CN110866766A

CN110866766A - Advertisement putting method, method for determining popularization crowd, server and client

Info

Publication number: CN110866766A
Application number: CN201810980484.9A
Authority: CN
Inventors: 杨子旭; 王月颖; 李莉; 王志勇; 张小洵
Original assignee: Alibaba Group Holding Ltd
Current assignee: Alibaba Group Holding Ltd
Priority date: 2018-08-27
Filing date: 2018-08-27
Publication date: 2020-03-06
Also published as: WO2020043001A1

Abstract

The application provides an advertisement putting method, a method for determining popularization crowd, a server and a client, wherein the method comprises the following steps: determining a significant feature group for delivering advertisements based on seed population parameters, extension parameters and extension target parameters set by a user, wherein the significant feature group comprises a plurality of significant features and weight values of the significant features; adjusting the set of salient features based on a user selection of the set of salient features; and determining similar crowds according to the adjusted significant feature group to obtain a popularization object for putting the advertisement. By the aid of the scheme, the technical problems that the extension direction of the user cannot be controlled by the user due to the fact that the existing crowd extension and screening are transparent and automatic to the user, and the extension result cannot meet the requirements of the user are solved, the technical effect that the extension will of the user can be effectively expressed is achieved, and user experience is greatly improved.

Description

Advertisement putting method, method for determining popularization crowd, server and client

Technical Field

The application belongs to the technical field of internet, and particularly relates to an advertisement putting method, a method for determining popularization crowd, a server and a client.

Background

Along with the continuous development of internet and electricity merchant, the demand of advertisement putting is also bigger and bigger, and at present, the mode that advertisement putting in-process advertiser circled the crowd mainly includes two kinds: one is a demographics-based approach and the other is a user behavior-based approach. The demographic-based dimension mainly defines the crowd based on the dimensions of gender, age, purchasing power and the like of the user, and the user behavior-based dimension mainly defines the crowd based on behaviors of the user such as browsing and clicking in an advertiser shop and the like.

However, no matter which way of enclosing people is adopted, the problem that the number of enclosed users is limited and the requirement of the marketing plan cannot be met exists.

In view of the above problems, no effective solution has been proposed.

Disclosure of Invention

The application aims to provide an advertisement putting method, a method for determining popularization crowd, a server and a client, which can realize advertisement putting meeting the demand of an advertiser and enable the intention of the advertiser to be effectively expressed.

The application provides an advertisement putting method, a method for determining popularization crowd, a server and a client, which are realized as follows:

a method of advertisement delivery, the method comprising:

determining a significant feature group for delivering advertisements based on seed population parameters, extension parameters and extension target parameters set by a user, wherein the significant feature group comprises a plurality of significant features and weight values of the significant features;

adjusting the set of salient features based on a user selection of the set of salient features;

and determining similar crowds according to the adjusted significant feature group to obtain a popularization object for putting the advertisement.

A method of determining a population to promote, the method comprising:

a setting interface for displaying seed population parameters, expansion parameters and expansion target parameters;

receiving seed crowd parameters, extension parameters and extension target parameters set by a user;

determining and displaying a significant feature group based on seed population parameters, extension parameters and extension target parameters set by a user, wherein the significant feature group comprises a plurality of significant features;

and receiving a selection operation of the user on the significant feature group, wherein the selection operation is used for adjusting the significant feature group, and determining similar people groups based on the adjusted significant feature group to obtain a popularization object.

A server comprising a processor and a memory for storing processor-executable instructions, the instructions when executed by the processor implementing the steps of the method of:

A client comprising a processor and a memory for storing processor-executable instructions, the instructions when executed by the processor implementing the steps of the method of:

A computer readable storage medium having stored thereon computer instructions which, when executed, implement the steps of the above-described method.

The advertisement putting method, the method for determining the popularization crowd, the server and the client side set the entrance which can set the seed crowd parameter, the extension target parameter and the significant feature group for the user with the popularization requirement, so that the extension intention of the user can be effectively expressed, the technical problems that the extension direction of the user cannot be controlled by the user due to the fact that the existing crowd extension and screening are transparent to the user and the extension result cannot meet the user requirement are solved, the technical effect that the extension intention of the user can be effectively expressed is achieved, and the user experience is greatly improved.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only some embodiments described in the present application, and for those skilled in the art, other drawings can be obtained according to the drawings without any creative effort.

FIG. 1 is an architecture diagram of an advertisement delivery system provided herein;

FIG. 2 is a block diagram of an advertising system provided herein;

FIG. 3 is a flow chart of a method of advertising provided by the present application;

FIG. 4 is a schematic diagram of extended parameter selection provided herein;

FIG. 5 is a schematic illustration of salient feature placement provided herein;

FIG. 6 is a flow chart of a method of advertising provided by the present application;

fig. 7 is a schematic architecture diagram of an advertisement delivery server provided in the present application.

Detailed Description

In order to make those skilled in the art better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, 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 application.

In order to expand the population, the current common mode is that an advertiser determines a part of target population, and then, through the similar population expansion function, the existing target people are used as seed users, and people similar to the seed users are found as the intention population to carry out advertisement putting. In implementation, generally, the crowd is defined based on the dimension such as gender, age, purchasing power and the like of the user, and the crowd is defined based on the behavior such as browsing and clicking of the user in an advertiser shop and the like based on the dimension of the user behavior. However, no matter which way of enclosing people is adopted, the problem that the number of enclosed users is limited and the requirement of the marketing plan cannot be met exists. Furthermore, in the current advertisement delivery system, after an advertiser generally selects an expansion scale, the system automatically generates expansion crowds meeting requirements, the advertiser cannot control the expansion direction of the intended crowds, the expansion willingness of the advertiser cannot be met and reflected, and the user experience effect is poor.

It is considered that if an advertisement delivery system is provided to provide a user with an option of expressing the expansion direction of the intended population, the expansion will of the advertiser can be satisfied, and the final expansion result can better meet the user demand. To this end, an advertisement delivery system is provided, as shown in fig. 1, which may include: a client 101 and a server 102. Under the condition that the user selects to perform crowd extension, the server 102 can push a setting interface for setting the extension intention, the user can set the expected crowd extension direction through the setting interface displayed on the client, after the user selects the crowd extension intention, the server 102 can perform crowd extension according to the extension intention set by the user, and advertisement putting is performed on the extended crowd.

The population expansion is based on the seed population, and similar populations of the seed population are determined to obtain populations of a preset number and scale. For example, the seed population is 10, and population expansion can be performed to obtain 90 people, so that 100 people exist, namely, population expansion is realized.

In an embodiment, the server 102 may be a single server, or a server cluster, or a server disposed in a cloud, and which type of server is specifically adopted may be selected according to actual needs, which is not limited in this application.

The client may be a terminal device or software operated and used by a user. Specifically, the client may be a terminal device such as a smart phone, a tablet computer, a notebook computer, a desktop computer, a smart watch, or other wearable devices. Of course, the user terminal may also be software that can run in the terminal device. For example: APP, browser and other application software.

For the service end, in order to effectively express the expansion intention of the user (i.e., the advertiser), so that the group to which the final advertisement is promoted more meets the requirements of the advertiser, an entrance capable of expressing the expansion intention can be provided for the advertiser, and the expanded population is determined by combining the promotion condition and information expected by the advertiser.

When implemented, portals in which advertisers participate may be set up, but are not limited to, at one or more of the following nodes:

1) selecting a seed population;

2) setting a crowd expansion target;

3) setting a feature weight.

The contents of the above nodes are described as follows:

1) selecting seed population:

the advertiser can set some parameter information of the crowd expected to be pushed in the interface for selecting the seed crowd, such as: target population, population characteristics, etc. The target population may be, for example: non-triggering users, potential users, action users, and purchasing users, etc., the user may select or set what type of user the desired target group is as desired. The demographic characteristics may be, for example: gender, age, city, etc.

In the actual implementation process, if the number of the promotion persons desired by the advertiser is small, two conditions are set: after the target population and the population characteristics, the population number in the formed population set reaches the advertisement promotion amount desired by the advertiser, and then population expansion is not needed. Therefore, a reminding interface can be set to remind the advertiser of finishing the advertisement promotion requirement without crowd expansion.

The population expansion is performed only when the number of people to be promoted is not satisfied by the number of people in the population group obtained by setting the two conditions.

For example, if the number of people the advertiser wishes to promote is 30 ten thousand, and after the target population and the demographic characteristics are set, the number of people obtained is 35 ten thousand, then population expansion is not necessary. If the number of people the advertiser wishes to promote is 100 ten thousand and the number of people obtained after setting the target people and the crowd characteristics is 35 ten thousand, then crowd expansion is required.

Through the mode of determining whether the crowd expansion is needed or not according to the number of the expected promoted crowds and reminding the user, the accuracy and the effectiveness of the user reached by advertisement promotion can be effectively improved, unnecessary operation of the system cannot be caused, and the cost and the efficiency can be saved.

The above two conditions are set: after the target population and the population characteristics, the formed population set can be used as a seed population for use in subsequent population popularization.

However, it should be noted that the option information listed in the target group and the group characteristics set forth above is only a schematic description, and may be selected according to needs in practical use, and the application is not limited thereto.

In the event that it is determined that population expansion is desired, the advertiser may be provided with the right to select an expanded population. For example, the advertiser may input the crowd expansion amount, set the expansion multiple, input the number, drag the progress bar, set the expansion multiple or the expansion amount, and select the mode for the advertiser, which is specifically adopted, and may be selected according to actual needs.

2) Setting a crowd extension target:

in the event that it is determined that population expansion is required, the advertiser may be provided with a plurality of expansion targets for selection by the advertiser, such as: brand promotion, effect promotion, click priority, conversion priority, and the like. That is, one or more of these expansion targets may be, but are not limited to being, selected by the advertiser.

Based on the selectable expansion targets, the advertiser can select the desired expansion target according to the advertisement promotion requirement. For example, if the promotion of the current advertisement by the advertiser is mainly to realize brand promotion, the brand promotion can be selected as an extension target, and if the advertiser wants that the promotion of the current advertisement can be willing to be clicked by more users, the click priority can be selected as an extension target.

It should be noted, however, that the above listed expansion objectives are only an exemplary description, and that other expansion objectives, such as: purchase priority, etc., which the present application does not limit.

3) Setting a characteristic weight:

considering that only the right for the user to select the seed population and the population expansion target is provided, the aim of accurately reflecting the intention of the advertiser cannot be achieved.

Therefore, considering that the advertiser can be provided with the authority for setting the feature weight, after the advertiser selects the seed crowd, sets the number of expansion people and sets the crowd expansion target, different expansion targets can be associated to different feature groups and combined.

For example, since an advertiser desires to promote a certain product and needs to define a certain scale of brand-oriented population, the advertiser selects "brand promotion" when selecting an expansion target. The server selects a feature group corresponding to "brand promotion", for example, a brand preference degree feature in each major brand, or the like, in a case where it is determined that the advertiser selects the brand promotion as the extension target. If "effect promotion" and "conversion priority" are selected at the extension target option, a feature group corresponding to "effect promotion" and "conversion priority" may be selected, for example: buying intent demographics under each store.

Based on the seed population, the expansion target and the expansion population selected by the advertiser, the feature group provided by the server is subjected to expansion modeling, and the significant feature group is screened out, namely, the features which have great influence on the expansion result are determined, the features are displayed to the advertiser to enable the advertiser to select whether to adopt the features, and the advertiser has the right to delete the undesired features of the advertiser.

Further, in order to further increase the control granularity of the advertiser on the expansion result, a weight value may be set for each feature in the significant feature group, where a high weight value indicates a large influence on the expansion result, and a low weight value indicates a small influence on the expansion result. The server can give a default significant feature group according to the seed population, the extended target and the extended population, and the weight value of each feature in the significant feature group is displayed to the advertiser. The advertiser can modify the weighted value of each characteristic according to the demand and the intention of the advertiser, so that the control on the crowd expansion result is realized, namely, the control on finer granularity is realized.

When generating the set of salient features, the set of salient features may be generated in one or more of, but not limited to: point-to-point information index, information gain, chi-square test value, etc. The method may be to select one of the manners to generate the significant feature group and the suggested weight value of each feature, or to select two or more manners to generate the significant feature group and the suggested weight value of each feature, and then cross-verify or perform weighted average on the results obtained in the two or more manners to obtain a more reasonable and accurate result.

For example, the features in the generated salient feature group may include: buyer star rating, last monthly spending limit, last week to store times, make-up preferences, suit-dress preferences, and the like.

In order to determine the salient feature group, the importance degrees of the obtained plurality of features need to be ranked, and a certain number of features, i.e., features with higher importance degrees, are selected as salient features. For example: and ranking the importance scores of the features, selecting the top K features with the highest score as the significant features, and taking the importance scores as the suggested weights corresponding to the significant features.

Considering that the population covered by the features has a certain overlap ratio, in order to ensure that the number of the final expansion result can meet the set expansion population number, the accumulated total number of the population covered by the previous K features can be controlled to be α times of the expected expansion population size, wherein α is an overlap coefficient.

For example: the advertiser chooses to down-adjust the feature "age" weight to 0, i.e., chooses to delete the feature. The weight value of the feature "the latest purchase amount" is adjusted from 0.3 to 1.0, and the weight value of the feature "the suit dress preference" is adjusted from 0.4 to 0.1. After the weight is adjusted for each significant feature value, normalization processing can be performed on the weight values to serve as final weight values.

In one embodiment, crowd expansion may be triggered after the advertiser completes setting the weight values for each salient feature. In order to realize crowd extension, each user in a full user group formed by user groups corresponding to the significant features can be scored based on each significant feature, the users are sorted according to scores, and a preset extended number of users with the scores higher than the scores are selected as users to be popularized.

However, it should be noted that the above-mentioned manner for determining the extended population based on the weighted values of the salient features is only an exemplary description, and other manners may be adopted in practical implementation, for example, the number of users of the product of the total number of extensions and the corresponding weighted value is selected from the group of people corresponding to the weighted value of each salient feature as the final extended user according to the weighted value after the normalization processing. The specific mode is selected according to actual needs, and the method is not limited in the application.

The foregoing advertisement delivery system and method will be described with reference to a specific embodiment, however, it should be noted that the specific embodiment is only for better describing the present application and is not to be construed as limiting the present application.

As shown in fig. 2, the advertisement delivery system may include: the system comprises a parameter setting module, a significant feature extraction module, a significant feature screening module and a crowd output module.

Based on the system, the advertisement putting can be carried out according to the method shown in FIG. 3, and the method can comprise the following steps:

s1: before the advertiser performs crowd extension, the extension information is set through the advertisement putting setting interface, for example, the advertiser can set a proper seed crowd according to the popularization requirement of the advertiser, and then, parameters such as extension scale are set.

Specifically, when selecting the seed population, the seed population may be selected by setting the target population and the population filtering parameters, as shown in fig. 4. For example, the target population may include, but is not limited to: untouched users, potential users, store-in users, action users, purchase users, and the like; the crowd filter parameters may include, but are not limited to: age, sex, region, etc.

Further, the advertiser also needs to set the population expansion number, or the population expansion multiple, and the number may be set as shown in fig. 4, and the advertiser may select the expansion multiple and display the corresponding estimated number of expansion users.

S2: the advertiser sets an expansion target parameter (such as a crowd expansion target), and particularly, the server can set different attribute characteristics according to different crowd expansion targets for expanding modeling.

For example, the following population expansion targets may be preset for selection by the advertiser: brand promotion, effect promotion, conversion priority, click priority and the like. The server can select candidate crowd characteristic expression under the corresponding scene according to different crowd extension targets. Therefore, the similarity between people can have various expression dimensions, different characteristic expressions are abstracted according to the similarity characteristics of different people groups, and different people group expansion directions can be customized. Under the specific characteristic expression, the crowd expanded according to the expansion target is the crowd most similar to the seed crowd.

The features of interest to advertisers may vary under different promotional goals, such as: under the brand promotion scene, advertisers pay more attention to the preference degree of consumers to each brand. For example: if the advertiser wants to promote a certain commodity, a certain scale of brand-oriented population needs to be defined, the advertiser selects a brand promotion option in the expansion target option, and then, for the server, a feature group which is matched with a brand promotion target can be selected for the candidate population according to the selection, for example, brand preference degree features under various large brands. During the promotion period, the advertiser carries out sales promotion on a certain commodity, and selects the options of 'effect promotion' and 'conversion priority' in the expansion target option, so that the server personnel can select conversion-related characteristic expressions for candidate groups according to the selection, for example, purchasing intention group characteristics under each shop.

S3: after the advertiser selects seed person parameters (e.g., seed population), expansion parameters (e.g., population expansion size), expansion targets, and the like, the server may perform expansion modeling based on these selections of the advertiser to screen out salient features and recommend the screened salient features to the advertiser. The salient features are a plurality of matched features which are matched with the expansion direction desired by the advertiser, and the features form a salient feature group.

Among other things, methods of determining salient features may include, but are not limited to: one or more of point-to-point information index, information gain, chi-square check value and the like, and the server can display the weight value corresponding to each significant feature while outputting the significant features.

1) The point mutual information index is used for describing the correlation between each characteristic and the positive example sample;

2) the information gain is used for measuring the importance degree of a single feature by calculating the information variation brought by the model when the feature exists and the model does not exist;

3) the chi-square test value is used for expressing the association degree of each characteristic with the positive sample by calculating the deviation degree of each characteristic under the assumption that the characteristic and the positive sample are independent from each other by using a hypothesis test method.

Specifically, the seed population can be defined as a true example sample, other candidate populations serve as unmarked samples, a classification model is constructed, then the importance degree of each preset feature group in the classification model is calculated, and the importance degree of each feature is measured through the relevance degree score of a single feature and the true example sample.

Considering that the covered people have certain overlap ratio for different characteristics, in order to ensure that enough people can be found, the accumulated total number of the candidate people covered by the first K characteristics can be set to be α times of the expected expanded people size, wherein α is the overlap coefficient.

Further, because the coverage of the population is different under different feature expressions, the corresponding coincidence coefficient can be obtained through the experiment experience under the line aiming at different feature groups, so that the selected significant feature list can be ensured to cover enough extended populations, for example, α is set to be 4.5 under the brand feature group, α is set to be 5.0 under the click feature group, α is set to be 3.5 under the effect feature group, and the like, so that different α values can be set based on the difference of the selected extended targets.

The salient features are described below in connection with an uti implementation:

assuming that an advertiser sets seed population S, the population size is seed _ cnt being 10000, and defining candidate population; the advertiser sets the desired expanded population size crown _ cnt to 50000, and all other parameters are default parameters. The system selects a default crowd expression feature group, uses a point mutual information index as an importance score calculation method, and the calculation result is shown in an importance score and inherent attribute (descending order according to the importance score) comparison table of each feature in table 1:

TABLE 1

Traversing table 1 above and sequentially calculating the population size cnt covered by the previous K features in an accumulated manner until cnt > becomes crown _ cnt α (the "click first" optimization goal is selected by default, and the corresponding overlap coefficient α becomes 5.0), and when the traversal is terminated, K becomes 5, at this time, cnt becomes 332474> -round (50000 x 5.0) becomes 250000(round means rounding), so that the first 5 features can be selected as significant features to be recommended to the advertiser, and the recommendation weights (i.e., importance degree scores) corresponding to the features are shown in table 2 and fig. 5:

TABLE 2

Feature ID	Feature name	Importance score
			1	Buyer star class	0.81
2	Cosmetic preferences	0.76
			3	Amount of recent purchase	0.73
4	Preference of women's dress	0.57
			5	Age (age)	0.23

S4: the advertiser can screen out interesting features from the recommended significant feature list and adjust the features on the basis of the feature weight recommended by the system to obtain a final significant feature list.

The server displays the generated salient features and corresponding weights for the advertisers, the advertisers can interactively select interested feature lists on the basis of system recommendation, the advertisers can actively carry out weighting adjustment on the weights of the interested features, and the weights of the features which are not interested can be reduced and adjusted, even set to be zero (namely, the features are deleted).

For example, an advertiser may want to define a group of active buyers with high purchasing power to promote a high-end beauty treasure. Based on the promotion will, the advertiser may choose to down-regulate the feature "age" weight to 0, i.e., choose to delete the feature; meanwhile, the weighting adjustment is carried out on the characteristic 'the latest purchase amount' to be 1.0, and the weight reduction adjustment is carried out on the characteristic 'the suit-dress preference' to be 0.1.

By the mode, the control capability of the advertiser on the intended crowd is enhanced, so that the advertiser can flexibly customize the expansion direction of the crowd according to the popularization requirement of the advertiser, and the user experience is improved.

S5: and the server sorts and cuts off the candidate crowd according to the significant feature list selected by the advertiser and the adjusted feature weight to obtain the expanded crowd.

Specifically, the candidate population may be scored one by one according to the salient features and the corresponding weights finally selected by the advertiser, and the specific scoring method may be:

wherein aid_jRepresenting the jth user in the candidate population, f_iRepresenting the eigenvalues, w, of the corresponding users under the i-dimensional characteristics_iRepresenting the (normalized) weight of the ith dimension feature, and the feature weight normalization represents the normalization of the adjusted weight of the advertiser.

And performing descending order arrangement on the users in the candidate crowd according to the similarity scores with the seed crowd, and truncating the crowd according to the crowd expansion scale to finally generate the expanded crowd.

For example, the advertiser selects some features, such as buyer star rating, cosmetic preference, recent purchase amount, and woman dress preference, as the features of interest, and adjusts the feature weights to be 0.81, 0.76, 1.0, and 0.1, respectively, and then the server may normalize the salient feature set to be 0.30, 0.28, 0.38, and 0.04, respectively, after normalization.

Then, the total candidate population is scored based on the feature weights of the features in the normalized significant feature group, for example, the candidate weight set is U ═ { U1, U2, U3, …, un } and the feature expression thereof are shown in table 3:

TABLE 3

User' s	Buyer star class	Cosmetic preferences	Amount of recent purchase	Preference of women's dress
					u1	0.29	0.72	0.23	0.77
u2	0.32	0.83	0.34	0.32
					u3	0.21	0.29	0.87	0.63
u4	0.15	0.64	0.12	0.52
					…
un	0.30	0.41	0.92	0.12

Calculating the similarity score of each user and the seed population according to the following modes:

Score(u1)＝0.29*0.30+0.72*0.28+0.23*0.38+0.77*0.04＝0.4068

Score(u2)＝0.32*0.30+0.83*0.28+0.34*0.38+0.32*0.04＝0.4704

Score(u3)＝0.21*0.30+0.29*0.28+0.87*0.38+0.63*0.04＝0.5

Score(u4)＝0.15*0.30+0.64*0.28+0.12*0.38+0.52*0.04＝0.2906

…

Score(un)＝0.30*0.30+0.41*0.28+0.92*0.38+0.12*0.04＝0.5592

sorting candidate population U according to similarity scores to obtain …, un, …, U3, …, U2, …, U1, …, U4 and …

Since the advertiser sets the population expansion size to 50000, assuming that the user u2 scores in the global sequence of candidate population are just sorted to 50000, the final expanded population is { …, un, …, u3, …, u2 }.

In the above example, on the basis of automatically recommending the salient features, a flexible entrance is provided for the advertiser to accurately control the extension direction of the promoted crowd and enhance the user experience, and specifically, in the present example, the interaction setting of the extended crowd optimization target and the adjustment of the salient features of the recommended crowd are mainly performed in two aspects. The specific implementation manner given in the above example is only an exemplary description, and other manners may also be adopted when the implementation is actually performed, for example: the system can automatically generate a plurality of different salient feature lists, and an advertiser can freely select one of the salient feature lists as a final salient feature group and other interaction forms, such as a specific presentation form, which is not limited in the application.

Specifically, in the embodiment, the intention crowd meeting the similarity expectation with the seed crowd can be fully automatically expanded for the advertiser, meanwhile, an entrance for the advertiser to promote the intention expression is provided, the control capability of the advertiser on the expansion direction of the intention crowd is greatly enhanced, and the user experience is enhanced. Based on the method, the advertiser can selectively set different crowd expansion optimization targets, and the crowd with higher conversion/click effect is screened out from the candidate crowd, or the intention crowd with brand preference is positioned. Aiming at different expansion targets, the system can select corresponding crowd characteristic expressions so as to achieve the aim that the crowd similar to the seed crowd is the intended crowd with the popularization requirement under the corresponding characteristic expressions, so that an advertiser can set an expansion crowd optimization target and pertinently guide the system to carry out expansion crowd optimization.

On the other hand, in the salient feature list recommended by the system, the advertiser can interactively select the interested feature list, carry out weighting adjustment on the interested feature weight, and carry out weight reduction adjustment on the uninteresting feature weight until the weight is set to zero, namely, the feature is deleted. Based on the method, the advertiser can control the expansion direction of the crowd in a fine granularity mode on a characteristic level according to popularization wishes, and the control capability of the expanded crowd is greatly enhanced. Through the control on the crowd extension optimization direction, the advertiser can accurately define the intended crowd meeting the self popularization desire, so that the advertiser can accurately put the crowd in a targeted manner.

Fig. 6 is a flow chart of an embodiment of a method for advertisement delivery according to the present application. Although the present application provides method operational steps or apparatus configurations as illustrated in the following examples or figures, more or fewer operational steps or modular units may be included in the methods or apparatus based on conventional or non-inventive efforts. In the case of steps or structures which do not logically have the necessary cause and effect relationship, the execution sequence of the steps or the module structure of the apparatus is not limited to the execution sequence or the module structure described in the embodiments and shown in the drawings of the present application. When the described method or module structure is applied in an actual device or end product, the method or module structure according to the embodiments or shown in the drawings can be executed sequentially or executed in parallel (for example, in a parallel processor or multi-thread processing environment, or even in a distributed processing environment).

Specifically, as shown in fig. 6, an advertisement delivery method provided in an embodiment of the present application may include:

step 601: acquiring seed population parameters (such as seed population), expansion parameters (such as expansion quantity) and expansion target parameters (such as an expansion target and a popularization target) set by a user;

that is, an interface for setting seed population, the expansion amount, and the expansion target can be displayed on the terminal side for the advertiser to set as needed and desired. Wherein the extended target may include, but is not limited to, at least one of: brand promotion, effect promotion, click priority and conversion priority.

Step 602: determining a significant feature group according to the seed population, the expansion number and the expansion target, wherein the significant feature group comprises a plurality of significant features;

step 603: acquiring the setting operation of the user on the salient feature group;

that is, the resulting salient feature groups may be displayed on the terminal side for modification by the advertiser so that the final salient feature groups are reflective of the advertiser's intent.

Step 604: setting the salient feature group according to the setting operation;

step 605: determining similar people groups according to the set significant feature groups to obtain the extended number of popularization objects;

step 606: and advertising is put to the expanded number of promotion objects.

Specifically, the setting operation may include, but is not limited to: deletion of the features and modification of the weight values of the features, namely, the advertiser can delete the features in the significant feature group and modify the weight values of the features in the significant feature group, so that the final expanded population can meet the will of the advertiser.

When the crowd expansion is carried out, the crowd expansion can be carried out according to the following steps:

s1: taking a set of crowds covered by each feature in the set significant feature group as a candidate crowd;

s2: according to the feature value of each significant feature in the set significant feature group, carrying out relevancy sorting on the candidate population;

s3: and taking the objects with the highest correlation degree in the expansion number as promotion objects.

The acquiring of the seed population set by the user may include: acquiring target population and population characteristics set by a user; and taking the target population and the population set defined by the population characteristics as the seed population.

The target population may include, but is not limited to, at least one of: non-triggering users, potential users, action users, and purchasing users; the demographic characteristics include at least one of: sex, age, region.

In step 602, determining a salient feature group for delivering an advertisement based on the seed population parameter, the expansion parameter and the expansion target parameter set by the user may include:

s1: matching an adaptive feature group corresponding to the extended target parameter according to a corresponding relation between a pre-established target parameter and the adaptive feature group;

s2: determining the importance degree of each feature in the adaptive feature group by taking the seed population parameter as a positive sample;

s3: and taking the feature with the importance degree meeting the preset requirement as the feature in the determined significant feature group.

That is, the feature group corresponding to each extension target may be set in advance, for example: the method comprises the steps that a feature group corresponding to brand promotion, a feature group corresponding to effect promotion, a feature group corresponding to click priority and the like are obtained, after an expansion target parameter is obtained, the corresponding feature group can be matched, the feature group serves as an adaptive feature group, then importance degrees of the features are determined by seed groups, and a final significant feature group is generated according to the determined importance degrees.

Further, considering that the number of people that can be matched by the feature group is required to be ensured to meet the requirement of the expanded number under the condition of actually determining the feature group, the requirement that the number reaches the expanded number can be set as a basic criterion, that is, the feature of which the importance degree meets the preset requirement can be used as the feature in the determined significant feature group under the condition that the candidate people corresponding to the feature in the significant feature group meet the expanded number in the expanded parameters.

The method embodiments provided in the above embodiments of the present application may be executed in a mobile terminal, a computer terminal, or a similar computing device. Taking the operation on the server side as an example, fig. 7 is a hardware structure block diagram of the server side of the advertisement delivery method according to the embodiment of the present invention. As shown in fig. 7, the server 10 may include one or more processors 102 (only one is shown in the figure) (the processor 102 may include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data, and a transmission module 106 for communication functions. It will be understood by those skilled in the art that the structure shown in fig. 7 is only an illustration and is not intended to limit the structure of the electronic device. For example, the server side 10 may also include more or fewer components than shown in FIG. 7, or have a different configuration than shown in FIG. 7.

The memory 104 may be used to store software programs and modules of application software, such as program instructions/modules corresponding to the advertisement delivery method in the embodiment of the present invention, and the processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, so as to implement the above-mentioned advertisement delivery method of the application program. The memory 104 may include high speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include memory located remotely from the processor 102, which may be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The transmission module 106 is used to receive or transmit data via a network. Specific examples of the network described above may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission module 106 includes a Network adapter (NIC) that can be connected to other Network devices through a base station to communicate with the internet. In one example, the transmission module 106 may be a Radio Frequency (RF) module, which is used for communicating with the internet in a wireless manner.

In a software aspect, the advertisement delivery apparatus may include: the device comprises a determining module, an adjusting module and an expanding module, wherein:

the system comprises a determining module, a judging module and a display module, wherein the determining module is used for determining a significant feature group for delivering advertisements based on seed crowd parameters, extension parameters and extension target parameters set by a user, and the significant feature group comprises a plurality of significant features and weight values of the significant features;

an adjustment module for adjusting the salient feature group based on a user selection of the salient feature group;

and the expansion module is used for determining similar crowds according to the adjusted significant feature group to obtain a popularization object for putting the advertisement.

In one embodiment, the user selectable set of salient features includes: deletion of salient features, modification of weight values of salient features.

In one embodiment, the determining module may include: the matching unit is used for matching out the adaptive feature group corresponding to the extended target parameter according to the corresponding relation between the pre-established target parameter and the adaptive feature group; the determining unit is used for determining the importance degree of each feature in the adaptive feature group by taking the seed population parameter as a positive sample; and the generating unit is used for taking the feature with the importance degree meeting the preset requirement as the feature in the determined significant feature group.

In one embodiment, the determining the feature group as a feature of the determined salient feature group, where the importance degree meets a preset requirement, may include: and under the condition that the candidate population corresponding to the features in the significant feature group is determined to meet the expansion quantity in the expansion parameters, taking the features with the importance degrees meeting the preset requirements as the features in the determined significant feature group.

In one embodiment, the determining module may specifically determine the salient feature group for placement of the advertisement according to one of, but not limited to: point-to-point information index, information gain, chi-square test value.

In one embodiment, determining similar people groups according to the adjusted significant feature group to obtain a promotion object for advertisement delivery may include: taking a set of people covered by each feature in the adjusted significant feature group as a candidate people, wherein the adjusted significant feature group comprises: the number of expansions; according to the adjusted characteristic value of each significant characteristic in the significant characteristic group, carrying out relevancy sorting on the candidate population; and taking the objects with the highest relevance in the expansion number as promotion objects.

In one embodiment, the seed population parameters include: the target population and the population characteristics, wherein the population group defined by the target population and the population characteristics is a seed population.

In one embodiment, the target population may include at least one of: non-triggering users, potential users, action users, and purchasing users; the demographic may include at least one of: sex, age, region.

In one embodiment, the extended target parameters may include, but are not limited to, at least one of: brand promotion, effect promotion, click priority and conversion priority.

At a software level, an apparatus for determining a promotion group is provided, which may include: display module, first receiving module, confirm module and second receiving module, wherein:

the display module is used for displaying a setting interface of the seed population parameters, the expansion parameters and the expansion target parameters;

the first receiving module is used for receiving seed crowd parameters, extension parameters and extension target parameters set by a user;

the determination module is used for determining and displaying a significant feature group based on seed crowd parameters, extension parameters and extension target parameters set by a user, wherein the significant feature group comprises a plurality of significant features;

and the second receiving module is used for receiving the selection operation of the user on the significant feature group, wherein the selection operation is used for adjusting the significant feature group, and determining similar people groups based on the adjusted significant feature group to obtain the popularization object.

In one embodiment, the second receiving module may specifically display the salient feature group and the weight value of each feature in the salient feature group; receiving a selection operation of the salient feature group by a user, wherein the selection operation comprises: deletion operation of the features in the significant feature group and modification operation of the weight values.

In a software aspect, the advertisement delivery apparatus may include: the system comprises a first acquisition module, a determination module, a second acquisition module, a setting module, an expansion module and a delivery module, wherein:

the first acquisition module is used for acquiring seed crowds, expansion quantity and expansion targets set by a user;

the determining module is used for determining a significant feature group according to the seed population, the expansion number and the expansion target, wherein the significant feature group comprises a plurality of significant features;

the second acquisition module is used for acquiring the setting operation of the user on the salient feature group;

the setting module is used for setting the salient feature group according to the setting operation;

the expansion module is used for determining similar crowds according to the set significant feature group to obtain the expansion quantity of popularization objects;

and the releasing module is used for releasing advertisements to the extended number of popularization objects.

In one embodiment, the second obtaining module is specifically configured to obtain a setting operation of a user, where the setting operation includes: deletion of features, modification of weight values of features.

In one embodiment, the expansion module may be specifically configured to use a set of people covered by each feature in the set salient feature group as a candidate group; according to the feature value of each significant feature in the set significant feature group, carrying out relevancy sorting on the candidate population; and taking the objects with the highest correlation degree in the expansion number as promotion objects.

In one embodiment, the first obtaining module may specifically obtain a target population and a population characteristic set by a user; and taking the target population and the population set defined by the population characteristics as the seed population.

In one embodiment, the target population may include, but is not limited to, at least one of: non-triggering users, potential users, action users, and purchasing users; the demographic characteristics include at least one of: sex, age, region.

In one embodiment, the expansion targets may include, but are not limited to, at least one of: brand promotion, effect promotion, click priority and conversion priority.

In the software aspect, the device for determining the promoted population is located in the terminal, and may include: first display module, first receiving module, second display module and second receiving module, wherein:

the first display module is used for displaying seed crowd, expansion quantity and expansion target setting interfaces;

the first receiving module is used for receiving seed crowds, the expansion amount and the expansion target set by a user;

the second display module is used for displaying a significant feature group determined by the seed population, the expansion number and the expansion target, wherein the significant feature group comprises a plurality of significant features;

and the second receiving module is used for receiving the setting operation of the user on the significant feature group, wherein the setting operation is used for setting the significant feature group, and determining similar people groups based on the set significant feature group to obtain the expanded number of popularization objects.

In one embodiment, the second display module may display the salient feature group and the weight value of each feature in the salient feature group; accordingly, the second receiving module may receive a deletion operation and a modification operation of the weight value of the feature in the salient feature group.

This application has reached following effect: the method has the advantages that the entrance capable of setting seed groups, the number of the expansion targets and showing the feature groups is set for users with popularization requirements, so that the expansion willingness of the users can be effectively expressed, the technical problems that the expansion direction cannot be controlled by the users and the expansion results cannot meet the requirements of the users due to the fact that the existing automatic mode that the population expansion and the screening are transparent to the users is solved, the technical effect that the expansion willingness of the users can be effectively expressed is achieved, and the user experience is greatly improved.

Although the present application provides method steps as described in an embodiment or flowchart, additional or fewer steps may be included based on conventional or non-inventive efforts. The order of steps recited in the embodiments is merely one manner of performing the steps in a multitude of orders and does not represent the only order of execution. When an actual apparatus or client product executes, it may execute sequentially or in parallel (e.g., in the context of parallel processors or multi-threaded processing) according to the embodiments or methods shown in the figures.

The apparatuses or modules illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. For convenience of description, the above devices are described as being divided into various modules by functions, and are described separately. The functionality of the modules may be implemented in the same one or more software and/or hardware implementations of the present application. Of course, a module that implements a certain function may be implemented by a plurality of sub-modules or sub-units in combination.

The methods, apparatus or modules described herein may be implemented in computer readable program code to a controller implemented in any suitable manner, for example, the controller may take the form of, for example, a microprocessor or processor and a computer readable medium storing computer readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, Application Specific Integrated Circuits (ASICs), programmable logic controllers and embedded microcontrollers, examples of which include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320, the memory controller may also be implemented as part of the control logic for the memory. Those skilled in the art will also appreciate that, in addition to implementing the controller as pure computer readable program code, the same functionality can be implemented by logically programming method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Such a controller may therefore be considered as a hardware component, and the means included therein for performing the various functions may also be considered as a structure within the hardware component. Or even means for performing the functions may be regarded as being both a software module for performing the method and a structure within a hardware component.

Some of the modules in the apparatus described herein may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, classes, etc. that perform particular tasks or implement particular abstract data types. The application may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

From the above description of the embodiments, it is clear to those skilled in the art that the present application can be implemented by software plus necessary hardware. Based on such understanding, the technical solutions of the present application may be embodied in the form of software products or in the implementation process of data migration, which essentially or partially contributes to the prior art. The computer software product may be stored in a storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, mobile terminal, server, or network device, etc.) to perform the methods described in the various embodiments or portions of the embodiments of the present application.

The embodiments in the present specification are described in a progressive manner, and the same or similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. All or portions of the present application are operational with numerous general purpose or special purpose computing system environments or configurations. For example: personal computers, server computers, hand-held or portable devices, tablet-type devices, mobile communication terminals, multiprocessor systems, microprocessor-based systems, programmable electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like.

While the present application has been described with examples, those of ordinary skill in the art will appreciate that there are numerous variations and permutations of the present application without departing from the spirit of the application, and it is intended that the appended claims encompass such variations and permutations without departing from the spirit of the application.

Claims

1. An advertisement delivery method, the method comprising:

2. The method of claim 1, wherein the user selection of the salient feature group comprises: deletion of salient features, modification of weight values of salient features.

3. The method of claim 1, wherein determining the salient feature group for advertisement delivery based on the seed population parameter, the expansion parameter and the expansion target parameter set by the user comprises:

matching an adaptive feature group corresponding to the extended target parameter according to a corresponding relation between a pre-established target parameter and the adaptive feature group;

determining the importance degree of each feature in the adaptive feature group by taking the seed population parameter as a positive sample;

and taking the feature with the importance degree meeting the preset requirement as the feature in the determined significant feature group.

4. The method according to claim 3, wherein the step of using the feature with the importance degree meeting the preset requirement as the feature in the determined significant feature group comprises the following steps:

and under the condition that the candidate population corresponding to the features in the significant feature group is determined to meet the expansion quantity in the expansion parameters, taking the features with the importance degrees meeting the preset requirements as the features in the determined significant feature group.

5. The method of claim 1, wherein determining the salient feature group for advertisement delivery based on the seed population parameter, the expansion parameter and the expansion target parameter set by the user comprises:

determining a set of salient features for placement of an advertisement in at least one of: point-to-point information index, information gain, chi-square test value.

6. The method of claim 2, wherein determining similar groups of people from the adjusted salient feature groups to obtain promotion objects for advertisement delivery comprises:

taking a set of people covered by each feature in the adjusted significant feature group as a candidate people, wherein the adjusted significant feature group comprises: the number of expansions;

according to the adjusted characteristic value of each significant characteristic in the significant characteristic group, carrying out relevancy sorting on the candidate population;

and taking the objects with the highest relevance in the expansion number as promotion objects.

7. The method of claim 1, wherein the seed population parameters comprise: the target population and the population characteristics, wherein the population group defined by the target population and the population characteristics is a seed population.

8. The method of claim 7, wherein the target population comprises at least one of: non-triggering users, potential users, action users, and purchasing users; the demographic characteristics include at least one of: sex, age, region.

9. The method of claim 1, wherein the extended objective parameters include at least one of: brand promotion, effect promotion, click priority and conversion priority.

10. A method for determining a population to promote, the method comprising:

11. The method of claim 10, wherein receiving a user selection of the salient feature group comprises:

displaying the salient feature group and the weight value of each feature in the salient feature group;

receiving a selection operation of the salient feature group by a user, wherein the selection operation comprises: deletion operation of the features in the significant feature group and modification operation of the weight values.

12. A server comprising a processor and a memory for storing processor-executable instructions which, when executed by the processor, implement the steps of the method of any one of claims 1 to 9.

13. A client comprising a processor and a memory for storing processor-executable instructions which, when executed by the processor, implement the steps of the method of any one of claims 10 to 11.

14. A computer readable storage medium having stored thereon computer instructions which, when executed, implement the steps of the method of any one of claims 1 to 9.