CN111652649B - Advertisement targeted delivery method, system, device and storage medium - Google Patents
Advertisement targeted delivery method, system, device and storage medium Download PDFInfo
- Publication number
- CN111652649B CN111652649B CN202010496625.7A CN202010496625A CN111652649B CN 111652649 B CN111652649 B CN 111652649B CN 202010496625 A CN202010496625 A CN 202010496625A CN 111652649 B CN111652649 B CN 111652649B
- Authority
- CN
- China
- Prior art keywords
- information
- orientation information
- directional
- product
- 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.)
- Active
Links
- 238000002716 delivery method Methods 0.000 title description 5
- 238000000034 method Methods 0.000 claims abstract description 31
- 230000006399 behavior Effects 0.000 claims description 14
- 230000008685 targeting Effects 0.000 claims description 13
- 230000006870 function Effects 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 4
- 230000035772 mutation Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 7
- 230000009471 action Effects 0.000 description 5
- 239000013598 vector Substances 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 3
- 230000002068 genetic effect Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000006467 substitution reaction Methods 0.000 description 3
- 238000003491 array Methods 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000003252 repetitive effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009432 framing Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
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
-
- 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
- G06Q30/0247—Calculate past, present or future revenues
-
- 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
Abstract
The invention discloses a method, a system, a device and a storage medium for targeted advertisement delivery, wherein the method comprises the following steps: acquiring product attribute information and advertisement putting limiting conditions; the product attribute information comprises a product type, a product target crowd attribute and a product target crowd pack; generating original orientation information according to the product attribute information and the advertisement putting limiting conditions; and carrying out advertisement delivery according to the original orientation information. The invention can enable the user to automatically complete the directional generation and advertisement delivery by only providing a small amount of information such as product attributes, delivery limiting conditions and the like; compared with the existing manual delivery, the method can reduce the cost and improve the delivery effect, and can be widely applied to the technical field of internet advertisements.
Description
Technical Field
The invention relates to the technical field of internet advertisements, in particular to an advertisement targeted delivery method, an advertisement targeted delivery system, an advertisement targeted delivery device and a storage medium.
Background
At present, convergence of information flow advertisement orientation is mainly carried out by means of manual adjustment by an operation planner, and the specific process comprises the following steps: the method comprises the steps of obtaining advertisement putting data, analyzing the advertisement putting data, adjusting the orientation according to putting data and product attributes and verifying online advertisements according to personal putting experience.
At present, an operation planner needs to spend a great deal of time to collect and analyze data, the work flow is complex, a great deal of repeated labor work exists, the adjustment of the orientation has a certain subjectivity, and the orientation convergence effect is different from person to person.
Disclosure of Invention
In view of the above, the embodiment of the invention provides an advertisement targeted delivery method, system, device and storage medium with low labor cost and good delivery effect.
The first aspect of the present invention provides for obtaining product attribute information and advertisement delivery constraints; the product attribute information comprises a product type, a product target crowd attribute and a product target crowd pack;
generating original orientation information according to the product attribute information and the advertisement putting limiting conditions;
and carrying out advertisement delivery according to the original orientation information.
In some embodiments, the method further comprises the steps of:
optimizing the original orientation information, and determining high-quality orientation information;
and carrying out factor exchange on the high-quality directional information and/or carrying out factor variation on the high-quality directional information, and then carrying out advertisement delivery based on the high-quality directional information.
In some embodiments, the generating original targeting information according to the product attribute information and the advertisement delivery constraint includes:
determining a directional label according to the advertisement putting limiting conditions;
determining a directional user group according to the product target crowd pack in the product attribute information;
determining a directed interest behavior tag word according to the product target crowd attribute in the product attribute information;
combining the oriented labels, the oriented user groups and the oriented interest behavior label words to generate original oriented information; the original orientation information comprises general projection orientation information, crowd pack orientation information, rocard orientation information and mixed orientation information.
In some embodiments, the generating original targeting information according to the product attribute information and the advertisement delivery constraint condition further comprises:
and when the original orientation information meets a first preset condition, calculating the fitness score of each tag factor in the original orientation information, selecting the tag factors with the fitness score larger than a first threshold as factor combinations, determining intersections of the factor combinations and the advertisement putting limiting conditions, and finally determining the original orientation information according to the intersections.
In some embodiments, the optimizing the original orientation information to determine the quality orientation information includes:
summarizing to obtain targeted delivery data according to the obtained advertisement delivery data;
calculating advertisement putting cost and directional fitness according to the directional putting data;
and determining high-quality orientation information according to the advertisement putting cost and the orientation adaptability.
In some embodiments, the factor exchanging the high quality orientation information comprises:
when the number of the high-quality orientations in the high-quality orientation information is more than or equal to 2, carrying out orientation grouping on the high-quality orientations in the high-quality orientation information;
and acquiring all the directional labels of each group of high-quality directional information obtained by grouping, and exchanging the directional factors in each group of high-quality directional information based on the exploration probability.
In some embodiments, the performing a factor variation on the quality orientation information comprises:
acquiring an orientation label list of the high-quality orientation information;
generating random numbers according to all the tags in the oriented tag list;
when the random number is smaller than the threshold value of the exploration probability and each label in the oriented label list meets a second preset condition, carrying out label variation on each label in the oriented label list;
the performing tag mutation on each tag in the oriented tag list specifically includes:
and calculating the fitness score of each label, selecting a label factor with the fitness score larger than a second threshold value as a factor combination, determining an intersection of the factor combination and the advertisement putting limiting condition, and finally determining high-quality orientation information according to the intersection.
A second aspect of the present invention provides an advertisement targeting system comprising:
the acquisition module is used for acquiring product attribute information and advertisement putting limiting conditions; the product attribute information comprises a product type, a product target crowd attribute and a product target crowd pack;
the generation module is used for generating original orientation information according to the product attribute information and the advertisement putting limiting conditions;
and the delivery module is used for delivering advertisements according to the original orientation information.
A third aspect of the present invention provides an advertisement targeting device, comprising a processor and a memory;
the memory is used for storing programs;
the processor is configured to perform the method according to the first aspect of the invention according to the program.
A fourth aspect of the present invention provides a storage medium storing a program for execution by a processor to perform the method of the first aspect of the present invention.
According to the embodiment of the invention, the original directional information is generated according to the product attribute information and the advertisement putting limiting conditions, and finally the advertisement putting is carried out according to the original directional information, so that a user can automatically complete the directional generation and the advertisement putting only by providing a small amount of information such as the product attribute, the putting limiting conditions and the like; compared with the existing manual throwing, the invention can reduce the cost and improve the throwing effect.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart illustrating the overall steps of an embodiment of the present invention.
Detailed Description
The invention is further explained and illustrated below with reference to the drawing and the specific embodiments of the present specification. The step numbers in the embodiments of the present invention are set for convenience of illustration, and the order of steps is not limited in any way, and the execution order of the steps in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.
Referring to fig. 1, the embodiment of the invention provides an advertisement directional delivery method with low labor cost and good delivery effect, which specifically comprises the following steps:
s1, acquiring product attribute information and advertisement putting limiting conditions; the product attribute information comprises a product type, a product target crowd attribute and a product target crowd pack;
specifically, in this embodiment, attribute information and delivery limiting conditions of delivered products are collected through an intelligent delivery system page.
The specific product attribute information comprises a product type, a product target crowd attribute and a product target crowd package, wherein the product target crowd attribute is a behavior interest tag seed word of a target crowd and is used for expanding a directional behavior interest attribute; the specific delivery limiting conditions comprise a target value of advertisement delivery Cost (CPA), and a plurality of items of targeted categories such as delivery region, gender, age, school, wedding status, working status, equipment type and the like.
S2, generating original orientation information according to the product attribute information and the advertisement putting limiting conditions;
step S2 of the present embodiment includes steps S21 to S25:
s21, determining a directional label according to the advertisement putting limiting conditions;
specifically, in this embodiment, according to the input constraint condition of the user, the orientation label is selected by framing, for example: the advertisement delivery user limits the sex (male) and age (18-50 years) of the delivery object, and the determined targeting label in the embodiment is the sex and age label.
S22, determining a directional user group according to the product target crowd pack in the product attribute information;
specifically, in this embodiment, whether the user group is included or excluded is defined according to whether the user provides the product target crowd pack and the type of the product target crowd pack, for example, when the service of advertising is online acquisition of credit cards, the user group may be selected as the targeted crowd pack by using the crowd who applied for credit cards, or the user group may be excluded by using the lost crowd pack.
S23, determining a directed interest behavior tag word according to the product target crowd attribute in the product attribute information;
specifically, according to the behavior interest tag seed words provided by the user, the embodiment uses an API tool provided by the media side to obtain recommended oriented interest behavior tag words as alternative tag words. Then, word2Vec model is utilized to calculate alternative tag Word vector Vec target Average word vector Vec with behavioral interest tag seed words root And calculating the similarity of the alternative tag word and the action interest tag seed word according to a cosine distance formula, and if the similarity of the alternative tag word and a certain action interest tag seed word is greater than a threshold value or the average similarity of the alternative tag word and all action interest tag seed words is greater than a set threshold value, selecting the alternative tag word as an available word and adding the available word into the oriented action interest tag.
The Word2Vec model is a general neural network model for generating Word vectors; if the word vectors of the n words are Vec respectively 1 ,Vec 2 …Vec n Then the average word vector isThe cosine distance formula for calculating the similarity between the candidate tag word and the behavior interest tag seed word is as follows:
s24, combining the oriented labels, the oriented user groups and the oriented interest behavior label words to generate original oriented information; the original orientation information comprises general projection orientation information, crowd pack orientation information, rocard orientation information and mixed orientation information.
According to the method, a general projection orientation, a crowd pack orientation, a rocard orientation and a mixed orientation are generated by combining according to the orientation label list framed in the steps S21-S23. The general projection orientation means only comprises the necessary orientation label framed by S21; the crowd-sourced orientation refers to crowd-sourced containing only the necessary orientation labels framed by S21 and S22; the rocard orientation means only comprises the necessary orientation labels and the action interest labels which are framed by S21 and S23; the mixed orientation means that all orientation tags defined in S21 to S23 are contained.
And S25, calculating the fitness score of each tag factor in the original orientation information when the original orientation information meets a first preset condition, selecting the tag factors with the fitness score larger than a first threshold as factor combinations, determining intersections of the factor combinations and the advertisement putting limiting conditions, and finally determining the original orientation information according to the intersections.
Specifically, the first preset condition in this embodiment refers to determining, for a certain tag, whether tag data satisfies a condition. The judgment criterion is whether the feeding amount of the past n days is larger than m.
If the tag data does not meet the conditions, the tag data is directly generated by using the release limiting information provided by the user, if the user limits only the release of the male crowd, the sex tag after the directed generation is a male.
If the tag data meets the conditions, tag data is used for tag generation, specifically, the fitness score of each tag factor is calculated according to the tag data, the tag factor with the fitness score larger than a first threshold value is selected as a factor combination of the tag, and the intersection of the factor combination and the putting limiting conditions is taken as the tag factor for directional generation.
Wherein the fitness score Reward is calculated according to a fitness formula: reward= (Norm (profit) +norm (inton um)). Times.0.5-tanh (cost) empty ) The method comprises the steps of carrying out a first treatment on the surface of the The normalization function Norm (x) is for the sequence x= (x) 1 ,x 2 ,…,x n ) The normalization process is carried out, the processing is carried out,wherein x is min Is the minimum value of the sequence, x max Is the maximum value of the sequence; empty consumption
S3, optimizing the original orientation information, and determining high-quality orientation information;
step S3 of the present embodiment includes steps S31 to S33:
s31, summarizing the obtained advertisement delivery data to obtain directional delivery data;
s32, calculating advertisement delivery cost and directional fitness according to the directional delivery data;
s33, determining high-quality orientation information according to the advertisement putting cost and the orientation fitness.
In the embodiment, the collected advertisement putting data is utilized to sum up the directional data, the directional data is utilized to calculate the directional fitness, and the high-quality orientation is selected according to the directional fitness;
in this embodiment, the step of utilizing the collected advertisement delivery data to sum up the directional data refers to classifying and combining the advertisement delivery data using the same directional advertisement data after collecting the advertisement delivery data, calculating the directional delivery data, and calculating the advertisement delivery cost CPA and the virtual profit value profit according to a formula.
Wherein,in the formula, cost refers to consumption of single orientation, intonum refers to feeding amount of single orientation, CPA target Refers to the CPA target value.
The calculation of the directional fitness by using the directional data means that the fitness score of each direction is calculated according to a fitness formula.
The selection of the high-quality orientation according to the orientation fitness refers to the selection of the orientation with the fitness score larger than a threshold value or the orientation virtual profit value larger than 0 as the high-quality orientation.
S4, performing factor exchange on the high-quality directional information;
specifically, step S4 of the present embodiment includes S41 to S42:
s41, when the number of the high-quality orientations in the high-quality orientation information is greater than or equal to 2, carrying out orientation grouping on the high-quality orientations in the high-quality orientation information;
s42, acquiring all the orientation labels of each group of high-quality orientation information obtained by grouping, and exchanging orientation factors in each group of high-quality orientation information based on the exploration probability.
In the embodiment, whether the number of high-quality orientations in the high-quality orientation information is smaller than 2 is firstly judged, and if yes, factor exchange cannot be carried out on the high-quality orientation information; otherwise, a factor exchange step may be performed.
For the high-quality orientation information with the high-quality orientation number being more than or equal to 2, the embodiment groups the high-quality orientations according to the group of 2 high-quality orientations; and obtaining all orientation labels of each group of high-quality orientations obtained by grouping, and exchanging factors in the orientations according to the exploration probability p for each orientation label. If the age label is aimed at, the age label of the direction 1 is 18-49 years old, the age label of the direction 2 is 25-55 years old, if the generated random number is smaller than the exploration probability p, the age labels of the 2 directions are exchanged, namely the age label of the direction 1 is 25-55 years old, and the age label of the direction 2 is 18-49 years old; if the orientation 2 does not contain the age tag, the new orientation 1 does not contain the age tag after exchange, and the new orientation 2 contains the age tag 18-49 years old.
S5, performing factor variation on the high-quality orientation information;
step S5 of the present embodiment includes S51-S53:
s51, acquiring an orientation label list of the high-quality orientation information;
s52, generating random numbers according to all the tags in the oriented tag list;
s53, when the random number is smaller than a threshold value of the exploration probability and each label in the oriented label list meets a second preset condition, carrying out label variation on each label in the oriented label list;
the second preset condition is to judge whether the feeding amount of each label in the past n days is larger than m.
The embodiment obtains a directed tag list, generates random numbers for each tag, and if the random numbers are smaller than the exploration probability p, performs the following steps;
for a certain tag, it is determined whether the tag data satisfies a condition. The judgment criterion is whether the feeding amount of the past n days is larger than m.
If the tag data meets the second preset condition, performing tag variation by using the tag data; if the label data does not meet the second preset condition, generating in the range of the throwing limiting condition.
The step of performing label variation by using label data is to calculate the fitness score of each label factor according to the label data, and select the label factor with the fitness score larger than a threshold value as a new factor combination of the label. Wherein the fitness score is calculated as described in step S2.
S6, carrying out advertisement delivery according to the original orientation information.
According to the embodiment, advertisement update can be performed according to the directional information generated in any one of the steps S2-S5, and online advertisement delivery can be performed.
In addition, the present embodiment further provides step S7:
acquiring media data such as advertisement delivery data and orientation label data through an API (application program interface) provided by a media delivery platform, wherein the acquired advertisement delivery data comprises, but is not limited to, exposure, click-through, consumption, feed amount and orientation corresponding to each advertisement; the obtained orientation label data includes, but is not limited to, exposure amount, click amount, consumption amount, and feed amount of orientation labels (age, sex, region, interest, and behavior).
Meanwhile, the invention can return the advertisement delivery data to execute the steps S2, S3 or S5 based on the data analysis and the genetic algorithm principle according to the advertisement delivery data, and refer to the arrow flow direction of the flow chart of FIG. 1, thereby realizing updating iteration of the directional information and finally realizing directional convergence.
Corresponding to the method of fig. 1, the embodiment of the invention also provides an advertisement targeting system, which comprises:
the acquisition module is used for acquiring product attribute information and advertisement putting limiting conditions; the product attribute information comprises a product type, a product target crowd attribute and a product target crowd pack;
the generation module is used for generating original orientation information according to the product attribute information and the advertisement putting limiting conditions;
and the delivery module is used for delivering advertisements according to the original orientation information.
The embodiment of the invention also provides an advertisement targeted delivery device, which comprises a processor and a memory;
the memory is used for storing programs;
the processor is configured to execute the method according to the present invention according to the program.
The embodiment of the invention also provides a storage medium which stores a program which is executed by a processor to complete the method of the invention
In summary, the invention is based on the genetic algorithm principle and the API (interface technology) provided by the delivery media (Tencent, head, hundred degrees, etc.), and can enable the user to automatically complete the directional generation and convergence by providing only a small amount of information such as product attributes, delivery limiting conditions, etc. Compared with the manual flow, the advertisement delivery technology provided by the invention has the following advantages:
reducing repetitive labor for users: according to the invention, only a small amount of product information is provided by a user in an initial stage, the system can finish directional generation and convergence, and a large amount of repetitive work such as user data collection, data analysis, directional adjustment and the like is reduced;
the directional convergence logic is objective: the invention updates and iterates the orientation based on the analysis of the advertisement putting data and the principle of the genetic algorithm to realize the orientation convergence, in the process, the orientation adjustment is carried out only according to the objective advertisement putting data, the convergence effect depends on the new putting effect data on the advertisement, the intervention of different manual experiences is avoided, and the stability is higher than that of manual adjustment;
directional convergence intelligence: according to the invention, learning is performed according to a new putting result on the advertisement, the history high-quality orientation is selected, the oriented high-quality factors are reserved, and the exploration is performed on the basis of the high-quality orientation, so that the problem of sinking into a local optimal solution is avoided. Meanwhile, the invention can add the orientation factor data collected in the advertisement putting process in the initial value selecting and mutation process to accelerate the orientation convergence speed.
In some alternative embodiments, the functions/acts noted in the block diagrams may occur out of the order noted in the operational illustrations. For example, two blocks shown in succession may in fact be executed substantially concurrently or the blocks may sometimes be executed in the reverse order, depending upon the functionality/acts involved. Furthermore, the embodiments presented and described in the flowcharts of the present invention are provided by way of example in order to provide a more thorough understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed, and in which sub-operations described as part of a larger operation are performed independently.
Furthermore, while the invention is described in the context of functional modules, it should be appreciated that, unless otherwise indicated, one or more of the described functions and/or features may be integrated in a single physical device and/or software module or one or more functions and/or features may be implemented in separate physical devices or software modules. It will also be appreciated that a detailed discussion of the actual implementation of each module is not necessary to an understanding of the present invention. Rather, the actual implementation of the various functional modules in the apparatus disclosed herein will be apparent to those skilled in the art from consideration of their attributes, functions and internal relationships. Accordingly, one of ordinary skill in the art can implement the invention as set forth in the claims without undue experimentation. It is also to be understood that the specific concepts disclosed are merely illustrative and are not intended to be limiting upon the scope of the invention, which is to be defined in the appended claims and their full scope of equivalents.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
Logic and/or steps represented in the flowcharts or otherwise described herein, e.g., a ordered listing of executable instructions for implementing logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.
More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer readable medium may even be paper or other suitable medium on which the program is printed, as the program may be electronically captured, via, for instance, optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.
It is to be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above-described embodiments, the various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, may be implemented using any one or combination of the following techniques, as is well known in the art: discrete logic circuits having logic gates for implementing logic functions on data signals, application specific integrated circuits having suitable combinational logic gates, programmable Gate Arrays (PGAs), field Programmable Gate Arrays (FPGAs), and the like.
In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
While embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that: many changes, modifications, substitutions and variations may be made to the embodiments without departing from the spirit and principles of the invention, the scope of which is defined by the claims and their equivalents.
While the preferred embodiment of the present invention has been described in detail, the present invention is not limited to the embodiments described above, and various equivalent modifications and substitutions can be made by those skilled in the art without departing from the spirit of the present invention, and these equivalent modifications and substitutions are intended to be included in the scope of the present invention as defined in the appended claims.
Claims (7)
1. An advertisement targeting method is characterized by comprising the following steps:
acquiring product attribute information and advertisement putting limiting conditions; the product attribute information comprises a product type, a product target crowd attribute and a product target crowd pack;
generating original orientation information according to the product attribute information and the advertisement putting limiting conditions;
according to the original orientation information, advertisement delivery is carried out;
the generating original targeting information according to the product attribute information and the advertisement putting limiting condition comprises the following steps:
determining a directional label according to the advertisement putting limiting conditions;
determining a directional user group according to the product target crowd pack in the product attribute information;
determining a directed interest behavior tag word according to the product target crowd attribute in the product attribute information;
combining the oriented labels, the oriented user groups and the oriented interest behavior label words to generate original oriented information; the original orientation information comprises general projection orientation information, crowd pack orientation information, rocard orientation information and mixed orientation information;
the generating original targeting information according to the product attribute information and the advertisement putting limiting condition further comprises:
calculating fitness scores of all the tag factors in the original orientation information when the original orientation information meets a first preset condition, selecting the tag factors with the fitness scores larger than a first threshold as factor combinations, determining intersections of the factor combinations and the advertisement putting limiting conditions, and finally determining the original orientation information according to the intersections;
the calculation formula of the fitness score is as follows:
Reward＝(Norm(profit)+Norm(intonum))×0.5-tann(cost empty )
in the method, in the process of the invention,
profit＝(CPA target -CPA)×intonum
wherein, reward is the fitness score, norm () is a normalization function, cost empty For empty consumption, profit is virtual profit value, intonum is single directional feed amount, tanh () is hyperbolic tangent function, CPA is advertisement delivery cost, CPA target For the target value of the advertising cost, cost is a single targeted consumption.
2. The method for targeted delivery of advertisements of claim 1, further comprising the steps of:
optimizing the original orientation information, and determining high-quality orientation information;
performing factor exchange on the high-quality directional information and/or performing factor variation on the high-quality directional information, and then performing advertisement delivery based on the high-quality directional information;
the optimizing the original orientation information to determine high-quality orientation information comprises the following steps:
summarizing to obtain targeted delivery data according to the obtained advertisement delivery data;
calculating advertisement putting cost and directional fitness according to the directional putting data;
determining high-quality orientation information according to the advertisement putting cost and the orientation fitness;
the calculating the directional fitness according to the directional delivery data refers to calculating the fitness score of each direction according to the calculation formula of the fitness score.
3. The method for targeted delivery of advertisements according to claim 2, wherein said factor exchanging said quality targeted information comprises:
when the number of the high-quality orientations in the high-quality orientation information is more than or equal to 2, carrying out orientation grouping on the high-quality orientations in the high-quality orientation information;
and acquiring all the directional labels of each group of high-quality directional information obtained by grouping, and exchanging the directional factors in each group of high-quality directional information based on the exploration probability.
4. The method for targeted delivery of advertisements according to claim 2, wherein said performing a factor variation on said quality targeted information comprises:
acquiring an orientation label list of the high-quality orientation information;
generating random numbers according to all the tags in the oriented tag list;
when the random number is smaller than the threshold value of the exploration probability and each label in the oriented label list meets a second preset condition, carrying out label variation on each label in the oriented label list;
the performing tag mutation on each tag in the oriented tag list specifically includes:
and calculating the fitness score of each label, selecting a label factor with the fitness score larger than a second threshold value as a factor combination, determining an intersection of the factor combination and the advertisement putting limiting condition, and finally determining high-quality orientation information according to the intersection.
5. An ad targeting system comprising:
the acquisition module is used for acquiring product attribute information and advertisement putting limiting conditions; the product attribute information comprises a product type, a product target crowd attribute and a product target crowd pack;
the generation module is used for generating original orientation information according to the product attribute information and the advertisement putting limiting conditions;
the generating original targeting information according to the product attribute information and the advertisement putting limiting condition comprises the following steps:
determining a directional label according to the advertisement putting limiting conditions;
determining a directional user group according to the product target crowd pack in the product attribute information;
determining a directed interest behavior tag word according to the product target crowd attribute in the product attribute information;
combining the oriented labels, the oriented user groups and the oriented interest behavior label words to generate original oriented information; the original orientation information comprises general projection orientation information, crowd pack orientation information, rocard orientation information and mixed orientation information;
the generating original targeting information according to the product attribute information and the advertisement putting limiting condition further comprises:
calculating fitness scores of all the tag factors in the original orientation information when the original orientation information meets a first preset condition, selecting the tag factors with the fitness scores larger than a first threshold as factor combinations, determining intersections of the factor combinations and the advertisement putting limiting conditions, and finally determining the original orientation information according to the intersections;
the calculation formula of the fitness score is as follows:
Reward＝(Norm(profit)+Norm(intonum))×0.5-tanh(cost empty )
in the method, in the process of the invention,
profit＝(CPA target -CPA)×intonum
wherein, reward is the fitness score, norm () is a normalization function, cost empty For empty consumptionProfile is a virtual profit value, intonum is a single directional feed amount, tanh () is a hyperbolic tangent function, CPA is an advertisement delivery cost, CPA target The cost is the consumption of single orientation for the target value of the advertisement putting cost;
and the delivery module is used for delivering advertisements according to the original orientation information.
6. An advertisement targeting device is characterized by comprising a processor and a memory;
the memory is used for storing programs;
the processor is configured to perform the method according to any one of claims 1-4 according to the program.
7. A storage medium storing a program for execution by a processor to perform the method of any one of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010496625.7A CN111652649B (en) | 2020-06-03 | 2020-06-03 | Advertisement targeted delivery method, system, device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010496625.7A CN111652649B (en) | 2020-06-03 | 2020-06-03 | Advertisement targeted delivery method, system, device and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111652649A CN111652649A (en) | 2020-09-11 |
| CN111652649B true CN111652649B (en) | 2024-01-30 |
Family
ID=72345019
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010496625.7A Active CN111652649B (en) | 2020-06-03 | 2020-06-03 | Advertisement targeted delivery method, system, device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111652649B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112381570A (en) * | 2020-11-12 | 2021-02-19 | 广州易尊网络科技股份有限公司 | Target user-oriented advertisement delivery method and device |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB9901666D0 (en) * | 1999-01-27 | 1999-03-17 | Ibm | Intelligent adverts |
| CN102033883A (en) * | 2009-09-29 | 2011-04-27 | 阿里巴巴集团控股有限公司 | Method, device and system for improving data transmission speed of website |
| CN108830660A (en) * | 2018-06-27 | 2018-11-16 | 杭州祁睿信息技术有限公司 | A kind of digital advertisement put-on method based on depth machine learning and Combinatorial Optimization |
| CN108846694A (en) * | 2018-06-06 | 2018-11-20 | 厦门集微科技有限公司 | A kind of elevator card put-on method and device, computer readable storage medium |
| CN109377253A (en) * | 2018-09-04 | 2019-02-22 | 中国平安人寿保险股份有限公司 | Advertisement placement method, device, terminal and storage medium |
| CN109829536A (en) * | 2019-02-20 | 2019-05-31 | 杭州职业技术学院 | A kind of Network Synchronization performance optimization method based on genetic algorithm |
| CN110533437A (en) * | 2018-05-23 | 2019-12-03 | 北京国双科技有限公司 | Budget allocation method and device is launched in advertisement |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070143186A1 (en) * | 2005-12-19 | 2007-06-21 | Jeff Apple | Systems, apparatuses, methods, and computer program products for optimizing allocation of an advertising budget that maximizes sales and/or profits and enabling advertisers to buy media online |
-
2020
- 2020-06-03 CN CN202010496625.7A patent/CN111652649B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB9901666D0 (en) * | 1999-01-27 | 1999-03-17 | Ibm | Intelligent adverts |
| CN102033883A (en) * | 2009-09-29 | 2011-04-27 | 阿里巴巴集团控股有限公司 | Method, device and system for improving data transmission speed of website |
| CN110533437A (en) * | 2018-05-23 | 2019-12-03 | 北京国双科技有限公司 | Budget allocation method and device is launched in advertisement |
| CN108846694A (en) * | 2018-06-06 | 2018-11-20 | 厦门集微科技有限公司 | A kind of elevator card put-on method and device, computer readable storage medium |
| CN108830660A (en) * | 2018-06-27 | 2018-11-16 | 杭州祁睿信息技术有限公司 | A kind of digital advertisement put-on method based on depth machine learning and Combinatorial Optimization |
| CN109377253A (en) * | 2018-09-04 | 2019-02-22 | 中国平安人寿保险股份有限公司 | Advertisement placement method, device, terminal and storage medium |
| CN109829536A (en) * | 2019-02-20 | 2019-05-31 | 杭州职业技术学院 | A kind of Network Synchronization performance optimization method based on genetic algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111652649A (en) | 2020-09-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7092315B2 (en) | Automated systems and methods for creating machine-generated ads | |
| Kao et al. | A Bayesian latent variable model with classification and regression tree approach for behavior and credit scoring | |
| CN107967575B (en) | Artificial intelligence platform system for artificial intelligence insurance consultation service | |
| CN103927615B (en) | Entity is associated with classification | |
| CN108416384A (en) | A kind of image tag mask method, system, equipment and readable storage medium storing program for executing | |
| US20110289025A1 (en) | Learning user intent from rule-based training data | |
| CN111401963B (en) | Method and device for training user behavior prediction model | |
| CN102812460A (en) | Crowd-sourcing and contextual reclassification of rated content | |
| US20110191170A1 (en) | Similarity function in online advertising bid optimization | |
| CN102222299A (en) | Inventory management | |
| CN107133315A (en) | A kind of smart media based on semantic analysis recommends method | |
| CN111080338A (en) | User data processing method and device, electronic equipment and storage medium | |
| WO2023273295A1 (en) | Enterprise esg index determination method based on clustering technology, and related product | |
| CN111967971A (en) | Bank client data processing method and device | |
| CN102880696A (en) | Advertisement placement | |
| CN110069699A (en) | Order models training method and device | |
| WO2017079824A1 (en) | Markov decision process-based decision support tool for financial planning, budgeting, and forecasting | |
| CN111612519B (en) | Method, device and storage medium for identifying potential customers of financial products | |
| CN114331543A (en) | Advertisement propagation method for large-scale crowd orientation and dynamic scene matching | |
| CN107194672B (en) | Review distribution method integrating academic expertise and social network | |
| CN111652649B (en) | Advertisement targeted delivery method, system, device and storage medium | |
| CN108665156A (en) | Evaluation method is selected based on markovian supply chain under block chain | |
| Horky et al. | Price dynamics of non-fungible tokens: The case of the digital arts market | |
| CN109242671A (en) | A kind of credit violation correction method and system based on multi-angle of view deficiency of data | |
| Yu | The role and influence of artificial intelligence on advertising industry |
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 |