CN115311104A

CN115311104A - Opinion leader dynamic selection method, device and equipment for digital community propaganda

Info

Publication number: CN115311104A
Application number: CN202210988295.2A
Authority: CN
Inventors: 单菲
Original assignee: Avatr Technology Chongqing Co Ltd
Current assignee: Avatr Technology Chongqing Co Ltd
Priority date: 2022-08-17
Filing date: 2022-08-17
Publication date: 2022-11-08

Abstract

The embodiment of the invention relates to the field of social networks, and discloses a method, a device and equipment for dynamically selecting opinion leaders for digital community propaganda, wherein the method comprises the following steps: acquiring the propagation volume of each opinion leader in a target digital community, wherein the propagation volume corresponds to the propaganda cost of the opinion leader; combining at least one opinion leader to obtain different candidate sets; and determining a target set from the different candidate sets based on a preset screening rule, wherein the target set is the candidate set with the maximum set sound volume in the total propaganda budget. By applying the technical scheme of the invention, the opinion leader combination with the largest total propagation volume and no more than the total propaganda budget can be selected from a plurality of candidate sets, the selection of the opinion leaders in the target digital community can be realized, the dynamic planning of maximum propagation in the budget range and the maximization of the propagation value can be realized, and the operational optimization of propaganda resource allocation can be realized.

Description

Opinion leader dynamic selection method, device and equipment for digital community propaganda

Technical Field

The embodiment of the invention relates to the field of social networks, in particular to a method, a device and equipment for dynamically selecting an opinion leader for digital community propaganda.

Background

The digital community belongs to a social network, and users can participate in operations of forwarding, commenting, praising and the like on community topics. In the digital community, a part of users have more connection with surrounding users, can influence their followers from own viewpoints, and even dominate public opinion trends, and the part of participants is called opinion leader. The opinion leader plays an important role in information dissemination, public opinion guiding supervision, promotion of business development and the like. In some implementations of propaganda opinion leaders in digital communities, all opinion leaders in a digital community are usually listed first; then calculating the importance degree value of the opinion leaders in the digital community based on an algorithm, and sequencing the opinion leaders according to the importance degree value; and finally, selecting the opinion leader with the front importance for propaganda. However, in the case of a limited budget, due to the difference in propaganda cost of opinion leaders, the opinion leaders ranked first are selected in a complete sequence for propaganda, and sometimes the best propagation effect is not obtained.

Disclosure of Invention

In view of the above problems, the present application provides a method, an apparatus, and a device for dynamically selecting opinion leaders for digital community propaganda, which are used to solve the problem that in the prior art, opinion leaders ranked at the top are selected according to an importance order for propaganda, and sometimes the best propagation effect cannot be obtained.

According to an aspect of an embodiment of the present invention, there is provided a method for dynamically selecting an opinion leader for digital community promotion, the method including:

acquiring the propagation volume of each opinion leader in a target digital community, wherein the propagation volume corresponds to the propaganda cost of the opinion leader;

combining at least one opinion leader to obtain different candidate sets, wherein the candidate sets correspond to set cost and set sound volume, the set cost is the sum of propaganda cost of the candidate sets containing the opinion leaders, and the set sound volume is the sum of propagation sound volume of the candidate sets containing the opinion leaders;

and determining a target set from the different candidate sets based on a preset screening rule, wherein the target set is a candidate set with the largest aggregate sound volume in the total propaganda budget, and the aggregate cost corresponding to the target set is less than or equal to the total propaganda budget of the target digital community.

In an optional manner, after the determining the target set from the different candidate sets based on the preset filtering rule, the method further includes:

monitoring opinion leader information existing in the target digital community, wherein the opinion leader information at least comprises one or more combinations of opinion leader quantity, propaganda cost and propagation volume;

and when the opinion leader information changes, determining a target set from the different candidate sets based on the preset screening rule again.

monitoring the total propaganda budget of the target digital community;

and when the total promotion budget is changed, executing the step of determining the target set from the different candidate sets based on the preset screening rule again.

In an alternative mode, the obtaining the propagation volume of each opinion leader in the target digital community further includes:

acquiring the number of users of the target digital community, the number of fans of each user and the number of text pushing reprints of each user;

determining users with the number of fans exceeding a first threshold and/or the number of text pushing reprints exceeding a second threshold as opinion leaders;

and determining the propagation volume of the opinion leaders based on the number of the vermicelli of the opinion leaders, the number of the tweet reprints and the number of the users of the target digital community.

In an optional manner, after determining the target set from the different candidate sets based on the preset screening rule, the method further includes:

when the users of the target digital community meet preset conditions, the users are regarded as important opinion leaders, and the preset conditions at least comprise one or more combinations of highest grade value, highest popularity value or most vermicelli number;

and when the important opinion leader of the target digital community changes or the propaganda cost of the important opinion leader changes, determining a target set from the different candidate sets based on the preset screening rule again.

In an alternative mode, the combining at least one of the opinion leaders to obtain different candidate sets further comprises:

obtaining a total propagation evaluation index based on the propagation volume sum of the opinion leaders and the total propaganda budget of the target digital community;

determining the personal propagation evaluation index of each opinion leader based on the propagation volume of each opinion leader and the corresponding propaganda cost;

and combining opinion leaders with the personal propagation evaluation indexes more than or equal to the overall propagation evaluation index to obtain different candidate sets.

In an optional manner, the determining a target set from the different candidate sets based on a preset filtering rule further includes:

creating a comparison matrix, wherein the comparison matrix comprises the different candidate sets and corresponding set costs;

when the row-by-row traversal is carried out to the ith row in the comparison matrix, a first candidate set, a second candidate set and a third candidate set are obtained; the first candidate set is a candidate set corresponding to the i-1 row vector with the same cost; the second candidate set is a candidate set corresponding to the i-row vector, and the set acoustic quantity of the second candidate set further comprises an acoustic quantity corresponding to the residual budget; the third candidate set is a candidate set with the largest set volume obtained in the traversal process;

when the set cost of the candidate set corresponding to the ith row vector is smaller than the total propaganda budget, determining the candidate set with the largest set sound volume in the first candidate set, the second candidate set and the third candidate set as the target set; otherwise, the first candidate set is determined as the target set.

According to another aspect of the embodiments of the present invention, there is provided an opinion leader dynamic selection apparatus for digital community publicity, the apparatus including:

the acquisition module is used for acquiring the propagation volume of each opinion leader in the target digital community, wherein the propagation volume corresponds to the propaganda cost of the opinion leader;

the analysis module is used for combining at least one opinion leader to obtain different candidate sets, wherein the candidate sets correspond to set cost and set acoustic volume, the set cost is the sum of propaganda cost of the opinion leaders contained in the candidate sets, and the set acoustic volume is the sum of transmission acoustic volumes of the opinion leaders contained in the candidate sets;

and the execution module is used for determining a target set from the different candidate sets based on a preset screening rule, wherein the target set is a candidate set with the largest set sound volume in the total propaganda budget, and the set cost corresponding to the target set is less than or equal to the total propaganda budget of the target digital community.

According to another aspect of the embodiments of the present invention, there is provided an opinion leader dynamic selection device for digital community publicity, including: the system comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete mutual communication through the communication bus;

the memory is used for storing at least one executable instruction, and the executable instruction enables the processor to execute the operation of any one opinion leader dynamic selection method for digital community publicity.

According to another aspect of the embodiments of the present invention, there is provided a computer-readable storage medium, wherein at least one executable instruction is stored in the storage medium, and when the executable instruction is executed on an opinion leader dynamic selection device/apparatus for digital community promotion, the opinion leader dynamic selection device/apparatus for digital community promotion performs any one of the operations of the opinion leader dynamic selection method for digital community promotion.

The beneficial effect of this application: by acquiring the propagation volume, the propagation effect of the opinion leader can be obtained; by constructing a candidate set, different opinion leaders in the digital community can be combined; by constructing the set cost and the set volume, the propagation effect and the propaganda expense of each candidate set can be determined; by determining the target set, the opinion leader combination with the largest total propagation volume and no more than the total propaganda budget can be selected from a plurality of candidate sets, the opinion leaders in the target digital community can be selected, the dynamic planning of maximum propagation in the budget range and the maximization of the propagation value can be realized, and the operational optimization of propaganda resource allocation can be realized.

The foregoing description is only an overview of the technical solutions of the embodiments of the present invention, and the embodiments of the present invention can be implemented according to the content of the description in order to make the technical means of the embodiments of the present invention more clearly understood, and the detailed description of the present invention is provided below in order to make the foregoing and other objects, features, and advantages of the embodiments of the present invention more clearly understandable.

Drawings

The drawings are only for purposes of illustrating embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:

FIG. 1 is a flow chart illustrating a first embodiment of the method for dynamic selection of digital community publicity opinion leaders in accordance with the present invention;

FIG. 2 is a flow chart illustrating a second embodiment of the method for dynamic selection of digital community publicity opinion leaders in accordance with the present invention;

FIG. 3 is a flow chart illustrating a third embodiment of the method for dynamic selection of the opinion leaders for digital community promotion according to the present invention;

FIG. 4 is a flow chart illustrating a fourth embodiment of the method for dynamic selection of the opinion leaders for digital community promotion of the present invention;

FIG. 5 is a flow chart illustrating a fifth embodiment of the method for dynamic selection of digital community publicity opinion leaders in accordance with the present invention;

FIG. 6 is a flow chart illustrating a sixth embodiment of the method for dynamic selection of digital community publicity opinion leaders in accordance with the present invention;

FIG. 7 is a schematic diagram of an embodiment of the dynamic opinion leader election device for digital community promotion according to the present invention;

fig. 8 is a schematic structural diagram of an embodiment of the dynamic opinion leader election device for digital community promotion according to the present invention.

Detailed Description

Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be embodied in various forms and should not be limited to the embodiments set forth herein.

The Opinion Leader (KOL) is generally defined as a user with more accurate information, and the Opinion of the Opinion Leader is accepted or trusted by a related group and has a great influence on the behavior of the group.

The impact of digital propagation is increasing. Opinion leadership with digital communities is an imperative choice for suppliers, or brand parties.

Therefore, it is a target of advertising business to select a plurality of proper opinion leaders within a budget range so that the total amount of propagation sound of each opinion leader becomes maximum.

Fig. 1 is a flowchart illustrating a first embodiment of the method for dynamic selection of opinion leaders for digital community promotion according to the present invention, which may be performed by an opinion leader dynamic selection method apparatus/device for digital community promotion. As shown in fig. 1, the method comprises the steps of:

step 110: and acquiring the propagation volume of each opinion leader in the target digital community, wherein the propagation volume corresponds to the propaganda cost of the opinion leader.

The propagation volume of the opinion leader can be understood as the sum of times that the opinion leader publishes opinions mentioned in a certain period of time, the mentioned sum can include operations such as publishing, commenting, reprinting, and like operations, and the propagation volume can reflect the influence of the opinion leader on other users in the digital community.

For example, opinion leader a published tweets in digital community a to promote brand a products;

then, the propagation volume of the opinion leader A can be determined by counting the sum of the times that brand A products appear in the digital community A, appear in the form of graphic notes, videos or the like and are mentioned.

It will be appreciated that the amount of sound transmitted will vary from opinion leader to opinion leader, and that the promotional cost will typically vary from opinion leader to opinion leader.

For example, the dimensionless value of the propagated sound volume of the opinion leader A is 15, the dimensionless value of the propagated sound volume of the opinion leader B is 30; the hype cost of the a opinion leader may be configured to be 300 dollars and the hype cost of the B opinion leader may be configured to be 600 dollars.

Generally, in the same digital community, the higher the propagation volume of the opinion leader is, the higher the promotion cost is.

It should be noted that the cost of advertising the opinion leader is usually determined by the supplier; the supplier may correct the promotion cost when the opinion leader disagrees with the promotion cost determined by the supplier.

Step 120: and combining at least one opinion leader to obtain different candidate sets, wherein the candidate sets correspond to a set cost and a set sound volume, the set cost is the sum of the propaganda costs of the candidate sets containing the opinion leaders, and the set sound volume is the sum of the propagation sound volumes of the candidate sets containing the opinion leaders.

When the target digital community is advertised, because the total advertising budget is usually limited and fixed, how to select the opinion leader is involved to obtain the maximum total propagation sound volume, which is the collective sound volume corresponding to the candidate set.

In the target digital community, opinion leaders in different combinations are selected to obtain different candidate sets 1, 2, \ 8230and N.

Based on the obtained candidate set N, the promotion costs of the opinion leaders are added to obtain the set cost of the candidate set N.

Accordingly, the propagation acoustic volume of each opinion leader included in the candidate set N is added to obtain the set acoustic volume of the candidate set N.

For example, the total number of opinion leaders in the target digital community is 50, and the candidate set 1 and the candidate set 2 can be obtained by combining the 50 opinion leaders.

Candidate set 1 may include: an opinion leader 1 and an opinion leader 2;

wherein, the propagation volume of the opinion leader 1 is 10, and the propaganda cost is 10;

the propagation volume of the opinion leader 2 is 20, and the propaganda cost is 20;

candidate set 1 has a set cost of 30 and a set volume of 30.

Candidate set 2 may include: opinion leader 1, opinion leader 9, opinion leader 50;

the spreading sound volume of the opinion leader 9 is 50, and the propaganda cost is 60;

the spreading sound volume of the opinion leader 50 is 200, and the propaganda cost is 300;

candidate set 2 has an aggregate cost of 370 and an aggregate acoustic volume of 260.

Although the candidate set 1 includes only 2 opinion leaders and the candidate set 2 includes only 3 opinion leaders in the above example, the application does not specifically limit the number of opinion leaders included in the candidate set, and does not specifically limit the propagation volume and the promotion cost corresponding to the opinion leaders.

It will be appreciated that in theory the candidate set may include a maximum number of opinion leaders of 50 and a minimum number of 1.

Step 130: and determining a target set from the different candidate sets based on a preset screening rule, wherein the target set is a candidate set with the largest aggregate sound volume in the total propaganda budget, and the aggregate cost corresponding to the target set is less than or equal to the total propaganda budget of the target digital community.

Among the candidate sets, the candidate set which has the largest set sound volume and the set cost not exceeding the total target digital community propaganda budget is determined as the target set.

It can be understood that under the condition that the total budget of target digital community propaganda is fixed, a plurality of opinion leaders included in a target set are selected for promotion and propaganda, and the maximum total propagation sound volume can be obtained.

The target collection includes a combination of opinion leaders that may affect other users of the target digital community to a maximum extent.

When the supplier puts in the propaganda in the target digital community, the supplier selects a plurality of opinion leaders included in the target set, so that the promoted content can be mentioned in the target digital community most frequently.

For example, a target digital community a is advertised, the total advertised budget for the target digital community a is 550 yuan, and the target digital community a includes the following opinion leaders:

an opinion leader 1, the propagation volume of sound is 10, and the propaganda cost is 100;

opinion leaders 2, the propagation volume is 20, and the propaganda cost is 200;

opinion leader 3, propagation volume 30, and advertising cost 300.

Combining the 4 opinion leaders to obtain a candidate set as follows:

candidate set 1 (opinion leader 1), set volume 10, set cost 100;

candidate set 2 (opinion leader 2), set volume 20, set cost 200;

candidate set 3 (opinion leader 3), set volume 30, set cost 300;

candidate set 4 (opinion leader 1, opinion leader 2), aggregate volume 30, aggregate cost 300;

candidate set 5 (opinion leader 1, opinion leader 3), aggregate volume 40, aggregate cost 400;

candidate set 6 (opinion leader 2, opinion leader 3), aggregate volume 50, aggregate cost 500;

candidate set 7 (opinion leader 1, opinion leader 2, opinion leader 3), aggregate volume 60, and aggregate cost 600.

First, removing the candidate set 7 with the set cost greater than the total promotion budget 550 elements from the plurality of candidate sets to obtain a candidate set with the set cost less than or equal to the total promotion budget 550 elements, including:

candidate set 1 (opinion leader 1), set volume 10, set cost 100;

candidate set 2 (opinion leader 2), set volume 20, set cost 200;

candidate set 3 (opinion leader 3), set volume 30, set cost 300;

candidate set 6 (opinion leader 2, opinion leader 3), aggregate volume 50, and aggregate cost 500.

Then, a candidate set 6 with the largest set volume is selected from the 6 candidate sets, and the candidate set 6 is determined as a target set.

It can be understood that under the condition that the expense is not more than 550 yuan of the total propaganda budget, the opinion leader 2 and the opinion leader 3 in the target set are selected for promotion and propaganda, the maximum set sound volume, namely the total of the opinion leader propagation sound volumes, can be obtained, and the best propagation effect can be obtained.

In the implementation process of the opinion leader dynamic selection method for digital community propaganda, the propagation effect of the opinion leader can be obtained by acquiring the propagation volume; by constructing a candidate set, different opinion leaders in the digital community can be combined; by constructing the set cost and the set volume, the propagation effect and the propaganda expense of each candidate set can be determined; by determining the target set, the opinion leader combination with the largest total propagation volume and no more than the total propaganda budget can be selected from a plurality of candidate sets, the opinion leaders in the target digital community can be selected, the dynamic planning of maximum propagation in the budget range and the maximization of the propagation value can be realized, and the operational optimization of propaganda resource allocation can be realized.

Fig. 2 is a flowchart illustrating a second embodiment of the method for dynamically selecting opinion leaders for digital community promotion according to the present invention, which may be implemented by the apparatus/device for dynamically selecting opinion leaders for digital community promotion. As shown in fig. 2, after determining the target set from the different candidate sets based on the preset filtering rule, the method further includes the following steps:

step 210: and monitoring opinion leader information existing in the target digital community, wherein the opinion leader information at least comprises one or more combinations of opinion leader quantity, propaganda cost and propagation volume.

Wherein, the opinion leader quantity can be changed frequently according to the affirmation rule of the opinion leader in the target digital community, including the increase and decrease of the opinion leader.

The cost of advertising the opinion leaders may also vary.

For example, the supplier actively adjusts the corresponding relation between the propagation volume and the promotion cost in the target digital community, or the opinion leader actively initiates the promotion cost change.

And monitoring the sum of times of referring the opinion leader opinions in a selected time period, and determining the transmission sound volume of the opinion leader in the time period.

The influence factors of the propagation volume can include operation data such as publishing, commenting, reprinting and praise, and the propagation volume can reflect the influence of the opinion leader on other users in the digital community.

It can be understood that after the target set is determined, data such as the number of the expected leaders, the propaganda cost, the propagation volume and the like in the target digital community are monitored, and whether the target set needs to be determined again can be judged.

When the opinion leader information is changed, the opinion leader combination selected by the current target set may not be the optimal combination, and the set sound volume of the current target set may not be the maximum propagation sound volume total within the budget range.

Step 220: and when the opinion leader information changes, determining a target set from the different candidate sets based on the preset screening rule again.

After the opinion leader information changes, the aggregation cost and the aggregation volume of the candidate aggregation may change.

Therefore, the target set needs to be determined again in a different candidate set.

And in the regenerated candidate set, determining the candidate set with the largest set sound volume and the set cost not exceeding the total budget of the target digital community propaganda as a new target set.

For example, when the promotion cost of the opinion leader 1 is adjusted from 200 to 300 or the propagation sound volume thereof is adjusted from 30 to 45, the set sound volume and the set cost of any candidate set in which the opinion leader 1 is located will change, and therefore, it is necessary to re-determine the target set among the plurality of candidate sets.

In the implementation process of the opinion leader dynamic selection method for digital community propaganda, through monitoring opinion leader information, increase and decrease of opinion leaders in a target digital community, and changes of propaganda cost and propagation volume can be found in time; by re-determining the target set, the selection of the opinion leaders can be adjusted in time, dynamic planning of maximum propagation in a budget range and maximum propagation value are realized when the change of the opinion leader information is observed, and operational optimization of propaganda resource allocation is realized.

Fig. 3 is a flowchart illustrating a third embodiment of the method for dynamically selecting opinion leaders for digital community promotion according to the present invention, which may be implemented by the apparatus/device for dynamically selecting opinion leaders for digital community promotion. As shown in fig. 3, after determining the target set from the different candidate sets based on the preset filtering rule, the method further includes the following steps:

step 310: and monitoring the total propaganda budget of the target digital community.

When the target digital community is publicized, the total publicizing budget is usually limited and fixed, and the supplier usually obtains the largest total opinion leader propagation sound volume, i.e. the collective sound volume, within the budget.

It can be understood that when the opinion leader information is kept stable and the total promotion budget changes, the collective sound volume of the currently determined target set may not be the maximum collective sound volume available within the budget.

Step 320: and when the total propaganda budget changes, executing the step of determining a target set from the different candidate sets based on the preset screening rule again.

The promotion total budget change comprises promotion total budget increase and promotion total budget decrease.

With the increase of the total budget for promotion, the number of candidate sets that may become target sets will also increase, which will result in the volume of the current target set, which is no longer the maximum volume of the available volume of the set.

When the total campaign budget is reduced, the number of candidate sets that may become target sets is also reduced, which may result in the set cost of the current target set exceeding the total campaign budget after the change.

Therefore, when the total promotion budget varies, the target set needs to be determined based on a different candidate set again.

For example, a digital promotion is made for a target digital community a, with a total promotion budget of 550 dollars, which includes the following opinion leaders:

opinion leader 3, propagation volume 30, and propaganda cost 300.

Combining the opinion leaders in the step 4 to obtain a candidate set as follows:

candidate set 1 (opinion leader 1), set volume 10, set cost 100;

candidate set 2 (opinion leader 2), set volume 20, set cost 200;

candidate set 3 (opinion leader 3), set volume 30, set cost 300;

candidate set 4 (opinion leader 1, opinion leader 2), set volume 30, set cost 300;

candidate set 5 (opinion leader 1, opinion leader 3), set volume 40, set cost 400;

candidate set 6 (opinion leader 2, opinion leader 3), set volume 50, set cost 500;

When the total promotion budget of the digital community A is 550 yuan, the target set is a candidate set 6, the set sound volume is 50, and the set cost is 500.

When the total promotion budget of the digital community A is reduced to 450 yuan, the target set is determined based on different candidate sets again, the new target set is a candidate set 5, the set volume is 40, and the set cost is 400.

When the total promotion budget of the digital community A is increased to 700 yuan, the target set is determined based on different candidate sets again, the new target set is a candidate set 7, the set sound volume is 60, and the set cost is 600.

In the implementation process of the opinion leader dynamic selection method for digital community propaganda, the total budget of propaganda is monitored, and the correction of a target set can be triggered in time; the target set is re-determined when the total propaganda budget changes, the selection of the opinion leader can be adjusted in time, the situation that the set cost of the target set exceeds the total propaganda budget can be avoided, the situation that the set volume of the target set cannot be maximized due to the fact that the total propaganda budget cannot be fully utilized can be avoided, dynamic planning of maximum propagation in the budget range and the maximization of the propagation value can be achieved when the total propaganda budget changes, and operational planning optimization of propaganda resource allocation is achieved.

Fig. 4 is a flowchart illustrating a fourth embodiment of the method for dynamically selecting opinion leaders for digital community promotion according to the present invention, which can be performed by the apparatus/device of the method for dynamically selecting opinion leaders for digital community promotion. As shown in fig. 4, the obtaining the propagation volume of each opinion leader in the target digital community further includes the following steps:

step 410: and acquiring the number of users of the target digital community, the number of fans of each user and the number of text pushing reprints of each user.

When a supplier advertises a target digital community, the number of users of the target digital community can be obtained first, and the number of the users is usually the number of registered users.

After the user registers, the digital community opens the operation authority for publishing, commenting, reprinting, approving and the like of the user.

Then, the number of fans of each user can be counted, that is, how many other users pay attention to the user in the target digital community.

Generally, the more fans a user has, the larger the amount of sound transmitted by the user will be.

In an alternative mode, the number of reprints of the published tweet of the user can be counted, and the counted value can represent the number of reprints of the published tweet of the user by other users of the target digital community. The number of comments and the number of praises of the user in the target digital community can be counted.

Step 420: and (4) identifying the users with the number of fans exceeding the first threshold and/or the number of text pushing reprints exceeding the second threshold as opinion leaders.

The first threshold value can be determined according to the number of registered users of the target digital community.

For example, the first threshold may be configured to be 5% of the number of users, so that users having a fan count of 5% or more of the number of users may be considered as opinion leaders.

It is to be appreciated that when the number of user fans is determined to exceed the first threshold, the user can be identified as an opinion leader, which can affect more other users.

The second threshold may be configured according to the liveness of the targeted digital community.

For example, the second threshold may be configured to be 100 times, and when the tweet published by user a is reloaded more than the second threshold 100 times within a set time period, the user may be considered as opinion leader.

It is understood that one or a combination of a number of fans and a number of tweets may be used as one of the conditions for identifying opinion leaders, depending on the actual situation.

In an alternative way, the opinion leader is determined in the digital community, and can also be realized by the following method.

And comprehensively evaluating the data of the number of text pushing creations, the number of fans, the number of forwarding, the number of comments, the number of praise and the like of the user so as to determine the opinion leader.

For example, in the target digital community, indexes such as the number of text pushing creations, the number of fans, the number of forwarding, the number of comments, the number of praise and the like of all users are respectively scored according to the overall sequence;

then, setting corresponding proportionality coefficients for the indexes respectively;

multiplying the scores of all indexes by a proportional coefficient for each user, and adding a plurality of values obtained after multiplication, wherein the larger the obtained value is, the larger the probability that the user is the opinion leader is;

conversely, the smaller the value obtained, the smaller the probability that the user is the opinion leader.

A threshold value may be set, and a user whose value obtained by adding the above-mentioned plurality of values is equal to or greater than the threshold value is determined as an opinion leader.

Step 430: and determining the propagation volume of the opinion leaders based on the number of the vermicelli of the opinion leaders, the number of the tweets and the number of the users of the target digital community.

The influence factors of the propagation volume can include the number of fans of the opinion leader, the number of users of the target digital community, and the number of text pushing and transfer of the opinion leader.

For example, a neural network model for determining the propagated sound volume may be trained by inputting the number of fans, the number of users, and the number of tweets reprints, and then manually scaling the output propagated sound volume.

After the optimal parameters of the neural network model are obtained, the parameters of the opinion leader can be input in real time, and accordingly the opinion leader propagation sound volume which is output in real time is obtained.

In an alternative mode, the propagation sound volume can be manually configured according to actual conditions.

In some small digital communities, manual configuration of the propagated sound volume will be simpler and easier to implement. For example, the propagation volume of the opinion leader may also be determined by determining an interaction assessment value of the opinion leader within the digital community.

The determining factors of the interactive evaluation value can comprise forwarding amount, evaluation amount, praise amount and the like;

and then weighting the factors and adding the factors to obtain the interaction evaluation value of the opinion leader in the digital community. The interaction evaluation value can be simply expressed as:

interaction assessment = forwarding quantity + assessment quantity + weighting 2+ praise quantity + weighting 3.

And the interactive evaluation value can reflect the interactive willingness of other users to the opinion leader in the digital community.

It can be understood that the volume of the transmitted sound is large, which means that more users in the digital community receive the resource on one hand, so that the situation that more users have missing feedback behaviors does not exist;

on the other hand, the large volume of the propagated sound also means that the resource has the characteristic of being able to propagate across user levels, because the volume of the propagated sound may be increased only if more people at user levels propagate the resource.

In the implementation process of the opinion leader dynamic selection method for digital community propaganda, the liveness of users in a community can be identified by acquiring the number of users, the number of vermicelli and the number of reprints of the pushed text; by constructing the first threshold and the second threshold, users meeting the fan count condition and/or the transfer capacity condition can be screened and identified as opinion leaders; by determining the propagation volume, the influence of each opinion leader on other users in the target digital community or the propagation capacity of the opinion leader can be determined.

Fig. 5 is a flowchart illustrating a fifth embodiment of the method for dynamic selection of opinion leaders for digital community promotion according to the present invention, which can be performed by the apparatus/device of the method for dynamic selection of opinion leaders for digital community promotion. As shown in fig. 5, the combining at least one opinion leader to obtain different candidate sets further comprises the following steps:

step 510: and obtaining a total propagation evaluation index based on the total propagation volume of the opinion leaders and the total propaganda budget of the target digital community.

The overall propagation evaluation index can be implemented as a ratio of the total propagation volume of all opinion leaders in the target digital community to the total propaganda budget of the target digital community.

It is understood that the overall propagation rating index is proportional to the total opinion leader propagation volume and inversely proportional to the total promotion budget.

The overall spread rating index, which may reflect the average of all opinion leaders about the ratio.

Step 520: and determining the personal propagation evaluation index of each opinion leader based on the propagation volume of each opinion leader and the corresponding propaganda cost.

The personal propagation evaluation index can be implemented as the ratio of the propagation volume of the individual opinion leader to the propaganda cost corresponding to the opinion leader.

It is understood that the individual propagation evaluation index is proportional to the amount of propagated sound and inversely proportional to the promotion cost.

The index of merit is propagated individually, and may reflect the above ratio of individual opinion leaders.

Step 530: and combining the opinion leaders with the individual propagation evaluation indexes more than or equal to the overall propagation evaluation index to obtain different candidate sets.

When the number of target digital community users is particularly large, the calculation amount for obtaining the candidate set by combining all opinion leaders is particularly large.

Therefore, the candidate set is obtained by combining the opinion leaders with the individual propagation evaluation indexes more than or equal to the overall propagation evaluation index, and the combination calculation amount can be effectively reduced.

The opinion leader with the personal propagation evaluation index more than or equal to the total propagation evaluation index is selected, so that the beneficial effects of spending less expense on propaganda budget and acquiring larger propagation volume can be realized.

The multiple opinion leaders with the individual propagation evaluation index being more than or equal to the overall propagation evaluation index are combined to obtain the candidate set, so that the calculation amount and the calculation time can be reduced, and the obtained target set has the same effect as the target set obtained in the embodiment.

In the implementation process of the opinion leader dynamic selection method for digital community propaganda, the ratio of the total propagation volume of all opinion leaders in the target digital community to the total propaganda budget in the target digital community can be obtained by constructing the total propagation evaluation index, and the average value of all opinion leaders about the ratio can be reflected; the personal ratio of each opinion leader can be obtained by constructing a personal propagation evaluation index; by combining the opinion leaders with the personal propagation evaluation index more than or equal to the overall propagation evaluation index, the combined calculation amount can be reduced, the propaganda budget with less expense can be realized, and the larger propagation volume can be achieved.

Fig. 6 shows a flowchart of a sixth embodiment of the invention for a method for dynamic selection of opinion leaders for digital community promotion, which may be performed by the apparatus/device for dynamic selection of opinion leaders for digital community promotion. As shown in fig. 6, after determining the target set from the different candidate sets based on the preset filtering rule, the method further includes the following steps:

step 610: and when the users of the target digital community meet preset conditions, the users are regarded as important opinion leaders, and the preset conditions at least comprise one or more combinations of highest grade value, highest popularity value or highest fan number.

Wherein, when the grade value of the opinion leader is highest in the target digital community, or the popularity value is highest, or the number of fans is the most, the opinion leader can be determined as the important opinion leader.

The preset conditions may also be combined, such as satisfying the highest rank value and the highest popularity value, or satisfying the highest popularity value and the highest number of fans, or satisfying the highest rank value, the highest popularity value and the highest number of fans, or satisfying other combinations of the above.

Although the content included in the preset condition is exemplified in the present application, the content included in the preset condition may also be configured according to the actual situation, and the present application does not specifically limit the content.

It is to be appreciated that when a user is deemed to be an important opinion leader, the user's impact on the target digital community may be deemed most important. When the information of the important opinion leader changes, the target digital community propaganda will be greatly influenced.

Step 620: and when the important opinion leader of the target digital community changes or the propaganda cost of the important opinion leader changes, determining a target set from the different candidate sets based on the preset screening rule again.

The important opinion leaders are changed, and the important opinion leaders can include the generation of new important opinion leaders, the logout of original important opinion leaders in a target digital community, or the change of the transmission sound volume of the important opinion leaders.

When the important opinion leaders change or the advertising cost changes, the suppliers need to immediately re-determine the target set because the important opinion leaders have the greatest impact on other users of the target digital community.

In the implementation process of the opinion leader dynamic selection method for digital community propaganda, users meeting the conditions can be identified as important opinion leaders by constructing preset conditions, and the users are prevented from being identified as common opinion leaders and neglecting the influence; by re-identifying the target group when the important opinion leader changes or the advertising cost thereof changes, the influence of the change of the important opinion leader on the total amount of propagation sound of the target group can be eliminated.

In an alternative approach, the problem of the present application can be solved by a knapsack problem solving concept. Within the budget control range, the maximization of the transmission value or the transmission sound volume is realized, and the method belongs to the operation and research optimization problem.

And secondary factors such as the limitation of the number of selectable opinion leaders and the limitation of the number of final opinion leaders can be ignored, and the problem is further simplified into a 0-1 knapsack problem.

For example, given n numbers of items and a backpack, each of 1;

the weight of item i is wi, its value is vi, the capacity of the backpack is c;

the problems to be solved are as follows: how to select the items to be loaded into the backpack maximizes the total value of the items loaded into the backpack.

It is understood that alternative opinion leaders (KOL) correspond to items that may be packed into a backpack, the cost of promotion (cost) for each opinion leader corresponds to the weight of the item, the volume of propagated sound (Buzz) for each opinion leader corresponds to the value of the item, and the total budget of promotion (budget) corresponds to the capacity of the backpack.

Firstly, an alignment matrix is created, wherein the alignment matrix comprises the different candidate sets and corresponding set costs.

The comparison matrix may be implemented as a two-dimensional matrix, for example, the two-dimensional matrix may include different candidate sets including different opinion leader combinations, and each candidate set corresponds to a respective set cost and set sound volume.

Secondly, when traversing to the ith row line by line in the comparison matrix, obtaining a first candidate set, a second candidate set and a third candidate set; the first candidate set is a candidate set corresponding to the i-1 row vector with the same cost; the second candidate set is a candidate set corresponding to the i row vector, and the set sound volume of the second candidate set also comprises the sound volume corresponding to the residual budget; and the third candidate set is a candidate set with the largest set volume obtained in the traversal process.

For example, in a matrix of KOL numbers (i) and costs (j), iteratively putting different opinion leader candidate sets, an opinion leader target set with the largest total propagation volume can be found, and the logic is expressed as follows:

If cost[i]<j；

b = max (maximum propagated sound volume of preceding KOL combination b [ i-1] [ j ] at the same cost, propagated sound volume of current KOL + propagated sound volume of maximum residual budget, succeeding KOL b [ i-1] [ j-cost [ i ] ]) not including the current KOL combination)

Else:

b＝b[i-1][j]。

B [ i-1] [ j ] refers to the previous row value of the i row in the matrix, and j is the total propaganda budget; the propagation acoustic volume of the current KOL refers to the collective acoustic volume of the opinion leader candidate set corresponding to the i-row vector; the maximum transmission sound volume of the residual budget can be obtained by continuously cycling and iterating; KOL b [ i-1] [ j-cost [ i ] not including the current KOL combination subsequent thereto is the total of the propagation volumes of the currently known largest opinion leader KOL combination.

Fig. 7 is a schematic structural diagram of an embodiment of the opinion leader dynamic selection device for digital community publicity according to the present invention. As shown in fig. 7, the apparatus 700 includes: an acquisition module 710, an analysis module 720, and an execution module 730.

the analysis module is used for combining at least one opinion leader to obtain different candidate sets, wherein the candidate sets correspond to a set cost and a set acoustic quantity, the set cost is the sum of propaganda costs of the opinion leaders contained in the candidate sets, and the set acoustic quantity is the sum of propagation acoustic quantities of the opinion leaders contained in the candidate sets;

an acquisition module monitors opinion leader information existing in the target digital community, wherein the opinion leader information at least comprises one or more combinations of opinion leader quantity, propaganda cost and propagation volume;

and when the opinion leader information changes, the execution module executes the step of determining the target set from the different candidate sets based on the preset screening rule again.

the acquisition module monitors the total propaganda budget of the target digital community;

and when the total propaganda budget changes, the execution module executes the step of determining the target set from the different candidate sets based on the preset screening rule again.

the acquisition module acquires the number of users of the target digital community, the number of fans of each user and the number of text pushing reprints of each user;

the acquisition module determines the users with the number of the fans exceeding a first threshold and/or the number of the text pushing reprints exceeding a second threshold as opinion leaders;

and the acquisition module determines the propagation volume of the opinion leaders based on the number of the vermicelli of the opinion leaders, the number of the text pushing reprints and the number of the users of the target digital community.

the analysis module obtains a total propagation evaluation index based on the propagation volume sum of the opinion leaders and the total propaganda budget of the target digital community;

the analysis module determines the personal propagation evaluation index of each opinion leader based on the propagation volume of each opinion leader and the corresponding propaganda cost;

and the analysis module combines the opinion leaders with the personal propagation evaluation index more than or equal to the overall propagation evaluation index to obtain different candidate sets.

In an optional manner, after the determining, based on the preset filtering rule, a target set from the different candidate sets, the method further includes:

when the users of the target digital community meet preset conditions, the users are regarded as important opinion leaders by the analysis module, and the preset conditions at least comprise one or more combinations of the highest grade value, the highest popularity value or the highest fan number;

and the execution module executes the step of determining the target set from the different candidate sets based on the preset screening rule again when the important opinion leader of the target digital community changes or the propaganda cost of the important opinion leader changes.

the execution module creates a comparison matrix, wherein the comparison matrix comprises the different candidate sets and corresponding set costs;

the execution module obtains a first candidate set, a second candidate set and a third candidate set when traversing to the ith row line by line in the comparison matrix; the first candidate set is a candidate set corresponding to the i-1 row vector with the same cost; the second candidate set is a candidate set corresponding to the i-row vector, and the set acoustic quantity of the second candidate set further comprises an acoustic quantity corresponding to the residual budget; the third candidate set is a candidate set with the largest set volume obtained in the traversal process;

when the set cost of the candidate set corresponding to the ith row vector is smaller than the total promotion budget, the execution module determines the candidate set with the maximum set sound volume in the first candidate set, the second candidate set and the third candidate set as the target set; otherwise, the first candidate set is determined as the target set.

By applying the technical scheme of the invention, the transmission effect of the opinion leaders can be obtained by acquiring the transmission sound volume; by constructing a candidate set, different opinion leaders in the digital community can be combined; by constructing the set cost and the set volume, the propagation effect and the propaganda expense of each candidate set can be determined; by determining the target set, the opinion leader combination which has the largest total propagation sound volume and does not exceed the total propaganda budget can be selected from a plurality of candidate sets, and the opinion leaders in the target digital community can be selected; by monitoring opinion leader information, the increase and decrease of the opinion leaders in the target digital community can be found in time, and the changes of the propaganda cost and the propagation volume can be realized; by re-determining the target set, the selection of the opinion leaders can be adjusted in time; by monitoring the total propaganda budget, the correction of the target set can be triggered in time; the target set is re-determined when the total propaganda budget changes, the selection of the opinion leader can be adjusted in time, the set cost of the target set can be prevented from exceeding the total propaganda budget, and the condition that the set volume of the target set cannot be maximized due to the fact that the total propaganda budget cannot be fully utilized can be avoided; the number of users, the number of fans and the number of reprints of the tweets are obtained, so that the activeness of the users in a community can be identified; by constructing the first threshold and the second threshold, users meeting the fan count condition and/or the transfer capacity condition can be screened and identified as opinion leaders; by determining the propagation volume, the influence of each opinion leader on other users in the target digital community or the propagation capacity of the opinion leader can be determined; by constructing a total propagation evaluation index, the ratio of the total propagation volume of all opinion leaders in the target digital community to the total propaganda budget in the target digital community can be obtained, and the average value of all opinion leaders about the ratio can be reflected; the personal ratio of each opinion leader can be obtained by constructing a personal propagation evaluation index; by combining the opinion leaders with the personal propagation evaluation index more than or equal to the overall propagation evaluation index, the combined calculation amount can be reduced, the propaganda budget with less expense can be realized, and the larger propagation volume can be achieved; the preset conditions are constructed, users meeting the conditions can be identified as important opinion leaders, and the situation that the users are identified as common opinion leaders and influence of the users is ignored is avoided; by re-determining the target set when the important opinion leader changes or the propaganda cost changes, the influence on the total propagation sound volume of the target set when the important opinion leader changes can be eliminated; the dynamic planning of maximum transmission in a budget range and the maximization of transmission value are realized, and the operation optimization of propaganda resource allocation is realized.

Fig. 8 is a schematic structural diagram illustrating an embodiment of the device for dynamically selecting opinion leaders for digital community publicity according to the present invention, and the specific implementation of the device for dynamically selecting opinion leaders for digital community publicity is not limited by the specific embodiment of the present invention.

As shown in fig. 8, the opinion leader dynamic selection apparatus for digital community publicity may include: a processor (processor) 802, a Communications Interface 804, a memory 806, and a communication bus 808.

Wherein: the processor 802, communication interface 804, and memory 806 communicate with one another via a communication bus 808. A communication interface 804 for communicating with network elements of other devices, such as clients or other servers. The processor 802 is configured to execute the program 810, and may specifically execute the relevant steps in the embodiment of the opinion leader dynamic selection method for digital community promotion described above.

In particular, program 810 may include program code comprising computer-executable instructions.

The processor 802 may be a central processing unit CPU, or an Application Specific Integrated Circuit ASIC (Application Specific Integrated Circuit), or one or more Integrated circuits configured to implement embodiments of the present invention. The opinion leader dynamic selection device for the digital community publicity comprises one or more processors which can be the same type of processor, such as one or more CPUs; or may be different types of processors such as one or more CPUs and one or more ASICs.

The memory 806 stores a program 810. The memory 806 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk memory.

The program 810 may be specifically invoked by the processor 802 to cause an opinion leader dynamic selection device for digital community promotion to perform the following operations:

monitoring the total propaganda budget of the target digital community;

and determining the propagation volume of the opinion leaders based on the number of the vermicelli of the opinion leaders, the number of the tweets and the number of the users of the target digital community.

determining a personal propagation evaluation index of each opinion leader based on the propagation volume of each opinion leader and the corresponding propaganda cost;

and combining the opinion leaders with the individual propagation evaluation indexes more than or equal to the overall propagation evaluation index to obtain different candidate sets.

when the users of the target digital community meet preset conditions, the users are regarded as important opinion leaders, and the preset conditions at least comprise one or more combinations of the highest grade value, the highest popularity value or the highest fan number;

when the row-by-row traversal is carried out to the ith row in the comparison matrix, a first candidate set, a second candidate set and a third candidate set are obtained; the first candidate set is a candidate set corresponding to the i-1 row vector with the same cost; the second candidate set is a candidate set corresponding to the i row vector, and the set sound volume of the second candidate set also comprises the sound volume corresponding to the residual budget; the third candidate set is a candidate set with the largest set sound volume obtained in the traversal process;

By applying the technical scheme of the invention, the transmission effect of the opinion leaders can be obtained by acquiring the transmission sound volume; by constructing a candidate set, different opinion leaders in the digital community can be combined; by constructing the set cost and the set volume, the propagation effect and the propaganda expense of each candidate set can be determined; by determining the target set, the opinion leader combination with the largest total propagation volume and no more than the total propaganda budget can be selected from a plurality of candidate sets, and the selection of the opinion leaders in the target digital community is realized; by monitoring opinion leader information, the increase and decrease of the opinion leaders in the target digital community can be found in time, and the changes of the propaganda cost and the propagation volume can be realized; by re-determining the target set, the selection of opinion leaders can be adjusted in time; by monitoring the total propaganda budget, the correction of the target set can be triggered in time; the target set is re-determined when the total propaganda budget changes, the selection of the opinion leader can be adjusted in time, the set cost of the target set can be prevented from exceeding the total propaganda budget, and the condition that the set volume of the target set cannot be maximized due to the fact that the total propaganda budget cannot be fully utilized can be avoided; the activity of the users in the community can be identified by acquiring the number of the users, the number of the fans and the number of the reprinting of the tweets; by constructing the first threshold and the second threshold, users meeting the fan count condition and/or the transfer capacity condition can be screened and identified as opinion leaders; by determining the propagation volume, the influence of each opinion leader on other users in the target digital community or the propagation capacity of the opinion leader can be determined; by constructing a total propagation evaluation index, the ratio of the total propagation volume of all opinion leaders in the target digital community to the total propaganda budget in the target digital community can be obtained, and the average value of all opinion leaders about the ratio can be reflected; the personal ratio of each opinion leader can be obtained by constructing a personal propagation evaluation index; by combining the opinion leaders with the personal propagation evaluation index more than or equal to the total propagation evaluation index, the combined calculation amount can be reduced, the propaganda budget with less expense can be realized, and larger propagation volume can be achieved; the preset conditions are constructed, users meeting the conditions can be identified as important opinion leaders, and the situation that the users are identified as common opinion leaders and influence of the users is ignored is avoided; by re-determining the target set when the important opinion leader changes or the propaganda cost changes, the influence on the total propagation sound volume of the target set when the important opinion leader changes can be eliminated; the dynamic planning of maximum propagation in a budget range and the maximization of propagation value are realized, and the operational optimization of propaganda resource allocation is realized.

The embodiment of the invention provides a computer-readable storage medium, wherein the storage medium stores at least one executable instruction, and when the executable instruction runs on an opinion leader dynamic selection device/apparatus for digital community promotion, the opinion leader dynamic selection device/apparatus for digital community promotion executes the opinion leader dynamic selection method for digital community promotion in any method embodiment.

The executable instructions may be specifically configured to cause an opinion leader dynamic selection device/apparatus for digital community promotion to perform the following operations:

and determining a target set from the different candidate sets based on a preset screening rule, wherein the target set is a candidate set with the maximum set sound volume in the propaganda total budget, and the set cost corresponding to the target set is less than or equal to the propaganda total budget of the target digital community.

monitoring the total propaganda budget of the target digital community;

By applying the technical scheme of the invention, the propagation effect of the opinion leader can be obtained by acquiring the propagation volume; by constructing a candidate set, different opinion leaders in the digital community can be combined; by constructing the set cost and the set volume, the propagation effect and the propaganda expense of each candidate set can be determined; by determining the target set, the opinion leader combination with the largest total propagation volume and no more than the total propaganda budget can be selected from a plurality of candidate sets, and the selection of the opinion leaders in the target digital community is realized; by monitoring opinion leader information, the increase and decrease of the opinion leaders in the target digital community can be found in time, and the changes of cost and propagation volume can be publicized; by re-determining the target set, the selection of the opinion leaders can be adjusted in time; by monitoring the total propaganda budget, the correction of the target set can be triggered in time; the target set is re-determined when the total propaganda budget changes, the selection of the opinion leader can be adjusted in time, the set cost of the target set can be prevented from exceeding the total propaganda budget, and the condition that the set volume of the target set cannot be maximized due to the fact that the total propaganda budget cannot be fully utilized can be avoided; the activity of the users in the community can be identified by acquiring the number of the users, the number of the fans and the number of the reprinting of the tweets; by constructing the first threshold and the second threshold, users meeting the fan number condition and/or the load transfer amount condition can be screened and identified as opinion leaders; by determining the propagation volume, the influence of each opinion leader on other users in the target digital community or the propagation capacity of the opinion leader can be determined; by constructing a total propagation evaluation index, the ratio of the total propagation volume of all opinion leaders in the target digital community to the total propaganda budget in the target digital community can be obtained, and the average value of all opinion leaders about the ratio can be reflected; the personal ratio of each opinion leader can be obtained by constructing a personal propagation evaluation index; by combining the opinion leaders with the personal propagation evaluation index more than or equal to the overall propagation evaluation index, the combined calculation amount can be reduced, the propaganda budget with less expense can be realized, and the larger propagation volume can be achieved; the preset conditions are constructed, users meeting the conditions can be identified as important opinion leaders, and the situation that the users are identified as common opinion leaders and influence of the users is ignored is avoided; by re-determining the target set when the important opinion leader changes or the propaganda cost changes, the influence on the total propagation sound volume of the target set when the important opinion leader changes can be eliminated; the dynamic planning of maximum propagation in a budget range and the maximization of propagation value are realized, and the operational optimization of propaganda resource allocation is realized.

The algorithms or displays presented herein are not inherently related to any particular computer, virtual system, or other apparatus. In addition, embodiments of the present invention are not directed to any particular programming language.

In the description provided herein, numerous specific details are set forth. It is understood, however, that embodiments of the invention may be practiced without these specific details. Similarly, in the above description of exemplary embodiments of the invention, various features of the embodiments of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the invention and aiding in the understanding of one or more of the various inventive aspects. Where the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this invention.

Those skilled in the art will appreciate that the modules in the device in an embodiment may be adaptively changed and disposed in one or more devices different from the embodiment. The modules or units or components in the embodiments may be combined into one module or unit or component, and furthermore, may be divided into a plurality of sub-modules or sub-units or sub-components. Except that at least some of such features and/or processes or elements are mutually exclusive.

It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The usage of the words first, second and third, etcetera do not indicate any ordering. These words may be interpreted as names. The steps in the above embodiments should not be construed as limiting the order of execution unless specified otherwise.

Claims

1. An opinion leader dynamic selection method for digital community propaganda is characterized by comprising the following steps:

2. The dynamic choice method of opinion leaders for digital community promotion according to claim 1, wherein after determining a target set from said different candidate sets based on a predetermined filtering rule, the method further comprises:

3. The method of claim 1, wherein after determining a target set from the different candidate sets based on a predetermined filtering rule, the method further comprises:

monitoring the total propaganda budget of the target digital community;

4. The method of claim 1, wherein the obtaining the propagation volume of each opinion leader in the target digital community further comprises:

5. The method of claim 1, wherein said combining at least one of said opinion leaders to obtain different candidate sets further comprises:

6. The method of claim 1, wherein after determining a target set from the different candidate sets based on a predetermined filtering rule, the method further comprises:

7. The dynamic selection method of opinion leaders for digital community promotion according to claim 1, wherein said determining a target set from among said different candidate sets based on a predetermined filtering rule, further comprises:

when the row-by-row traversal is carried out to the ith row in the comparison matrix, a first candidate set, a second candidate set and a third candidate set are obtained; the first candidate set is a candidate set corresponding to the i-1 row vector with the same cost; the second candidate set is a candidate set corresponding to the i-row vector, and the set acoustic quantity of the second candidate set further comprises an acoustic quantity corresponding to the residual budget; the third candidate set is a candidate set with the largest set sound volume obtained in the traversal process;

when the set cost of the candidate set corresponding to the ith row vector is smaller than the total promotion budget, determining the candidate set with the largest set sound volume in the first candidate set, the second candidate set and the third candidate set as the target set; otherwise, the first candidate set is determined as the target set.

8. An opinion leader dynamic selection device for digital community publicity, the device comprising:

9. An opinion leader dynamic selection device for digital community propaganda, comprising: the system comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete mutual communication through the communication bus;

the memory is configured to store at least one executable instruction that causes the processor to perform the operations of the method for dynamic selection of opinion leaders for digital community promotion of any one of claims 1-7.

10. A computer-readable storage medium, wherein the storage medium has stored therein at least one executable instruction, which when running on an opinion leader dynamic selection device/apparatus for digital community promotion, causes the opinion leader dynamic selection device/apparatus for digital community promotion to perform the operations of the opinion leader dynamic selection method for digital community promotion according to any one of claims 1 to 7.