CN112950277A - Intelligent playing schedule generation method based on digital media - Google Patents

Intelligent playing schedule generation method based on digital media Download PDF

Info

Publication number
CN112950277A
CN112950277A CN202110270761.9A CN202110270761A CN112950277A CN 112950277 A CN112950277 A CN 112950277A CN 202110270761 A CN202110270761 A CN 202110270761A CN 112950277 A CN112950277 A CN 112950277A
Authority
CN
China
Prior art keywords
advertisement
intention
search
digital media
list
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.)
Granted
Application number
CN202110270761.9A
Other languages
Chinese (zh)
Other versions
CN112950277B (en
Inventor
苏同
袁克雄
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huayang Lianzhong Digital Technology Shenzhen Co ltd
Original Assignee
Huayang Lianzhong Digital Technology Shenzhen Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huayang Lianzhong Digital Technology Shenzhen Co ltd filed Critical Huayang Lianzhong Digital Technology Shenzhen Co ltd
Priority to CN202110270761.9A priority Critical patent/CN112950277B/en
Publication of CN112950277A publication Critical patent/CN112950277A/en
Application granted granted Critical
Publication of CN112950277B publication Critical patent/CN112950277B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • G06Q30/0272—Period of advertisement exposure
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0251—Targeted advertisements

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Strategic Management (AREA)
  • Finance (AREA)
  • Game Theory and Decision Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a play scheduling intelligent generation method based on digital media, which comprises the following steps: and step S1, acquiring the number of commuters at the digital media position by using the real-time monitoring equipment, and carrying out periodic statistical analysis on the number of commuters so as to realize the segmented quantification of the advertisement playing time frame and generate a playing segmented list. The method and the system perform segmented quantization on the advertisement playing time frame of the digital media according to the advertisement exposure effect by using the number of commuters in the place where the digital media is placed, simultaneously establish an intention reference table for knowing the favorite condition of the user for watching the advertisement by using an advertisement intention model, and finally perform positive correlation matching arrangement on the user order and the advertisement exposure effect according to the intention watched by the user and the advertisement income balance degree, thereby improving the user experience of three parties including an advertisement watching user, an advertisement order user and a digital media operator and the economic benefits of the advertisement order user and the digital media operator.

Description

Intelligent playing schedule generation method based on digital media
Technical Field
The invention relates to the technical field of software, in particular to a playing schedule intelligent generation method based on digital media.
Background
The digital media playing system is a very common system, and large-scale LED screens in business circles, advertisement screens in elevators and the like are managed by the system. The most core in the digital media playing system is the playing of scheduling (advertisement), the conventional playing technical algorithm mainly comprises ordinary carousel, standard period playing and the like, the ordinary carousel is to play the advertisement according to the order and times according to the user, and the complementary advertisement is played after the ordinary advertisement is played.
In the prior art, the standard period playing is to set a playing period as a range, for example, a period of 4 minutes, play different advertisements in one period, and then perform a loop. However, these playing technologies all have a certain problem, and the common carousel equally distributes all the advertisements to each time slot for playing, i.e. neglecting to distinguish the advertisement grades to cause the difference of economic benefit for the digital media, and at the same time, does not distinguish the grades of the playing time slots, so that the advertisement exposure effect is good when the number of people watching in the golden time slot is large, and the advertisement exposure effect is poor when the number of people watching in the non-golden time slot is small, which leads to the unreasonable resource distribution of the digital media as a whole, and also leads to the reduction of the company's income.
Disclosure of Invention
The invention aims to provide a digital media-based play schedule intelligent generation method, which aims to solve the technical problems that in the prior art, the resource distribution of digital media is unreasonable, and the income of a company is reduced.
In order to solve the technical problems, the invention specifically provides the following technical scheme:
a play schedule intelligent generation method based on digital media comprises the following steps:
step S1, acquiring the number of commuters at the digital media position by using real-time monitoring equipment, and carrying out periodic statistical analysis on the number of commuters to realize the segmented quantization of the advertisement playing time frame and generate a playing segmented list;
step S2, constructing an advertisement intention model of the advertisement in the digital media by using the network big data, and mapping and matching advertisement order data to realize coding quantization to generate a play schedule list;
and step S3, generating the advertisement scheduling list by the playing segment list and the playing scheduling list through quantitative mapping.
As a preferred embodiment of the present invention, in step S1, the shooting range of the real-time monitoring device has a consistent extent with the visual range of the digital media, and the specific method for generating the playing segment list includes:
s101, frame cutting is realized on continuous monitoring images of real-time monitoring equipment in an advertisement playing time frame of a digital medium to obtain a plurality of advertisement playing time branch frames;
s102, sequentially and respectively counting the number of commuters in each advertisement playing time branch frame and synchronously quantizing the number of commuters into a heat sequence of the advertisement playing time branch frame, wherein the calculation formula of the heat sequence is as follows:
Figure BDA0002974282740000021
wherein H is the number of commuters in the advertisement playing time frame, and H is the number of commuters in the advertisement playing time frame;
s103, sequencing all the advertisement playing time branch frames based on the heat sequence and synchronously generating a playing segment list by taking each advertisement playing time branch frame as an independent list item.
As a preferred embodiment of the present invention, in step S2, the specific manner of constructing the advertisement intention model of the advertisement in the digital media by using the network big data is as follows:
step S201, capturing relevant data of a user multilayer shopping search intention by utilizing the Internet to realize quantitative extraction of the intention degree of a search level, wherein a calculation formula of the intention degree is as follows:
Figure BDA0002974282740000022
step S202, the quantitative extraction of the correlation degree of the search level, wherein the calculation formula of the correlation degree is as follows:
Figure BDA0002974282740000023
step S203, the quantitative extraction of the attention of the search level is carried out, and the calculation formula of the attention is as follows:
Figure BDA0002974282740000024
s204, combining the intention degree, the relevance degree and the attention degree by using a genetic algorithm as a target function to construct the advertisement intention model;
wherein, ai，ajFor the intention keyword of the ith, j search level, MaiIs all pairs of aiSet of intended keywords of search tier where the search tier is out-linked, L (a)j) Is ajNumber of out-links of search layer, TaiIs aiThe number of search layers, N is the total number of search layers.
As a preferred embodiment of the present invention, in step S201, a specific manner of capturing relevant data of a user' S multi-level shopping search intention by using the internet to generate a search hierarchy is as follows:
a user grabs related data of multiple layers of shopping search intentions and synchronously extracts intention keywords in each layer of shopping search intentions in a layering manner;
and storing the intention keywords in a layered mode to form a single search layer, and connecting the hierarchical depths of the multilayer shopping search intention of all the search layers in a linked list mode to form a search layer.
As a preferred aspect of the present invention, in step S204, the intention, the relevance, and the attention are modified by minimization to generate an objective function with multi-objective search characteristics, specifically:
the intention degree, the correlation degree and the attention degree are respectively [ y ] by utilizing the minimum correction1(aj)]、[y2(aj)]、[y3(aj)]；
Will [ y1(aj)]、[y2(aj)]、[y3(aj)]The joint generation objective function is:
Figure BDA0002974282740000031
as a preferred scheme of the present invention, the ad intention model generates an ad intention result as a Pareto optimal solution set, and the step S2 further includes implementing local optimization of the ad intention result on the Pareto solution set by using a hill climbing algorithm, where the specific manner is as follows:
step one, constructing a local optimization function f2(j)＝ω1*[y1(aj)]+ω2*[y2(aj)]+ω3*[y3(aj)]；
Step two, calculating all a in the Pareto solution setjLocal optimization function value f2(aj) Updating a by using a neighbor-based node identification methodjForm ajnewAnd recalculate ajnewLocal optimization function value f2(ajnew)；
Step three, comparing f2(aj) And f2(ajnew)：
If f2(aj)<f2(ajnew) Then a in Pareto solution set is retainedj；
If f2(aj)>f2(ajnew) Then use ajnewReplacing a in Pareto solution setj；
Step four, repeating the step one to the step three until all a in the Pareto solution set are finishedjThe update of (a) obtains a Pareto optimal solution set representing the ad intent.
As a preferred solution of the present invention, the step S2 further includes extracting original sequences of all the intended keywords in the Pareto optimal solution set, and generating a play-scheduling intention reference system.
As a priority scheme of the present invention, the specific manner of generating the play schedule list is as follows:
extracting order keywords of all advertisement order data, and obtaining a first intention weight of the advertisement order data by referring to the play scheduling intention reference system;
respectively counting single advertisement revenue of all advertisement order data to obtain a second revenue weight of the advertisement order data;
and constructing a scoring function based on the first intention weight and the second profit weight to obtain scheduling scores for the advertisement order data, and ranking all the advertisement order data according to the scheduling scores and storing the advertisement order data as a playing scheduling list in a single mode.
As a preferred embodiment of the present invention, the playing segment list and the playing schedule list have consistent ordering rules, and the step S3 generates the advertisement schedule list by quantizing and mapping the playing segment list and the playing schedule list in the following specific manner:
and sequentially filling all the scheduling list items in the sequential playing scheduling list into all the segment list items of the playing segment list to obtain the advertisement scheduling list.
The invention also comprises an advertisement management system, wherein the advertisement management system is integrated in the digital media internal processor by a playing schedule intelligent generation method based on digital media by using a software technology and plays advertisements according to the sequence of an advertisement playing list.
Compared with the prior art, the invention has the following beneficial effects:
the method and the system utilize the number of commuters of the place where the digital media are placed to carry out segmented quantization on the advertisement playing time frame of the digital media according to the advertisement exposure effect, simultaneously establish an intention reference table of an advertisement intention model for knowing the favorite condition of users for watching advertisements, carry out coding quantization on advertisement orders according to the watching intentions of the users and the benefits of the advertisements, and finally carry out positive correlation matching arrangement on the user orders and the advertisement exposure effect according to the watching intentions of the users and the balance degree of the advertising benefits, thereby improving the user experience of three parties of advertisement watching users, advertisement order users and digital media operators and the economic benefits of the advertisement order users and the digital media operators.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. It should be apparent that the drawings in the following description are merely exemplary, and that other embodiments can be derived from the drawings provided by those of ordinary skill in the art without inventive effort.
Fig. 1 is a flowchart of an intelligent generation method of a play schedule based on digital media according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, the present invention provides a method for intelligently generating a play schedule based on digital media, which comprises the following steps:
step S1, acquiring the number of commuters at the digital media position by using real-time monitoring equipment, and carrying out periodic statistical analysis on the number of commuters to realize the segmented quantization of the advertisement playing time frame and generate a playing segmented list;
in step S1, the shooting range of the real-time monitoring device and the visual range of the digital media have the same extent, that is, the commuters located in the visual range of the digital media can view the advertisement played by the digital media, so that the real-time monitoring device can monitor all commuters in the visual range of the digital media, the monitoring accuracy is improved, and the specific method for generating the playing segment list includes:
s101, frame cutting is realized on continuous monitoring images of real-time monitoring equipment in an advertisement playing time frame of a digital medium to obtain a plurality of advertisement playing time branch frames;
s102, sequentially and respectively counting the number of commuters in each advertisement playing time branch frame and synchronously quantizing the number of commuters into a heat sequence of the advertisement playing time branch frame, wherein the calculation formula of the heat sequence is as follows:
Figure BDA0002974282740000051
wherein H is the number of commuters in the advertisement playing time frame, and H is the number of commuters in the advertisement playing time frame;
s103, sequencing all the advertisement playing time branch frames based on the heat sequence and synchronously generating a playing segment list by taking each advertisement playing time branch frame as an independent list item.
The commuter number and the advertisement exposure effect are positively correlated, the commuter number is more, the advertisement exposure effect is better, the advertisement exposure effect is used as the embodiment of the heat degree of the advertisement playing time frame, the commuter number is used as the embodiment of the advertisement exposure effect, the commuter number is converted into the heat degree of the advertisement playing time frame to be embodied, the whole advertisement playing time frame can be quantized into the advertisement playing time frame branch frame with the heat degree sequence in a segmented mode, the heat degree sequence is used for ascending or descending, if the commuter number is arranged in a descending mode, the list items in the front portion of the playing segmented list are golden time periods, the list items at the tail portion are non-golden time periods, and finally grading of the advertisement playing time frame is achieved.
Step S2, constructing an advertisement intention model of the advertisement in the digital media by using the network big data, and mapping and matching advertisement order data to realize coding quantization to generate a play schedule list;
in step S2, the specific way of constructing the advertisement intention model of the advertisement in the digital media by using the network big data is as follows:
step S201, capturing relevant data of a user multilayer shopping search intention by utilizing the Internet to realize quantitative extraction of the intention degree of a search level, wherein a calculation formula of the intention degree is as follows:
Figure BDA0002974282740000061
in step S201, a specific way of capturing relevant data of a multi-level shopping search intention of a user by using the internet to generate a search level is as follows:
a user grabs related data of multiple layers of shopping search intentions and synchronously extracts intention keywords in each layer of shopping search intentions in a layering manner;
and storing the intention keywords in a layered mode to form a single search layer, and connecting the hierarchical depths of the multilayer shopping search intention of all the search layers in a linked list mode to form a search layer.
The potential intention of the user can be deeply mined by grabbing the multi-layer shopping search intention of the user, and the grabbing precision of the intention of the user is improved.
Step S202, the quantitative extraction of the correlation degree of the search level, wherein the calculation formula of the correlation degree is as follows:
Figure BDA0002974282740000062
step S203, the quantitative extraction of the attention of the search level is carried out, and the calculation formula of the attention is as follows:
Figure BDA0002974282740000063
s204, combining the intention degree, the relevance degree and the attention degree by using a genetic algorithm as a target function to construct the advertisement intention model;
wherein, ai，ajFor the intention keyword of the ith, j search level, MaiIs all pairs of aiSet of intended keywords of search tier where the search tier is out-linked, L (a)j) Is ajNumber of out-links of search layer, TaiIs aiThe number of search layers, N is the total number of search layers.
In the step S204, the intention degree, the correlation degree, and the attention degree are modified by minimization to generate an objective function with multi-objective search characteristics, specifically:
the intention degree, the correlation degree and the attention degree are respectively [ y ] by utilizing the minimum correction1(aj)]、[y2(aj)]、[y3(aj)]；
Will [ y1(aj)]、[y2(aj)]、[y3(aj)]The joint generation objective function is:
Figure BDA0002974282740000071
the method has the advantages that the optimal intention of a user can be really obtained by utilizing a multi-target intelligent search strategy, the search steps are simplified, the one-sidedness of searching according to a single element is avoided, and the search result is more comprehensive and reliable.
The ad intention model generates an ad intention result as a Pareto optimal solution set, and the step S2 further includes implementing local optimization of the ad intention result on the Pareto solution set by using a hill-climbing algorithm, where the specific manner is as follows:
step one, constructing a local optimization function f2(j)＝ω1*[y1(aj)]+ω2*[y2(aj)]+ω3*[y3(aj)]；
Step two, calculating all a in the Pareto solution setjLocal optimization function value f2(aj) Updating a by using a neighbor-based node identification methodjForm ajnewAnd recalculate ajnewLocal optimization function value f2(ajnew)；
Step three, comparing f2(aj) And f2(ajnew)：
If f2(aj)<f2(ajnew) Then a in Pareto solution set is retainedj；
If f2(aj)>f2(ajnew) Then use ajnewReplacing a in Pareto solution setj；
Step four, repeating the step one to the step three until all a in the Pareto solution set are finishedjThe update of (a) obtains a Pareto optimal solution set representing the ad intent.
By utilizing the local search strategy, global convergence and local optimum trapping in the multi-target intelligent search strategy can be avoided, so that the accuracy of the optimal solution is further optimized and improved.
The step S2 further includes extracting original sequences from all the intention keywords in the Pareto optimal solution set and sorting the extracted original sequences to generate an intention frame of play and schedule.
The specific way of generating the play schedule list is as follows:
extracting order keywords of all advertisement order data, and obtaining a first intention weight of the advertisement order data by referring to the play scheduling intention reference system;
respectively counting single advertisement revenue of all advertisement order data to obtain a second revenue weight of the advertisement order data;
and constructing a scoring function based on the first intention weight and the second profit weight to obtain scheduling scores for the advertisement order data, and ranking all the advertisement order data according to the scheduling scores and storing the advertisement order data as a playing scheduling list in a single mode.
In the actual use process, the first intention weight and the second profit weight can be set by self, so that the intention and the profit are both considered in the scheduling scoring of the advertisement order data, and the digital media operator can obtain the best profit while providing the advertisement which meets the watching intention for the user.
And step S3, generating the advertisement scheduling list by the playing segment list and the playing scheduling list through quantitative mapping.
The order rule of the playing segment list is consistent with that of the playing schedule list, and the step S3 is implemented by performing quantitative mapping on the playing segment list and the playing schedule list to generate the advertisement schedule list in the specific manner:
and sequentially filling all the scheduling list items in the sequential playing scheduling list into all the segment list items of the playing segment list to obtain the advertisement scheduling list.
The step can realize that the advertisement order with the best scheduling score is placed in the hottest advertisement time branch frame to be played, so that the advertisement order has the best exposure effect, advertisements which accord with the watching intention can be provided for users, and the digital media operator can obtain the best profit and simultaneously provide the best exposure profit for the advertisement order users, which is all the beauty of the users.
The invention also provides an advertisement management system, which integrates the playing schedule intelligent generation method based on the digital media into the digital media internal processor by using the software technology and plays the advertisements according to the sequence of the advertisement playing list.
The method and the system utilize the number of commuters of the place where the digital media are placed to carry out segmented quantization on the advertisement playing time frame of the digital media according to the advertisement exposure effect, simultaneously establish an intention reference table of an advertisement intention model for knowing the favorite condition of users for watching advertisements, carry out coding quantization on advertisement orders according to the watching intentions of the users and the benefits of the advertisements, and finally carry out positive correlation matching arrangement on the user orders and the advertisement exposure effect according to the watching intentions of the users and the balance degree of the advertising benefits, thereby improving the user experience of three parties of advertisement watching users, advertisement order users and digital media operators and the economic benefits of the advertisement order users and the digital media operators.
The above embodiments are only exemplary embodiments of the present application, and are not intended to limit the present application, and the protection scope of the present application is defined by the claims. Various modifications and equivalents may be made by those skilled in the art within the spirit and scope of the present application and such modifications and equivalents should also be considered to be within the scope of the present application.

Claims (10)

1. A play schedule intelligent generation method based on digital media is characterized by comprising the following steps:
step S1, acquiring the number of commuters at the digital media position by using real-time monitoring equipment, and carrying out periodic statistical analysis on the number of commuters to realize the segmented quantization of the advertisement playing time frame and generate a playing segmented list;
step S2, constructing an advertisement intention model of the advertisement in the digital media by using the network big data, and mapping and matching advertisement order data to realize coding quantization to generate a play schedule list;
and step S3, generating the advertisement scheduling list by the playing segment list and the playing scheduling list through quantitative mapping.
2. The method of claim 1, wherein the method comprises: in step S1, the shooting range of the real-time monitoring device and the visual range of the digital media have a consistent extent, and the specific method for generating the playing segment list includes:
s101, frame cutting is realized on continuous monitoring images of real-time monitoring equipment in an advertisement playing time frame of a digital medium to obtain a plurality of advertisement playing time branch frames;
s102, sequentially and respectively counting the number of commuters in each advertisement playing time branch frame and synchronously quantizing the number of commuters into a heat sequence of the advertisement playing time branch frame, wherein the calculation formula of the heat sequence is as follows:
Figure FDA0002974282730000011
wherein H is the number of commuters in the advertisement playing time frame, and H is the number of commuters in the advertisement playing time frame;
s103, sequencing all the advertisement playing time branch frames based on the heat sequence and synchronously generating a playing segment list by taking each advertisement playing time branch frame as an independent list item.
3. The method of claim 2, wherein the method comprises: in step S2, the specific way of constructing the advertisement intention model of the advertisement in the digital media by using the network big data is as follows:
step S201, capturing relevant data of a user multilayer shopping search intention by utilizing the Internet to realize quantitative extraction of the intention degree of a search level, wherein a calculation formula of the intention degree is as follows:
Figure FDA0002974282730000012
step S202, the quantitative extraction of the correlation degree of the search level, wherein the calculation formula of the correlation degree is as follows:
Figure FDA0002974282730000021
step S203, the quantitative extraction of the attention of the search level is carried out, and the calculation formula of the attention is as follows:
Figure FDA0002974282730000022
s204, combining the intention degree, the relevance degree and the attention degree by using a genetic algorithm as a target function to construct the advertisement intention model;
wherein, ai，ajFor the intention keyword of the ith, j search level, MaiIs all pairs of aiSet of intended keywords of search tier where the search tier is out-linked, L (a)j) Is ajNumber of out-links of search layer, TaiIs aiThe number of search layers, N is the total number of search layers.
4. The method of claim 3, wherein the method comprises: in step S201, a specific way of capturing relevant data of a multi-level shopping search intention of a user by using the internet to generate a search level is as follows:
a user grabs related data of multiple layers of shopping search intentions and synchronously extracts intention keywords in each layer of shopping search intentions in a layering manner;
and storing the intention keywords in a layered mode to form a single search layer, and connecting the hierarchical depths of the multilayer shopping search intention of all the search layers in a linked list mode to form a search layer.
5. The method of claim 4, wherein the method comprises: in the step S204, the intention degree, the correlation degree, and the attention degree are modified by minimization to generate an objective function with multi-objective search characteristics, specifically:
the intention degree, the correlation degree and the attention degree are respectively [ y ] by utilizing the minimum correction1(aj)]、[y2(aj)]、[y3(aj)]；
Will [ y1(aj)]、[y2(aj)]、[y3(aj)]The joint generation objective function is:
Figure FDA0002974282730000023
6. the method of claim 5, wherein the method comprises: the ad intention model generates an ad intention result as a Pareto optimal solution set, and the step S2 further includes implementing local optimization of the ad intention result on the Pareto solution set by using a hill-climbing algorithm, where the specific manner is as follows:
step one, constructing a local optimization function f2(j)＝ω1*[y1(aj)]+ω2*[y2(aj)]+ω3*[y3(aj)]；
Step two, calculating all a in the Pareto solution setjLocal optimization function value f2(aj) Updating a by using a neighbor-based node identification methodjForm ajnewAnd recalculate ajnewLocal optimization function value of
Figure FDA0002974282730000031
Step three, comparing f2(aj) And
Figure FDA0002974282730000032
if it is
Figure FDA0002974282730000033
Then a in the Pareto solution set is retainedj；
If it is
Figure FDA0002974282730000034
Then use ajnewReplacing a in Pareto solution setj；
Step four, repeating the step one to the step three until all a in the Pareto solution set are finishedjThe update of (a) obtains a Pareto optimal solution set representing the ad intent.
7. The method of claim 6, wherein the method comprises: the step S2 further includes extracting original sequences from all the intention keywords in the Pareto optimal solution set and sorting the extracted original sequences to generate an intention frame of play and schedule.
8. The method of claim 7, wherein the method comprises: the specific way of generating the play schedule list is as follows:
extracting order keywords of all advertisement order data, and obtaining a first intention weight of the advertisement order data by referring to the play scheduling intention reference system;
respectively counting single advertisement revenue of all advertisement order data to obtain a second revenue weight of the advertisement order data;
and constructing a scoring function based on the first intention weight and the second profit weight to obtain scheduling scores for the advertisement order data, and ranking all the advertisement order data according to the scheduling scores and storing the advertisement order data as a playing scheduling list in a single mode.
9. The method according to claim 8, wherein the playlist segments are consistent with the ordering rules of the playlist, and in step S3, the advertisement schedule list is generated by quantitatively mapping the playlist segments and the playlist schedule by:
and sequentially filling all the scheduling list items in the sequential playing scheduling list into all the segment list items of the playing segment list to obtain the advertisement scheduling list.
10. The method as claimed in claim 9, further comprising an advertisement management system, wherein the advertisement management system is integrated in the digital media internal processor by the digital media-based play schedule intelligent generation method using software technology and plays the advertisement according to the advertisement play list sequence.
CN202110270761.9A 2021-03-12 2021-03-12 Intelligent playing schedule generation method based on digital media Active CN112950277B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110270761.9A CN112950277B (en) 2021-03-12 2021-03-12 Intelligent playing schedule generation method based on digital media

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110270761.9A CN112950277B (en) 2021-03-12 2021-03-12 Intelligent playing schedule generation method based on digital media

Publications (2)

Publication Number Publication Date
CN112950277A true CN112950277A (en) 2021-06-11
CN112950277B CN112950277B (en) 2021-12-14

Family

ID=76229654

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110270761.9A Active CN112950277B (en) 2021-03-12 2021-03-12 Intelligent playing schedule generation method based on digital media

Country Status (1)

Country Link
CN (1) CN112950277B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080033806A1 (en) * 2006-07-20 2008-02-07 Howe Karen N Targeted advertising for playlists based upon search queries
CN104252650A (en) * 2014-09-30 2014-12-31 北京工商大学 Automated instant digital sign advertisement release method
CN109118272A (en) * 2018-07-20 2019-01-01 山东星宏电讯有限责任公司 A kind of advertisement screen management system
CN110728538A (en) * 2019-09-29 2020-01-24 杭州联汇科技股份有限公司 Advertisement arrangement method based on following library and attribute library

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080033806A1 (en) * 2006-07-20 2008-02-07 Howe Karen N Targeted advertising for playlists based upon search queries
CN104252650A (en) * 2014-09-30 2014-12-31 北京工商大学 Automated instant digital sign advertisement release method
CN109118272A (en) * 2018-07-20 2019-01-01 山东星宏电讯有限责任公司 A kind of advertisement screen management system
CN110728538A (en) * 2019-09-29 2020-01-24 杭州联汇科技股份有限公司 Advertisement arrangement method based on following library and attribute library

Also Published As

Publication number Publication date
CN112950277B (en) 2021-12-14

Similar Documents

Publication Publication Date Title
CN103473291B (en) Personalized service recommendation system and method based on latent semantic probability models
CN103793537B (en) System for recommending individual music based on multi-dimensional time series analysis and achieving method of system
Apala et al. Prediction of movies box office performance using social media
CN105574159B (en) A kind of user&#39;s portrait method for building up and user&#39;s portrait management system based on big data
CN103049440B (en) A kind of recommendation process method of related article and disposal system
CN101981563B (en) The method and apparatus for the related content shown is selected in conjunction with media
CN101281520B (en) Interactive physical training video search method based on non-supervision learning and semantic matching characteristic
CN108595550A (en) A kind of music commending system and recommendation method based on convolutional neural networks
CN110874780A (en) Scenic spot playing system and recommendation method based on big data statistics
CN109241341A (en) A kind of video display of knowledge based map preferentially recommender system and method
CN102654860A (en) Personalized music recommendation method and system
CN103838857B (en) Automatic service combination system and method based on semantics
CN107918657A (en) The matching process and device of a kind of data source
CN110717093A (en) Spark-based movie recommendation system and method
CN106951471A (en) A kind of construction method of the label prediction of the development trend model based on SVM
CN111125429A (en) Video pushing method and device and computer readable storage medium
CN108595526B (en) Resource recommendation method and device
CN106604068B (en) A kind of method and its system of more new media program
Liu Music trend prediction based on improved LSTM and random forest algorithm
CN105608118B (en) Result method for pushing based on customer interaction information
CN112950277B (en) Intelligent playing schedule generation method based on digital media
CN109874032A (en) The program special topic personalized recommendation system and method for smart television
CN110213660A (en) Distribution method, system, computer equipment and the storage medium of program
KR102643159B1 (en) A matching method that finds empty space in lcl containers in real time during container import and export
Biramane et al. Relationships between classical factors, social factors and box office collections

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