first treatment on the surface of the Word element x 2 Corresponding weight vector a 3,2 The method comprises the steps of carrying out a first treatment on the surface of the Word element x 4 Corresponding weight directionQuantity a 3,4 The method comprises the steps of carrying out a first treatment on the surface of the Word element x 5 Corresponding weight vector a3,5 。
The formula for calculating the weight corresponding to each word element in the word element set is as follows:
the above equation (3) and equation (4) can be implemented by the softmax specification.
Specifically, the calculating the weight vector corresponding to each element includes: the weight vectors corresponding to the word elements, such as the attention vector g of the word vector x3, are obtained by carrying out weighted summation on each word element in the word element set and the weight vector corresponding to each word element 3 ＝x 1 *a 3,1 +x 2 *a 3,2 +x 4 *a 3,4 +x 5 *a 3,5 。
S35, obtaining a joint vector sequence according to the initial word segmentation vector and the weight vector corresponding to each element in the initial word segmentation vector.
Specifically, the initial word segmentation vector and the weight vector corresponding to each element in the initial word segmentation vector are spliced to obtain a joint vector sequence.
S5, constructing a neural network model, and predicting a predicted image pointed by a joint vector sequence corresponding to each existing theme based on the neural network model.
S7, obtaining a loss value based on the predicted image and the existing image with the corresponding relation, and carrying out back propagation optimization on parameters of the neural network based on the loss value until the neural network model reaches a preset convergence condition.
S1075, clustering the reference advertisements according to the advertisement portrait of each reference advertisement and the second control parameter.
Specifically, the clustering of the reference advertisements according to the advertisement portraits of the reference advertisements and the second control parameters, as shown in fig. 5, includes:
s10751, calculating the difference degree of the advertisement portrait of each reference advertisement and the advertisement portrait of other reference advertisements according to the advertisement portrait of each reference advertisement.
S10753, constructing a difference map according to each difference and other differences, wherein each vertex in the difference map represents an advertisement portrait of a reference advertisement, each vertex and a related node are respectively provided with a unique connecting line, the weight of each connecting line is the difference between the advertisement portrait of the reference advertisement represented by the vertex and the advertisement portrait of the reference advertisement represented by the related node, and the related node is other vertex adjacent to the vertex.
And S10755, calculating target difference degree according to the difference degree graph and the second control parameter.
Specifically, the calculating the target difference according to the difference map and the second control parameter, as shown in fig. 6, includes:
s10, acquiring a first vertex set and a first connecting set according to the difference degree graph.
Specifically, the first connection set records each connection in the variability map, and the value of each element in the first connection set identifies the weight, i.e. the variability, of the connection. The first vertex set comprises various vertices in the difference graph. Specifically, each vertex and each link may have its corresponding number.
S30, initializing a second vertex set and a second connecting line set, wherein the second vertex set has one element and only one element, and the second connecting line set is empty.
Specifically, the element may be any vertex in the first vertex set.
S50, constructing a first attribute set, wherein elements in the first attribute set are used for recording first attributes of all vertexes, and the first attributes represent minimum weights of connecting lines formed by all relevant elements in a second vertex set when the vertexes are in a difference set of the first vertex set and the second vertex set, and the vertexes corresponding to the relevant elements are vertexes adjacent to the vertexes of the difference map.
S70, constructing a second attribute set, wherein elements in the second attribute set are used for recording second attributes of all vertexes, the second attributes represent the other vertex which is different from the vertex in a connecting line with the minimum weight and hooked by each related element in the second vertex set when the vertex is in a difference set of the first vertex set and the second vertex set, and the vertex corresponding to the related element is the vertex adjacent to the vertex in the difference graph.
S90, executing preset operation, and updating the second vertex set, the second connection set, the first attribute set and the second attribute set until the preset requirement is met.
Specifically, the executing the preset operation, updating the second vertex set, the second connection set, the first attribute set and the second attribute set until reaching the preset requirement, includes:
the following operations are performed until the second vertex set and the first vertex set have the same element number:
(1) Selecting a target connecting line with the smallest weight value from the first connecting line set, wherein a first vertex of the target connecting line is positioned in a second vertex set, and a second vertex of the target connecting line is positioned in a difference set between the first vertex set and the second vertex set;
(2) Adding the vertex of the difference set between the first vertex set and the second vertex set in the target connecting line into the second vertex set, and adding the target connecting line into the second connecting line set;
(3) Updating the first set of attributes and the second set of attributes.
S110, arranging the elements in the first attribute set in descending order according to the numerical value, and determining the value of the N-1 element as the target degree of difference, wherein N is a third control parameter.
S10757, clustering the reference advertisements according to the target difference degree to obtain a plurality of clusters, so that the difference degree of advertisement figures of the reference advertisements in the same cluster is not larger than the target cluster difference degree, and the difference degree of advertisement figures of any two reference advertisements in different clusters is larger than the target cluster difference degree.
S1077, calculating the advertisement portrait pointed by the target advertisement putting theme based on a portrait extraction model, and determining a target cluster related to the advertisement portrait in the plurality of clusters according to a third control parameter.
Specifically, the distance between the advertisement portrait pointed by the target advertisement putting theme and the clustering center of each class cluster can be calculated, and N class clusters closest to the distance are taken as target class clusters, wherein N is a third control parameter.
S1079, determining target users based on the union of the reference forward user attribute sets of each reference advertisement in the target class cluster.
Specifically, users in the reference forward user attribute set may be selected randomly or all selected as target users.
S109, delivering the target advertisement to the target user.
The embodiment of the invention provides an advertisement putting control method, which is used for obtaining experiences enough for advertisement putting control by fully analyzing data of reference advertisements, and determining target users according to the experiences, so that the front users of the past advertisement putting participate in the determination of the target users, and accurate advertisement directional putting is realized on the premise of limited advertisement putting resources.
The embodiment of the invention also provides a computer storage medium which can store a plurality of instructions. The instructions may be adapted to be loaded and executed by a processor to perform a method of advertisement delivery control according to an embodiment of the present invention, the method comprising:
acquiring advertisement putting demand data of a target advertisement, wherein the advertisement putting demand data comprises a target advertisement putting theme, target advertisement putting funds and at least one target advertisement putting region;
according to the target advertisement putting funds and first control parameters corresponding to the target advertisement putting regions obtained from all the target advertisement putting regions, the first control parameters are used for characterizing the advertisement putting funds configured for each target advertisement putting region;
obtaining a second control parameter and a third control parameter which are adapted to each target advertisement putting region according to the first control parameter, wherein the second control parameter is used for controlling the clustering granularity of the reference advertisements of each target advertisement putting region, and the third control parameter is used for controlling the clustering target selection granularity of the reference advertisements of each advertisement putting region;
for each target advertisement putting region, determining a target user according to the second control parameter, the third control parameter and the target advertisement putting theme;
and delivering the target advertisement to the target user.
Preferably, the second control parameter is positively correlated with the first control parameter; a third control parameter is positively correlated with the first control parameter.
Preferably, the determining the target user according to the second control parameter, the third control parameter and the target advertisement delivery theme includes:
extracting all reference advertisements of the target advertisement putting region, wherein each reference advertisement comprises a reference advertisement putting theme and a reference forward user attribute set;
constructing advertisement portraits for each reference advertisement putting theme based on a portrait extraction model;
clustering the reference advertisements according to the advertisement portraits of the reference advertisements and the second control parameters;
calculating the advertisement portrait pointed by the target advertisement putting theme based on a portrait extraction model, and determining a target cluster related to the advertisement portrait in the plurality of clusters according to a third control parameter;
the target user is determined based on a union of the reference forward user attribute sets of the respective reference advertisements in the target class cluster.
Preferably, the clustering of the reference advertisements according to the advertisement portraits of the reference advertisements and the second control parameters comprises:
calculating the difference degree of the advertisement portrait of each reference advertisement and the advertisement portrait of other reference advertisements according to the advertisement portrait of each reference advertisement;
constructing a difference map according to each difference and other differences, wherein each vertex in the difference map represents an advertisement portrait of a reference advertisement, each vertex and a related node are provided with a unique connecting line, the weight of the connecting line is the difference between the advertisement portrait of the reference advertisement represented by the vertex and the advertisement portrait of the reference advertisement represented by the related node, and the related node is other vertex adjacent to the vertex;
calculating a target difference according to the difference map and the second control parameter;
clustering the reference advertisements according to the target difference degree to obtain a plurality of clusters, so that the difference degree of advertisement figures of the reference advertisements in the same cluster is not larger than the target cluster difference degree, and the difference degree of advertisement figures of any two reference advertisements in different clusters is larger than the target cluster difference degree.
Preferably, the calculating the target difference according to the difference map and the second control parameter includes:
acquiring a first vertex set and a first connecting set according to the difference degree graph;
initializing a second vertex set and a second connecting line set, wherein the second vertex set has one element and only one element, and the second connecting line set is empty;
constructing a first attribute set, wherein elements in the first attribute set are used for recording first attributes of each vertex, and the first attributes represent minimum weights of connecting lines formed by all relevant elements in a second vertex set when the vertices are in a difference set of the first vertex set and the second vertex set, and the vertices corresponding to the relevant elements are vertices adjacent to the vertices of the difference graph;
constructing a second attribute set, wherein elements in the second attribute set are used for recording second attributes of each vertex, the second attributes represent another vertex which is different from the vertex in a connecting line with the minimum weight and hooked by each related element in the second vertex set when the vertex is in a difference set of the first vertex set and the second vertex set, and the vertex corresponding to the related element is a vertex adjacent to the vertex in the difference graph;
executing a preset operation, and updating the second vertex set, the second connection set, the first attribute set and the second attribute set until a preset requirement is met;
and arranging the elements in the first attribute set in descending order according to the numerical value, and determining the value of the N-1 element as the target difference degree, wherein N is a third control parameter.
Preferably, the executing the preset operation, updating the second vertex set, the second connection set, the first attribute set and the second attribute set until reaching a preset requirement, includes:
the executing the preset operation, updating the second vertex set, the second connection set, the first attribute set and the second attribute set until reaching the preset requirement, includes:
the following operations are performed until the second vertex set and the first vertex set have the same element number:
selecting a target connecting line with the smallest weight value from the first connecting line set, wherein a first vertex of the target connecting line is positioned in a second vertex set, and a second vertex of the target connecting line is positioned in a difference set between the first vertex set and the second vertex set;
adding the vertex of the difference set between the first vertex set and the second vertex set in the target connecting line into the second vertex set, and adding the target connecting line into the second connecting line set;
(3) Updating the first set of attributes and the second set of attributes.
Preferably, the determining the target user based on the union of the reference forward user attribute sets of the reference advertisements in the target class cluster includes:
the users in the reference forward user attribute set may be selected randomly or all selected as target users.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. And the foregoing description has been directed to specific embodiments of this specification. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for the device and server embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, and references to the parts of the description of the method embodiments are only required.
It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program for instructing relevant hardware, where the program may be stored in a computer readable storage medium, and the storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.
The foregoing description of the preferred embodiments of the invention is not intended to limit the invention to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and scope of the invention are intended to be included within the scope of the invention.
Claims (3)
1. An advertisement delivery control method, characterized in that the method comprises the following steps:
acquiring advertisement putting demand data of a target advertisement, wherein the advertisement putting demand data comprises a target advertisement putting theme, target advertisement putting funds and at least one target advertisement putting region;
according to the target advertisement putting funds and first control parameters corresponding to the target advertisement putting regions obtained from all the target advertisement putting regions, the first control parameters are used for characterizing the advertisement putting funds configured for each target advertisement putting region;
obtaining a second control parameter and a third control parameter which are adapted to each target advertisement putting region according to the first control parameter, wherein the second control parameter is used for controlling the clustering granularity of the reference advertisements of each target advertisement putting region, and the third control parameter is used for controlling the clustering target selection granularity of the reference advertisements of each advertisement putting region; the second control parameter is positively correlated with the first control parameter; a third control parameter is positively correlated with the first control parameter;
for each target advertisement putting region, determining a target user according to the second control parameter, the third control parameter and the target advertisement putting theme; the determining the target user according to the second control parameter, the third control parameter and the target advertisement putting theme comprises the following steps: extracting all reference advertisements of the target advertisement putting region, wherein each reference advertisement comprises a reference advertisement putting theme and a reference forward user attribute set; constructing advertisement portraits for each reference advertisement putting theme based on a portrait extraction model; clustering the reference advertisements according to the advertisement portraits of the reference advertisements and the second control parameters; calculating the advertisement portrait pointed by the target advertisement putting theme based on a portrait extraction model, and determining a target cluster related to the advertisement portrait in a plurality of clusters according to a third control parameter; determining a target user based on the union of the reference forward user attribute sets of each reference advertisement in the target class cluster;
delivering the target advertisement to the target user;
wherein the clustering of the reference advertisements according to the advertisement portraits of the reference advertisements and the second control parameters comprises:
calculating the difference degree of the advertisement portrait of each reference advertisement and the advertisement portrait of other reference advertisements according to the advertisement portrait of each reference advertisement;
constructing a difference map according to each difference and other differences, wherein each vertex in the difference map represents an advertisement representation of a reference advertisement, each vertex and a related node are provided with a unique connecting line, the weight of the connecting line is the difference between the advertisement representation of the reference advertisement represented by the vertex and the advertisement representation of the reference advertisement represented by the related node, and the related node is other vertex adjacent to the vertex;
calculating a target difference according to the difference map and the second control parameter; clustering the reference advertisements according to the target difference degree to obtain a plurality of clusters, so that the difference degree of advertisement figures of the reference advertisements in the same cluster is not larger than the target cluster difference degree, and the difference degree of advertisement figures of any two reference advertisements in different clusters is larger than the target cluster difference degree;
wherein the calculating the target difference according to the difference map and the second control parameter includes:
acquiring a first vertex set and a first connecting set according to the difference degree graph;
initializing a second vertex set and a second connecting line set, wherein the second vertex set has one element and only one element, and the second connecting line set is empty;
constructing a first attribute set, wherein elements in the first attribute set are used for recording first attributes of each vertex, and the first attributes represent minimum weights of connecting lines formed by all relevant elements in a second vertex set when the vertices are in a difference set of the first vertex set and the second vertex set, and the vertices corresponding to the relevant elements are vertices adjacent to the vertices of the difference graph;
constructing a second attribute set, wherein elements in the second attribute set are used for recording second attributes of each vertex, the second attributes represent another vertex which is different from the vertex in a connecting line with the minimum weight and hooked by each related element in the second vertex set when the vertex is in a difference set of the first vertex set and the second vertex set, and the vertex corresponding to the related element is a vertex adjacent to the vertex in the difference graph;
executing a preset operation, and updating the second vertex set, the second connection set, the first attribute set and the second attribute set until a preset requirement is met;
arranging the elements in the first attribute set in descending order according to the numerical value, and determining the value of the N-1 element as the target difference degree, wherein N is a third control parameter;
the executing the preset operation, updating the second vertex set, the second connection set, the first attribute set and the second attribute set until reaching the preset requirement, includes:
the executing the preset operation, updating the second vertex set, the second connection set, the first attribute set and the second attribute set until reaching the preset requirement, includes:
the following operations are performed until the second vertex set and the first vertex set have the same element number:
selecting a target connecting line with the smallest weight value from the first connecting line set, wherein a first vertex of the target connecting line is positioned in a second vertex set, and a second vertex of the target connecting line is positioned in a difference set between the first vertex set and the second vertex set;
adding the vertex of the difference set between the first vertex set and the second vertex set in the target connecting line into the second vertex set, and adding the target connecting line into the second connecting line set;
updating the first set of attributes and the second set of attributes.
2. The method according to claim 1, characterized in that:
the determining the target user based on the union of the reference forward user attribute sets of each reference advertisement in the target class cluster comprises the following steps: the users in the reference forward user attribute set are selected randomly or all selected as target users.
3. A computer storage medium having stored thereon a plurality of instructions adapted to be loaded by a processor and to carry out an advertising control method according to claim 1 or 2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911368188.4A CN111127106B (en) | 2019-12-26 | 2019-12-26 | Advertisement putting control method and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911368188.4A CN111127106B (en) | 2019-12-26 | 2019-12-26 | Advertisement putting control method and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111127106A CN111127106A (en) | 2020-05-08 |
| CN111127106B true CN111127106B (en) | 2023-05-26 |
Family
ID=70503212
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911368188.4A Active CN111127106B (en) | 2019-12-26 | 2019-12-26 | Advertisement putting control method and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111127106B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105869021A (en) * | 2016-04-05 | 2016-08-17 | 乐视控股（北京）有限公司 | Internet advertisement putting method and apparatus |
| CN106886918A (en) * | 2017-02-06 | 2017-06-23 | 中国联合网络通信集团有限公司 | A kind of determination method of targeted customer, apparatus and system |
| CN109242575A (en) * | 2018-09-27 | 2019-01-18 | 武汉聚鸿传媒文化有限公司 | A kind of advertisement delivery system based on block chain |
| CN110210899A (en) * | 2019-05-23 | 2019-09-06 | 中国银行股份有限公司 | Advertisement sending method, device and equipment based on advertisement similitude |
| CN110210898A (en) * | 2019-05-23 | 2019-09-06 | 中国银行股份有限公司 | Advertisement sending method, device and equipment based on user behavior |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5895058B2 (en) * | 2012-08-02 | 2016-03-30 | グリー株式会社 | Target advertisement distribution system and target advertisement distribution method in the same system |
| US10085049B2 (en) * | 2016-07-09 | 2018-09-25 | N. Dilip Venkatraman | Method and system for serving advertisements during streaming of dynamic, adaptive and non-sequentially assembled video |
-
2019
- 2019-12-26 CN CN201911368188.4A patent/CN111127106B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105869021A (en) * | 2016-04-05 | 2016-08-17 | 乐视控股（北京）有限公司 | Internet advertisement putting method and apparatus |
| CN106886918A (en) * | 2017-02-06 | 2017-06-23 | 中国联合网络通信集团有限公司 | A kind of determination method of targeted customer, apparatus and system |
| CN109242575A (en) * | 2018-09-27 | 2019-01-18 | 武汉聚鸿传媒文化有限公司 | A kind of advertisement delivery system based on block chain |
| CN110210899A (en) * | 2019-05-23 | 2019-09-06 | 中国银行股份有限公司 | Advertisement sending method, device and equipment based on advertisement similitude |
| CN110210898A (en) * | 2019-05-23 | 2019-09-06 | 中国银行股份有限公司 | Advertisement sending method, device and equipment based on user behavior |
Non-Patent Citations (2)
| Title |
|---|
| 张清.面向精准广告投放的数据分析与可视化系统设计与实现.《信息科技》.2016,(第02期),全文. * |
| 曹阳等.基于聚类分析的互联网广告投放研究.《工程科技Ⅱ辑》.2016,全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111127106A (en) | 2020-05-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104778173B (en) | Target user determination method, device and equipment | |
| CN103502899B (en) | Dynamic prediction Modeling Platform | |
| KR20180091043A (en) | Method and apparatus for obtaining user portraits | |
| CN109583468A (en) | Training sample acquisition methods, sample predictions method and corresponding intrument | |
| TW201939400A (en) | Method and device for determining group of target users | |
| CN109816438B (en) | Information pushing method and device | |
| CN110971659A (en) | Recommendation message pushing method and device and storage medium | |
| CN107526810B (en) | Method and device for establishing click rate estimation model and display method and device | |
| CN108629608A (en) | User data processing method and processing device | |
| CN110321473B (en) | Multi-modal attention-based diversity preference information pushing method, system, medium and device | |
| CN110263821A (en) | Transaction feature generates the generation method and device of the training of model, transaction feature | |
| KR101725510B1 (en) | Method and apparatus for recommendation of social event based on users preference | |
| CN111768239A (en) | Property recommendation method, device, system, server and storage medium | |
| WO2023000491A1 (en) | Application recommendation method, apparatus and device, and computer-readable storage medium | |
| CN111861605A (en) | Business object recommendation method | |
| CN110263136B (en) | Method and device for pushing object to user based on reinforcement learning model | |
| CN112785390B (en) | Recommendation processing method, device, terminal equipment and storage medium | |
| CN111159241A (en) | Click conversion estimation method and device | |
| CN113034168A (en) | Content item delivery method and device, computer equipment and storage medium | |
| CN114861783A (en) | Recommendation model training method and device, electronic equipment and storage medium | |
| CN113761388A (en) | Recommendation method and device, electronic equipment and storage medium | |
| CN107729424B (en) | Data visualization method and equipment | |
| CN109583937A (en) | A kind of Products Show method and apparatus | |
| CN111127106B (en) | Advertisement putting control method and storage medium | |
| CN112288489A (en) | Advertisement reach rate estimation method and device, electronic equipment and 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |