CN115841351A - Method and device for determining recommended advertisements and electronic equipment - Google Patents
Method and device for determining recommended advertisements and electronic equipment Download PDFInfo
- Publication number
- CN115841351A CN115841351A CN202211571295.9A CN202211571295A CN115841351A CN 115841351 A CN115841351 A CN 115841351A CN 202211571295 A CN202211571295 A CN 202211571295A CN 115841351 A CN115841351 A CN 115841351A
- Authority
- CN
- China
- Prior art keywords
- sample
- advertisement
- user
- behavior index
- current
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
The application discloses a method, a device and electronic equipment for determining recommended advertisements, which relate to the technical field of big data intelligent analysis, and the method comprises the following steps: acquiring behavior index data of a plurality of sample users on each sample advertisement; determining the value of credit of each sample user on each advertisement category respectively according to the behavior index data, and forming a credit vector corresponding to each sample user according to the value of credit of each sample user on each advertisement category; representing the sample users by the grading vectors corresponding to the sample users, and clustering the plurality of sample users by adopting a clustering algorithm to obtain a plurality of user groups; acquiring behavior index data of a target user on each historical advertisement; determining a target user group which is most similar to a target user in a plurality of user groups according to the behavior index data of the target user; and determining advertisements recommended for the target users according to the advertisement sets which are interested by the users in the target user group. According to the scheme, the advertisements recommended to the target user are determined according to the preference of the sample user, and the accuracy of advertisement recommendation is improved.
Description
Technical Field
The application relates to the technical field of big data intelligent analysis, in particular to a method and a device for determining recommended advertisements and electronic equipment.
Background
With the rapid development of the current internet era, user information and behavior data also show a rapid growth trend, and the internet technology has become an important part in the life of people. Recommending personalized advertisements for a user according to the user's personal preferences generally results in a better service experience for the user.
Currently, the determination of personalized advertisements recommended for users generally includes obtaining personal basic information (such as sex, age, interested subject, and the like) and behavior data information of a target user, determining a user portrait according to the personal basic information and the behavior data information, and then taking a preset advertisement corresponding to the user portrait as an advertisement recommended for the target user.
However, in this method, if the preset advertisement corresponding to the user image is inaccurate or the method for determining the user image is inaccurate, the recommended advertisement is inaccurate, and the experience of the target user is affected.
Disclosure of Invention
The application aims to provide a method and a device for determining recommended advertisements and electronic equipment, so as to solve the problem that the existing method for determining recommended advertisements is inaccurate.
In order to solve the above technical problem, a first aspect of the present specification provides a method for determining a recommended advertisement, including: acquiring behavior index data of a plurality of sample users for each sample advertisement when browsing the advertisement; determining the value of credit of each sample user on each advertisement category according to the behavior index data, and forming a credit vector corresponding to each sample user according to the value of credit of each sample user on each advertisement category; representing the sample users by the grading vectors corresponding to the sample users, and clustering the plurality of sample users by adopting a clustering algorithm to obtain a plurality of user groups; acquiring behavior index data of each historical advertisement when a target user browses the historical advertisements; determining a target user group which is most similar to a target user in the plurality of user groups according to the behavior index data of the target user; and determining the advertisements recommended for the target users according to the advertisement set which is interested by each user in the target user group.
In some embodiments, before determining, according to the behavior index data, a score vector for each advertisement category for each sample user, the method further comprises: acquiring a set of sample advertisements browsed by the plurality of sample users; and clustering the sample advertisements in the set to obtain a plurality of advertisement categories.
In some embodiments, determining, from the behavior index data, respective values of ratings of the sample users for the respective advertisement categories includes: calculating the value of the credit of the current sample user to the current advertisement category according to the following method: and calculating the score value of the current sample user on the current advertisement category according to the behavior index data of the current sample user on all sample advertisements under the current advertisement category.
In some embodiments, calculating the value of the score of the current sample user for the current advertisement category according to the behavior index data of the current sample user for all sample advertisements under the current advertisement category comprises: calculating the value of the credit of the current sample user to each sample advertisement by the following method: calculating the value of the score of the current sample user on the current sample advertisement according to various behavior index data of the current sample user on the current sample advertisement; and calculating the grade value of the current sample user to the current advertisement category based on the grade value of each sample advertisement under the current category.
In some embodiments, calculating the value of the score of the current sample user for the current advertisement category according to the behavior index data of the current sample user for all sample advertisements under the current advertisement category comprises: respectively calculating the comprehensive values of each behavior index data corresponding to the current category: calculating a comprehensive value of the current behavior index data according to the current behavior index data of each sample advertisement under the current category; obtaining weights corresponding to various behavior index data respectively; and carrying out weighted summation on the comprehensive values of various behavior index data to obtain the value of the credit of the current sample user on the current advertisement category.
In some embodiments, clustering the plurality of sample users using a clustering algorithm to obtain a plurality of user groups includes: randomly initializing n points from all sample users as initial clustering centers; the following operations are executed in a loop until a preset cut-off condition is reached: calculating the distance between each sample user and each clustering center, determining the clustering center with the minimum distance from the sample user, and classifying the sample users into a class represented by the distance center; and re-determining the clustering center of each category according to all the updated sample users in each category.
In some embodiments, determining a target user group, which is closest to the target user, in the plurality of user groups according to the behavior index data of the target user includes: representing the target user by the behavior index data of the target user, and calculating the distance between the target user and each user group; and taking the user group with the minimum distance as a target user group which is closest to the target user.
In some embodiments, said: acquiring behavior index data of a plurality of sample users for each sample advertisement when browsing the advertisement; determining the value of credit of each sample user on each advertisement category respectively according to the behavior index data, and forming a credit vector corresponding to each sample user according to the value of credit of each sample user on each advertisement category respectively; representing the sample users by the grading vectors corresponding to the sample users, and clustering the plurality of sample users by adopting a clustering algorithm to obtain a plurality of user groups; when the data of the sample user reaches a preset threshold value or the number of the sample advertisements browsed by the sample user reaches the preset threshold value, the sample advertisement behavior index data with the earliest time is cleared, so that the number of the sample advertisements is in a preset numerical value range.
In some embodiments, determining the advertisement recommended for the target user according to the advertisement set in which the users in the target user group are interested includes: and taking the M sample advertisements with the highest score values in the target user group as advertisements recommended for the target users, wherein M is a preset natural number.
In some embodiments, determining the advertisement recommended for the target user according to the advertisement set in which the users in the target user group are interested includes: determining N sample users closest to the target user in the target user group, wherein N is a preset natural number; and determining the advertisements recommended for the target users according to the historical recommendation results of the N sample users.
In some embodiments, determining the advertisements recommended for the target user according to the advertisement sets in which the users in the target user group are interested includes: determining M sample advertisements with the highest scoring values in a target user group as a first advertisement set, wherein M is a preset natural number; acquiring actual behavior index data corresponding to each advertisement in a first advertisement set; determining N sample users closest to the target user in the target user group, and taking the historical recommendation results of the N sample users as a second advertisement set; acquiring behavior index prediction data corresponding to each advertisement in the second advertisement set; acquiring an intersection of the first advertisement set and the second advertisement set; according to the first behavior index and the behavior index prediction data, calculating behavior index prediction data of the target advertisement in the intersection; and according to the behavior index prediction data, selecting advertisements with target advertisement quantity from the intersection as target advertisements recommended for target users.
A second aspect of the present specification provides an apparatus for determining recommended advertisements, including: the first acquisition unit is used for acquiring behavior index data of each sample advertisement when a plurality of sample users browse the advertisements; the first determining unit is used for determining the scores of the sample users for the advertisement categories respectively according to the behavior index data, and forming score vectors corresponding to the sample users according to the scores of the sample users for the advertisement categories; the user clustering unit is used for representing the sample users by the grading vectors corresponding to the sample users, and clustering the plurality of sample users by adopting a clustering algorithm to obtain a plurality of user groups; the second acquisition unit is used for acquiring behavior index data of each historical advertisement when the target user browses the historical advertisements; a second determining unit, configured to determine, according to behavior index data of a target user, a target user group that is closest to the target user among the plurality of user groups; and the third determining unit is used for determining the advertisements recommended for the target users according to the advertisement sets which are interested by the users in the target user group.
In some embodiments, the apparatus further comprises: a third obtaining unit, configured to obtain a set of sample advertisements browsed by the plurality of sample users; and the advertisement clustering unit is used for clustering the sample advertisements in the set to obtain a plurality of advertisement categories.
In some embodiments, the first determination unit comprises: a first calculating subunit, configured to calculate a value of the score of the current sample user for the current advertisement category according to the following method: calculating the score value of the current sample user on the current advertisement category according to the behavior index data of the current sample user on all sample advertisements under the current advertisement category; and the construction subunit is used for forming a scoring vector corresponding to the sample user according to the scoring value of the sample user to each advertisement category.
In some embodiments, the first computing subunit comprises: the second calculating subunit is used for calculating the scoring value of the current sample user on each sample advertisement by the following method: calculating the value of the score of the current sample user on the current sample advertisement according to various behavior index data of the current sample user on the current sample advertisement; and the third calculating subunit is used for calculating the score value of the current sample user to the current advertisement category based on the score values of all sample advertisements in the current category.
In some embodiments, the first computing subunit comprises: a third calculating subunit, configured to calculate, respectively, a comprehensive value of each behavior index data corresponding to the current category: calculating a comprehensive value of the current behavior index data according to the current behavior index data of each sample advertisement under the current category; the first acquiring subunit is used for acquiring weights corresponding to various behavior index data respectively; and the summation subunit is used for carrying out weighted summation on the comprehensive values of various behavior index data to obtain the value of the credit of the current sample user on the current advertisement category.
In some embodiments, the user clustering unit comprises: the initialization subunit is used for randomly initializing n points from all sample users as initial clustering centers; the fourth calculating subunit and the updating subunit are used for circularly executing the operation until a preset cut-off condition is reached, wherein the fourth calculating subunit is used for calculating the distance between each sample user and each clustering center, determining the clustering center with the minimum distance from the sample user, and classifying the sample users into a class represented by the distance center; and the updating subunit is used for re-determining the clustering center of each category according to all the updated sample users in each category.
In some embodiments, the second determination unit comprises: the fifth calculating subunit is used for representing the target user by the behavior index data of the target user and calculating the distance between the target user and each user group; and the first determining subunit is used for taking the user group with the minimum distance as a target user group closest to the target user.
In some embodiments, said: acquiring behavior index data of a plurality of sample users for each sample advertisement when browsing the advertisement; determining the value of credit of each sample user on each advertisement category respectively according to the behavior index data, and forming a credit vector corresponding to each sample user according to the value of credit of each sample user on each advertisement category respectively; representing the sample users by the grading vectors corresponding to the sample users, and clustering the plurality of sample users by adopting a clustering algorithm to obtain a plurality of user groups; when the data of the sample user reaches a preset threshold value or the number of the sample advertisements browsed by the sample user reaches the preset threshold value, the sample advertisement behavior index data with the earliest time is cleared, so that the number of the sample advertisements is in a preset numerical value range.
In some embodiments, the third determining unit includes: the second determining subunit is used for determining M sample advertisements with the highest score values in the target user group as a first advertisement set, wherein M is a preset natural number; the second obtaining subunit is used for obtaining actual behavior index data corresponding to each advertisement in the first advertisement set; the third determining subunit is used for determining N sample users which are closest to the target user in the target user group, and taking the historical recommendation results of the N sample users as a second advertisement set; the third obtaining subunit is configured to obtain behavior index prediction data corresponding to each advertisement in the second advertisement set; a fourth obtaining subunit, configured to obtain an intersection of the first advertisement set and the second advertisement set; the sixth calculating subunit is used for calculating behavior index prediction data of the target advertisement in the intersection according to the first behavior index and the behavior index prediction data; and the screening subunit is used for selecting advertisements with target advertisement quantity from the intersection as the target advertisements recommended for the target users according to the behavior index prediction data.
A third aspect of the present specification provides an electronic apparatus comprising: a memory and a processor, the processor and the memory being communicatively connected to each other, the memory having stored therein computer instructions, the processor implementing the steps of the method of any one of the first aspect by executing the computer instructions.
A fourth aspect of the present description provides a computer storage medium storing computer program instructions which, when executed by a processor, implement the steps of the method of any one of the first aspects.
A fifth aspect of the present description provides a computer program product comprising a computer program which, when executed by a processor, performs the steps of the method of any one of the first aspects.
According to the method, the device and the electronic product for determining the recommended advertisements, the sample users are clustered according to the behavior index data of the sample users on the sample advertisements to obtain a plurality of user groups, the target user group closest to the target users is determined, then the advertisements recommended to the target users are determined according to the advertisement set which is interested by each user in the target user group, namely the advertisements recommended to the target users are determined according to the preference of the sample users, the portrait of the target users does not need to be determined, and therefore the problem that the recommended advertisements are inaccurate due to the fact that the relevant content of the portrait of the users is inaccurate is solved; the behavior index data of the sample advertisement of the sample user is adopted to cluster the sample user to obtain a plurality of user groups, namely the user groups are divided according to the behaviors of the user, the user groups are not divided according to the static index data (such as age, checking preference and the like) of the user, the division of the user groups is more accurate, and the advertisement recommended to the target user is determined according to the preference of the user group closest to the target user, so that the advertisement recommended to the target user is more accurate. Therefore, the method and the device improve the accuracy of recommending the advertisement.
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 creative efforts.
FIG. 1 is a flow chart illustrating a method of determining recommended advertisements provided by the present specification;
FIG. 2 is a flow chart illustrating another method of determining recommended advertisements provided by the present specification;
FIG. 3 illustrates a flowchart of one method for determining advertisements to recommend to a target user based on a set of advertisements in which each user in a target user group is interested;
FIG. 4 is a schematic block diagram of a device for determining recommended advertisements provided by the present specification;
fig. 5 illustrates a functional block diagram of an electronic device provided by the present specification.
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 obtained by a person of ordinary skill in the art without any inventive work based on the embodiments in the present application shall fall within the scope of protection of the present application.
According to the technical scheme, the data acquisition, storage, use, processing and the like meet relevant regulations of national laws and regulations.
The present specification provides a method for determining recommended advertisements, as shown in fig. 1, including the following steps:
s10: behavior index data of a plurality of sample users for each sample advertisement when browsing the advertisement is obtained.
The sample advertisement may be some advertisement delivered in advance. The purpose of the advertisement putting can be used for acquiring the behavior index data of the user on the advertisement besides the function of the advertisement. The sample ads served are diverse.
The behavior index data is a plurality of index data of behaviors of the user when the user views the advertisement. For example, for an advertisement presented on an electronic medium, the plurality of metric data may include a plurality of the following metric data: whether the advertisement link will be clicked, the browsing duration for the advertisement publicity page, the browsing duration for the content pointed by the advertisement link, whether the advertisement link will be forwarded, whether the advertisement will be selected to be shielded, etc.; for a physical advertising medium such as a billboard, the plurality of metric data may include a plurality of the following metric data: watching duration, expressions after watching, whether the advertisement and the peer will be talked about after watching, whether the photo will be taken, and the like; for a voice advertisement for voice broadcast, the plurality of indicator data may include a plurality of the following indicator data: whether to listen to the advertisement content, the expression after listening to the advertisement, whether to talk about the advertisement and talk about the time with a peer after listening to the advertisement, whether to imitate the advertisement words, etc.
Acquiring behavior index data, wherein for the advertisement displayed on the electronic medium, a buried point is preset at a proper position in a background program for displaying the advertisement, and the behavior index data is acquired through the buried point; for entity advertisement media and voice advertisement media such as billboards, cameras and voice collectors can be arranged on the periphery, and behavior index data can be determined through collected user videos and voices.
S20: and determining the value of the grade of each advertisement category of each sample user according to the behavior index data, and forming a grade vector corresponding to each sample user according to the value of the grade of each advertisement category of each sample user.
Before step S20, the category identification of each sample advertisement may be obtained.
In some embodiments, the sample advertisements may be divided according to a predefined division standard before placement, and then the category identification of each advertisement may be directly obtained.
In some embodiments, the advertisement category may be a result of performing a cluster classification algorithm according to at least one of keywords and/or sentences in the advertisement words, key frames, key content in pictures, advertisement presentation style (e.g., ink and wash style, comic style, vintage style, etc.), and the like.
For example, a set of sample advertisements viewed by the plurality of sample users is obtained; and clustering the sample advertisements in the set to obtain a plurality of advertisement categories. The method has the advantages that only the sample advertisements browsed by the sample users are clustered through a clustering mode to obtain the advertisement categories, the categories to which the advertisements belong can be adjusted according to the preference of the users, and therefore the advertisements recommended finally are changed along with the variation trend of the popular preferences.
The score values represent the user's attention to each advertisement category.
For example, if the sample users include a, B, and C, the sample advertisement categories include X, Y, and Z, and the attention of each sample user to each sample advertisement category is shown in table one below, the score vector may be expressed as:
wherein a1, b1, c1, a2, b2, c2, a3, b3, c3, etc. represent the score values.
Watch 1
| X | Y | Z | |
| A | a1 | b1 | c1 |
| B | a2 | b2 | c2 |
| C | a3 | b3 | c3 |
The method for determining recommended advertisements is based on data driving, the quality of collected data determines the performance of the whole system, and in order to ensure the accuracy and reliability of subsequent calculation results, a formula can be adopted firstly
Standardizing the data, and countingNormalizing the features by scaling to unit variance, wherein x represents the behavior index data before normalization, x * And represents the normalized behavior index data, μ represents the average value of the same behavior index data, and σ represents the variance of the same behavior index data, where the same behavior index data is the same behavior index data as the behavior index data x before normalization.
In some embodiments, as shown in fig. 2, S20 may include S21: calculating the value of the credit of the current sample user to the current advertisement category according to the following method: and calculating the value of the current sample user to the current advertisement category according to the behavior index data of the current sample user to all sample advertisements under the current advertisement category.
Taking table one as an example, under the condition that the current sample user is a and the current sample advertisement category is X, the score value of the sample user a to the sample advertisement category X is calculated according to the behavior index data of the sample user a to all sample advertisements under the category X.
Watch two
| Index 1 | Index 2 | Index 3 | Index 4 | |
| A | p1 | p2 | p3 | p4 |
| A | q1 | q2 | q3 | q4 |
| A | w1 | w2 | w3 | w4 |
In some embodiments, S21 may calculate the score value of each sample advertisement for the current sample user through the following steps S211 and S212:
s211: and calculating the value of the current sample advertisement to be scored by the current sample user according to various behavior index data of the current sample advertisement to be scored by the current sample user.
S212: and calculating the value of the credit of the current sample user to the current advertisement category based on the value of the credit of each sample advertisement under the current category.
The data in table two corresponds to a-X (i.e., the second row and the second column of data) in table one, the behavior index data has 4 types, p1, q1, w1, p2, q2, w2, p3, q3, w3, etc. represent values of corresponding indexes, and one row in table two represents data corresponding to one sample advertisement.
Taking table two as an example, S211 is to calculate a score value 1 according to p1, p2, p3, and p4, calculate a score value 2 according to q1, q2, q3, and q4, calculate a score value 3 according to w1, w2, w3, and w4, and S212 is to calculate a score value of the sample user a for the current advertisement category according to the score value 1, the score value 2, and the score value 3. For example, a weighted average of the score value 1, the score value 2, and the score value 3 is obtained as the score value for the advertisement category.
In some embodiments, S21 may calculate the integrated value of each behavior index data corresponding to the current category in steps S213 and S214 as follows:
s213: and calculating the comprehensive value of the current behavior index data according to the current behavior index data of each sample advertisement under the current category.
S214: and acquiring weights corresponding to various behavior index data respectively.
S215: and carrying out weighted summation on the comprehensive values of the various behavior index data to obtain the score value of the current sample user on the current advertisement category.
Taking table two as an example, a comprehensive value 1 may be obtained by calculation according to p1, q1, and w1, a comprehensive value 2 may be obtained by calculation according to p2, q2, and w2, and a comprehensive value 3 may be obtained by calculation according to p3, q3, and w3, and S214 is to calculate a score value of the sample user a for the current advertisement category according to the comprehensive value 1, the comprehensive value 2, and the comprehensive value 3.
S30: and clustering the sample users by adopting a clustering algorithm by taking the grading vector as a representative sample user to obtain a plurality of user groups.
For example, n points are initialized randomly from all sample users as initial clustering centers; then, the following operations S31 and S32 are cyclically performed until a preset cutoff condition is reached:
s31: and calculating the distance between each sample user and each clustering center, determining the clustering center with the minimum distance from the sample user, and classifying the sample users into the class represented by the distance center.
S32: and re-determining the clustering center of each category according to all the updated sample users in each category.
The sample users are clustered through a clustering algorithm, and the clustering basis is a scoring vector which is the scoring value of the sample users for each sample advertisement category, so that the sample users in the same category in the clustering result have more commonality in the interested advertisements, and the accuracy of recommending the advertisements for the target users according to the clustering result is higher.
S40: behavior index data of the target user on each historical advertisement when browsing the historical advertisements is obtained.
The historical advertisement acquired in S40 may be a historical advertisement viewed within a predetermined time span, for example, within one month.
The historical advertisement may be a sample advertisement that has been delivered, or may be a newly delivered advertisement that is different from the sample advertisement.
S50: and determining a target user group which is closest to the target user in the plurality of user groups according to the behavior index data of the target user.
S50, representing the target user by the behavior index data of the target user, and calculating the distance between the target user and each user group; and taking the user group with the minimum distance as a target user group which is closest to the target user.
When the distance between the target user and each user group is calculated, the distance between the target user and the group center of the user group can be used as the distance between the target user and the user group; or taking the minimum distance value in the distances between the target user and each user in the user group as the distance between the target user and the user group; the average value of the minimum distance value and the maximum distance value among the distances between the target user and each user in the user group may be used as the distance between the target user and the user group.
S60: and determining the advertisements recommended for the target users according to the advertisement set which is interested by each user in the target user group.
After the advertisements targeted for recommendation are determined, they may be presented to the user in a retention advertisement or carousel.
In some embodiments, S60 may use M sample advertisements with the highest score values in the target user group as the advertisements recommended for the target users, where M is a preset natural number. The above calculation method can be referred to as the score value here.
In some embodiments, S60 may first determine N sample users closest to the target user in the target user group, where N is a preset natural number; and then determining the advertisements recommended for the target users according to the historical recommendation results of the N sample users. For example, the historical recommendation results of the N sample users may be directly used as the advertisements recommended for the target user, or a part of the recommendation results may be further filtered out from the historical recommendation results of the N sample users as the advertisements recommended for the target user.
In some embodiments, S60 may also determine the recommended advertisement for the target user in combination with the advertisement most preferred by the target user group as a whole and the historical recommendation results of the sample users closest to the target user. Accordingly, as shown in FIG. 3, S60 may include the following steps S61-S67.
S61: and taking M sample advertisements with the highest scoring values in the target user group as a first advertisement set, wherein M is a preset natural number.
S62: and acquiring actual behavior index data corresponding to each advertisement in the first advertisement set.
The actual behavior index data herein refers to actually collected behavior index data, and is not predicted.
S63: and determining N sample users closest to the target user in the target user group, and taking the historical recommendation results of the N sample users as a second advertisement set.
S64: and acquiring behavior index prediction data corresponding to each advertisement in the second advertisement set.
When determining the advertisements recommended for the user, behavior index data of the user for each advertisement is calculated according to actually collected data, namely prediction data, but not actually collected data. The behavior index prediction data acquired in S64 is prediction data, not actually acquired data.
S65: and acquiring the intersection of the first advertisement set and the second advertisement set.
S66: and calculating the behavior index prediction data of the target advertisement in the intersection according to the actual behavior index data and the behavior index prediction data.
For example, the actual behavior index data may be predetermined to be weighted as η 1 The weight corresponding to the behavior index prediction data is eta 2 Then, the behavior index prediction data of the target advertisement in the intersection is: d = D 1 ×η 1 +D 2 ×η 2 Wherein D is 1 As actual behavior index data, D 2 Data is predicted for the behavior index.
S67: and according to the behavior index prediction data, selecting advertisements with target advertisement quantity from the intersection as target advertisements recommended for target users.
In some embodiments, S10, S20, and S30 described above are performed every predetermined span of time. That is, the user group is updated every predetermined time span so that the preference of the user group is changed following the current preference of the public.
When the data of the sample user reaches a preset threshold value or the number of the sample advertisements browsed by the sample user reaches the preset threshold value, the sample advertisement behavior index data with the earliest time is cleared, so that the number of the sample advertisements is in a preset numerical value range.
According to the method, the device and the electronic product for determining the recommended advertisements, the sample users are clustered according to the behavior index data of the sample users on the sample advertisements to obtain a plurality of user groups, the target user group closest to the target users is determined, then the advertisements recommended to the target users are determined according to the advertisement set which is interested by each user in the target user group, namely the advertisements recommended to the target users are determined according to the preference of the sample users, the portrait of the target users does not need to be determined, and therefore the problem that the recommended advertisements are inaccurate due to the fact that the relevant content of the portrait of the users is inaccurate is solved; the behavior index data of the sample advertisement of the sample user is adopted to cluster the sample user to obtain a plurality of user groups, namely the user groups are divided according to the behaviors of the user, the user groups are not divided according to the static index data (such as age, checking preference and the like) of the user, the division of the user groups is more accurate, and the advertisement recommended to the target user is determined according to the preference of the user group closest to the target user, so that the advertisement recommended to the target user is more accurate. Therefore, the advertisement recommending method and the advertisement recommending device improve the accuracy of advertisement recommending.
The method comprises the steps of obtaining user basic information, transaction data, product data and user behavior data on an application platform, collecting the user basic information, analyzing user preference by adopting a clustering algorithm, discovering personalized requirements and interest characteristics of a user by mining user behaviors, displaying advertisement positions and links which are possibly interested by the user to the user, and having strong pertinence. Meanwhile, the mobile phone bank is assisted in creating the digital payment capacity, the user flow is redirected to the bank, and the advertising income is maximized.
The present specification provides a device for determining recommended advertisements, which can be used to implement the method shown in fig. 1. As shown in fig. 4, the apparatus includes a first acquisition unit 10, a first determination unit 20, a user clustering unit 30, a second acquisition unit 40, a second determination unit 50, and a third determination unit 60.
The first acquiring unit 10 is configured to acquire behavior index data of a plurality of sample users for each sample advertisement when browsing the advertisement.
The first determining unit 20 is configured to determine, according to the behavior index data, the score values of the sample users for the advertisement categories, and form, by using the score values of the sample users for the advertisement categories, a score vector corresponding to the sample users.
The user clustering unit 30 is configured to represent the sample users with the scoring vectors corresponding to the sample users, and cluster the multiple sample users by using a clustering algorithm to obtain multiple user groups.
The second obtaining unit 40 is configured to obtain behavior index data of each history advertisement when the target user browses the history advertisement.
The second determining unit 50 is configured to determine, according to the behavior index data of the target user, a target user group that is closest to the target user in the plurality of user groups.
The third determining unit 60 is configured to determine the advertisement recommended for the target user according to the advertisement sets interested by the users in the target user group.
In some embodiments, the apparatus further comprises: a third obtaining unit, configured to obtain a set of sample advertisements browsed by the plurality of sample users; and the advertisement clustering unit is used for clustering the sample advertisements in the set to obtain a plurality of advertisement categories.
In some embodiments, the first determination unit comprises: a first calculating subunit, configured to calculate a score value of a current sample user for a current advertisement category according to the following method: calculating the value of the grade of the current sample user to the current advertisement category according to the behavior index data of the current sample user to all sample advertisements under the current advertisement category; and the construction subunit is used for forming a grading vector corresponding to the sample user according to the grading value of the sample user to each advertisement category.
In some embodiments, the first computing subunit comprises: the second calculating subunit is used for calculating the scoring value of the current sample user on each sample advertisement by the following method: calculating the score value of the current sample user on the current sample advertisement according to various behavior index data of the current sample user on the current sample advertisement; and the third calculating subunit is used for calculating the score value of the current sample user to the current advertisement category based on the score values of all sample advertisements in the current category.
In some embodiments, the first computing subunit comprises: a third calculating subunit, configured to calculate, respectively, a comprehensive value of each behavior index data corresponding to the current category: calculating a comprehensive value of the current behavior index data according to the current behavior index data of each sample advertisement under the current category; the first acquiring subunit is used for acquiring weights corresponding to various behavior index data respectively; and the summation subunit is used for carrying out weighted summation on the comprehensive values of the various behavior index data to obtain the score value of the current sample user on the current advertisement category.
In some embodiments, the user clustering unit comprises: the initialization subunit is used for randomly initializing n points from all sample users as initial clustering centers; the fourth calculating subunit and the updating subunit are used for circularly executing the operation until a preset cut-off condition is reached, wherein the fourth calculating subunit is used for calculating the distance between each sample user and each clustering center, determining the clustering center with the minimum distance from the sample user, and classifying the sample users into a class represented by the distance center; and the updating subunit is used for re-determining the clustering center of each category according to all the updated sample users in each category.
In some embodiments, the second determination unit comprises: the fifth calculating subunit is used for representing the target user by the behavior index data of the target user and calculating the distance between the target user and each user group; and the first determining subunit is used for taking the user group with the minimum distance as a target user group closest to the target user.
In some embodiments, said: acquiring behavior index data of a plurality of sample users for each sample advertisement when browsing the advertisement; determining the value of credit of each sample user on each advertisement category respectively according to the behavior index data, and forming a credit vector corresponding to each sample user according to the value of credit of each sample user on each advertisement category respectively; representing the sample users by the grading vectors corresponding to the sample users, and clustering the plurality of sample users by adopting a clustering algorithm to obtain a plurality of user groups; when the data of the sample user reaches a preset threshold value or the number of the sample advertisements browsed by the sample user reaches the preset threshold value, the sample advertisement behavior index data with the earliest time is cleared, so that the number of the sample advertisements is in a preset numerical value range.
In some embodiments, the third determination unit comprises: the second determining subunit is used for determining M sample advertisements with the highest score values in the target user group as a first advertisement set, wherein M is a preset natural number; the second obtaining subunit is used for obtaining actual behavior index data corresponding to each advertisement in the first advertisement set; the third determining subunit is used for determining N sample users which are closest to the target user in the target user group, and taking the historical recommendation results of the N sample users as a second advertisement set; the third obtaining subunit is configured to obtain behavior index prediction data corresponding to each advertisement in the second advertisement set; a fourth obtaining subunit, configured to obtain an intersection of the first advertisement set and the second advertisement set; the sixth calculating subunit is used for calculating behavior index prediction data of the target advertisement in the intersection according to the first behavior index and the behavior index prediction data; and the screening subunit is used for selecting advertisements with target advertisement quantity from the intersection as the target advertisements recommended for the target users according to the behavior index prediction data.
The description and the beneficial effects of the device for determining recommended advertisements may refer to the description and the beneficial effects of the method part, and are not described again.
An embodiment of the present invention further provides an electronic device, as shown in fig. 5, the electronic device may include a processor 501 and a memory 502, where the processor 501 and the memory 502 may be connected by a bus or in another manner, and fig. 5 takes the connection by the bus as an example.
The memory 502, which is a non-transitory computer-readable storage medium, may be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as program instructions/modules corresponding to the determination method of the recommended advertisement in the embodiment of the present invention (for example, the first obtaining unit 10, the first determining unit 20, the user clustering unit 30, the second obtaining unit 40, the second determining unit 50, and the third determining unit 60 shown in fig. 4). The processor 501 executes various functional applications and data classification of the processor by running non-transitory software programs, instructions and modules stored in the memory 502, namely, implementing the method for determining recommended advertisements in the above method embodiments.
The memory 502 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created by the processor 501, and the like. Further, the memory 502 may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, memory 502 optionally includes memory located remotely from processor 501, which may be connected to processor 501 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 one or more modules are stored in the memory 502 and, when executed by the processor 501, perform the method of determining recommended advertisements described above.
The details of the electronic device can be understood by referring to the relevant description and effects in the above embodiments, and are not described herein again.
The present specification provides a computer storage medium having stored thereon computer program instructions which, when executed by a processor, implement the steps of the method of determining recommended advertisements described above.
The present specification provides a computer program product comprising a computer program which, when executed by a processor, performs the steps of the method of determining recommended advertisements described above.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic Disk, an optical Disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a Flash Memory (Flash Memory), a Hard Disk (Hard Disk Drive, abbreviated as HDD), a Solid State Drive (SSD), or the like; the storage medium may also comprise a combination of memories of the kind described above.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments can be referred to each other, and each embodiment focuses on the differences from the other embodiments.
The systems, devices, modules or units described 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 units by function, and are described separately. Of course, the functionality of the units may be implemented in one or more software and/or hardware when implementing the present application.
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 a necessary general hardware platform. Based on such understanding, the technical solutions of the present application may be essentially or partially implemented in the form of software products, which may be stored in a storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and include instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods of some parts of the embodiments of the present application.
The application is 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, multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like.
The application 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, 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.
Although the present application has been described in terms of embodiments, those of ordinary skill in the art will recognize 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 (20)
1. A method for determining recommended advertisements, comprising:
acquiring behavior index data of a plurality of sample users on each sample advertisement when browsing the advertisement;
determining the value of credit of each sample user on each advertisement category according to the behavior index data, and forming a credit vector corresponding to each sample user according to the value of credit of each sample user on each advertisement category;
representing the sample users by the grading vectors corresponding to the sample users, and clustering the plurality of sample users by adopting a clustering algorithm to obtain a plurality of user groups;
acquiring behavior index data of each historical advertisement when a target user browses the historical advertisements;
determining a target user group which is most similar to a target user in the plurality of user groups according to the behavior index data of the target user;
and determining advertisements recommended for the target users according to the advertisement sets which are interested by the users in the target user group.
2. The method of claim 1, further comprising, prior to determining a score vector for each advertisement category for each sample user based on the behavior index data:
acquiring a set of sample advertisements browsed by the plurality of sample users;
and clustering the sample advertisements in the set to obtain a plurality of advertisement categories.
3. The method of claim 1, wherein determining a respective value of credit to each advertisement category for each sample user based on the behavior index data comprises:
calculating the value of the credit of the current sample user to the current advertisement category according to the following method: and calculating the score value of the current sample user on the current advertisement category according to the behavior index data of the current sample user on all sample advertisements under the current advertisement category.
4. The method of claim 3, wherein calculating the value of the score of the current sample user for the current advertisement category according to the behavior index data of the current sample user for all sample advertisements in the current advertisement category comprises:
calculating the rating value of the current sample user to each sample advertisement by the following method: calculating the value of the score of the current sample user on the current sample advertisement according to various behavior index data of the current sample user on the current sample advertisement;
and calculating the value of the credit of the current sample user to the current advertisement category based on the value of the credit of each sample advertisement under the current category.
5. The method of claim 3, wherein calculating the value of the score of the current sample user for the current advertisement category according to the behavior index data of the current sample user for all sample advertisements in the current advertisement category comprises:
respectively calculating the comprehensive values of the behavior index data corresponding to the current category: calculating a comprehensive value of the current behavior index data according to the current behavior index data of each sample advertisement under the current category;
acquiring weights corresponding to various behavior index data respectively;
and carrying out weighted summation on the comprehensive values of various behavior index data to obtain the value of the credit of the current sample user on the current advertisement category.
6. The method of claim 1, wherein clustering the plurality of sample users using a clustering algorithm to obtain a plurality of user groups comprises:
randomly initializing n points from all sample users as initial clustering centers;
the following operations are executed in a loop until a preset cut-off condition is reached:
calculating the distance between each sample user and each clustering center, determining the clustering center with the minimum distance from the sample user, and classifying the sample users into a class represented by the distance center;
and re-determining the clustering center of each category according to all the updated sample users in each category.
7. The method of claim 1, wherein determining a target user group of the plurality of user groups that is closest to the target user according to the behavior index data of the target user comprises:
representing the target user by the behavior index data of the target user, and calculating the distance between the target user and each user group;
and taking the user group with the minimum distance as a target user group which is closest to the target user.
8. The method according to claim 1, wherein said: acquiring behavior index data of a plurality of sample users on each sample advertisement when browsing the advertisement; determining the value of credit of each sample user on each advertisement category respectively according to the behavior index data, and forming a credit vector corresponding to each sample user according to the value of credit of each sample user on each advertisement category respectively; representing the sample users by the grading vectors corresponding to the sample users, and clustering the plurality of sample users by adopting a clustering algorithm to obtain a plurality of user groups;
when the data of the sample user reaches a preset threshold value or the number of the sample advertisements browsed by the sample user reaches the preset threshold value, the sample advertisement behavior index data with the earliest time is cleared, so that the number of the sample advertisements is in a preset numerical value range.
9. The method of claim 1, wherein determining the recommended advertisement for the target user according to the advertisement sets of interest to the users in the target user group comprises:
and taking the M sample advertisements with the highest scoring values in the target user group as advertisements recommended for the target user, wherein M is a preset natural number.
10. The method of claim 1, wherein determining the recommended advertisement for the target user according to the advertisement sets of interest to the users in the target user group comprises:
determining N sample users which are closest to the target user in the target user group, wherein N is a preset natural number;
and determining the advertisements recommended for the target users according to the historical recommendation results of the N sample users.
11. The method of claim 10, wherein determining the recommended advertisement for the target user according to the advertisement sets of interest to the users in the target user group comprises:
determining M sample advertisements with the highest scoring values in a target user group as a first advertisement set, wherein M is a preset natural number;
acquiring actual behavior index data corresponding to each advertisement in a first advertisement set;
determining N sample users closest to the target user in the target user group, and taking the historical recommendation results of the N sample users as a second advertisement set;
acquiring behavior index prediction data corresponding to each advertisement in the second advertisement set;
acquiring an intersection of the first advertisement set and the second advertisement set;
according to the first behavior index and the behavior index prediction data, calculating behavior index prediction data of the target advertisement in the intersection;
and according to the behavior index prediction data, selecting advertisements with target advertisement quantity from the intersection as target advertisements recommended for target users.
12. An apparatus for determining recommended advertisements, comprising:
the first acquisition unit is used for acquiring behavior index data of each sample advertisement when a plurality of sample users browse the advertisements;
the first determining unit is used for determining the scores of the sample users for the advertisement categories respectively according to the behavior index data, and forming score vectors corresponding to the sample users according to the scores of the sample users for the advertisement categories;
the user clustering unit is used for representing the sample users by the grading vectors corresponding to the sample users, and clustering the plurality of sample users by adopting a clustering algorithm to obtain a plurality of user groups;
the second acquisition unit is used for acquiring behavior index data of each historical advertisement when the target user browses the historical advertisements;
a second determining unit, configured to determine, according to behavior index data of a target user, a target user group that is closest to the target user among the plurality of user groups;
and the third determining unit is used for determining the advertisements recommended for the target users according to the advertisement sets which are interested by the users in the target user group.
13. The apparatus of claim 12, further comprising:
a third obtaining unit, configured to obtain a set of sample advertisements browsed by the plurality of sample users;
and the advertisement clustering unit is used for clustering the sample advertisements in the set to obtain a plurality of advertisement categories.
14. The apparatus of claim 13, wherein the first determining unit comprises:
a first calculating subunit, configured to calculate a score value of a current sample user for a current advertisement category according to the following method: calculating the score value of the current sample user on the current advertisement category according to the behavior index data of the current sample user on all sample advertisements under the current advertisement category;
and the construction subunit is used for forming a scoring vector corresponding to the sample user according to the scoring value of the sample user to each advertisement category.
15. The apparatus of claim 14, wherein the first computing subunit comprises:
the second calculating subunit is used for calculating the scoring value of the current sample user on each sample advertisement by the following method: calculating the score value of the current sample user on the current sample advertisement according to various behavior index data of the current sample user on the current sample advertisement;
and the third calculating subunit is used for calculating the score value of the current sample user to the current advertisement category based on the score values of all sample advertisements in the current category.
16. The apparatus of claim 14, wherein the first computing subunit comprises:
a third calculating subunit, configured to calculate, respectively, a comprehensive value of each behavior index data corresponding to the current category: calculating a comprehensive value of the current behavior index data according to the current behavior index data of each sample advertisement under the current category;
the first acquiring subunit is used for acquiring weights respectively corresponding to various behavior index data;
and the summation subunit is used for carrying out weighted summation on the comprehensive values of various behavior index data to obtain the value of the credit of the current sample user on the current advertisement category.
17. The apparatus of claim 12, wherein the third determining unit comprises:
the second determining subunit is used for determining M sample advertisements with the highest score values in the target user group as a first advertisement set, wherein M is a preset natural number;
the second obtaining subunit is used for obtaining actual behavior index data corresponding to each advertisement in the first advertisement set;
the third determining subunit is used for determining N sample users which are closest to the target user in the target user group, and taking the historical recommendation results of the N sample users as a second advertisement set;
the third obtaining subunit is configured to obtain behavior index prediction data corresponding to each advertisement in the second advertisement set;
a fourth obtaining subunit, configured to obtain an intersection of the first advertisement set and the second advertisement set;
the sixth calculating subunit is used for calculating behavior index prediction data of the target advertisement in the intersection according to the first behavior index and the behavior index prediction data;
and the screening subunit is used for selecting advertisements with target advertisement quantity from the intersection as the target advertisements recommended for the target users according to the behavior index prediction data.
18. An electronic device, comprising:
a memory and a processor, the processor and the memory being communicatively connected to each other, the memory having stored therein computer instructions, the processor implementing the steps of the method of any one of claims 1 to 11 by executing the computer instructions.
19. A computer storage medium, characterized in that it stores computer program instructions which, when executed by a processor, implement the steps of the method of any one of claims 1 to 11.
20. A computer program product comprising a computer program which, when executed by a processor, carries out the steps of the method according to any one of claims 1 to 11.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211571295.9A CN115841351A (en) | 2022-12-08 | 2022-12-08 | Method and device for determining recommended advertisements and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211571295.9A CN115841351A (en) | 2022-12-08 | 2022-12-08 | Method and device for determining recommended advertisements and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115841351A true CN115841351A (en) | 2023-03-24 |
Family
ID=85578289
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211571295.9A Pending CN115841351A (en) | 2022-12-08 | 2022-12-08 | Method and device for determining recommended advertisements and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115841351A (en) |
-
2022
- 2022-12-08 CN CN202211571295.9A patent/CN115841351A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107862553B (en) | Advertisement real-time recommendation method and device, terminal equipment and storage medium | |
| US10134053B2 (en) | User engagement-based contextually-dependent automated pricing for non-guaranteed delivery | |
| US20160379268A1 (en) | User behavior data analysis method and device | |
| US8893012B1 (en) | Visual indicator based on relative rating of content item | |
| US20080103886A1 (en) | Determining relevance of a term to content using a combined model | |
| EP2242016A1 (en) | Touchpoint customization system | |
| US20080077581A1 (en) | System and method for providing medical disposition sensitive content | |
| US20190163714A1 (en) | Search result aggregation method and apparatus based on artificial intelligence and search engine | |
| US20100100443A1 (en) | User classification apparatus, advertisement distribution apparatus, user classification method, advertisement distribution method, and program used thereby | |
| Jansen et al. | Real time search on the web: Queries, topics, and economic value | |
| CN113742567B (en) | Recommendation method and device for multimedia resources, electronic equipment and storage medium | |
| CN105916032A (en) | Video recommendation method and video recommendation terminal equipment | |
| US20170169096A1 (en) | Methods and systems for analyzing reading logs and documents thereof | |
| CN111061979A (en) | User label pushing method and device, electronic equipment and medium | |
| US20140074608A1 (en) | Rapid identification of search terms that surge in response to current events | |
| CN114330329A (en) | Service content searching method and device, electronic equipment and storage medium | |
| CN110428277B (en) | Touch method of recommended product, storage medium and program product | |
| CN113961823B (en) | News recommendation method, system, storage medium and equipment | |
| CN106383857A (en) | Information processing method and electronic equipment | |
| CN111177564B (en) | Product recommendation method and device | |
| US20230316106A1 (en) | Method and apparatus for training content recommendation model, device, and storage medium | |
| CN108229990B (en) | Advertisement title generation method, device and equipment | |
| CN115841351A (en) | Method and device for determining recommended advertisements and electronic equipment | |
| CN108241699B (en) | Method and device for pushing information | |
| US11803574B2 (en) | Clustering approach for auto generation and classification of regional sports |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |