CN116205673A

CN116205673A - Account screening method, account screening device, computer equipment and storage medium

Info

Publication number: CN116205673A
Application number: CN202310250116.XA
Authority: CN
Inventors: 占天顺; 曾晓阳; 温忠养
Original assignee: Industrial and Commercial Bank of China Ltd ICBC
Current assignee: Industrial and Commercial Bank of China Ltd ICBC
Priority date: 2023-03-13
Filing date: 2023-03-13
Publication date: 2023-06-02

Abstract

The embodiment of the application provides an account screening method, relates to the technical field of computers, and can be used in the field of financial science and technology or other related fields. The method comprises the following steps: acquiring an account to be screened aiming at a target object and a plurality of historical accounts for presetting user behaviors aiming at the existence of the target object; acquiring a target historical account from a plurality of historical accounts; acquiring a plurality of account characteristics of an account to be screened, and acquiring the similarity degree between the account to be screened and a target historical account according to the plurality of account characteristics; and obtaining a target account aiming at the potential preset user behavior of the target object from the accounts to be screened according to the similarity. In the account screening method provided by the embodiment of the application, the target account is obtained from the account to be screened by combining the analytic hierarchy process with the weighted Euclidean distance, so that the screening accuracy is improved, and the target account can be accurately positioned.

Description

Account screening method, account screening device, computer equipment and storage medium

Technical Field

The present disclosure relates to the field of internet, and in particular, to a method, an apparatus, a computer device, and a storage medium for screening accounts.

Background

With the continuous development of internet technology, data Mining technology is increasingly used in various industries, and Data Mining (Data Mining) is a method, tool and process for Mining hidden, unprecedented and beneficial relationships, modes and trends from a Data set or mass Data, establishing a model for decision support by using the mined knowledge and rules, and providing predictive decision support.

At present, in the process of mining relevant data of target clients aiming at target objects, a user portrait is generally used for focusing target client groups, but because of too many attribute tags related in the user portrait, the screening of the target clients is not accurate enough and the target client groups cannot be focused accurately, so that the problem of low accuracy exists in the current client screening technology.

Disclosure of Invention

Based on the foregoing, it is necessary to provide an account screening method, apparatus, computer device and storage medium, which can improve screening accuracy and precisely locate a target account.

In a first aspect, the present application provides an account screening method. The method comprises the following steps:

acquiring accounts to be screened aiming at a target object, and presetting a plurality of historical accounts of user behaviors aiming at the existence of the target object;

Acquiring a target historical account from the plurality of historical accounts; the target historical account is a historical account with the greatest similarity with other historical accounts in the plurality of historical accounts; the other historical accounts are historical accounts of the plurality of historical accounts except the target historical account;

acquiring a plurality of account characteristics of the account to be screened, and acquiring the similarity degree between the account to be screened and the target historical account according to the plurality of account characteristics;

and acquiring a target account aiming at the potential preset user behavior of the target object from the account to be screened according to the similarity.

In one embodiment, the obtaining the similarity between the account to be screened and the target historical account according to the plurality of account features includes: according to the importance degrees among the account features, weight values corresponding to the account features are obtained; and obtaining the weighted Euclidean distance between each account to be screened and the target historical account according to the weight value respectively corresponding to each account characteristic, and taking the weighted Euclidean distance as the similarity degree between each account to be screened and the target historical account.

In one embodiment, the obtaining the weight value corresponding to each account feature according to the importance degree among the plurality of account features includes: constructing a feature comparison matrix according to the importance degrees among the account features; and carrying out consistency check on the feature comparison matrix, and acquiring weight values corresponding to the account features respectively based on the feature comparison matrix under the condition that the check result of the consistency check is that the check is passed.

In one embodiment, the obtaining, based on the feature comparison matrix, a weight value corresponding to each account feature includes: normalizing the characteristic comparison matrix to obtain a normalized comparison matrix; and acquiring weight values respectively corresponding to the account features based on the normalized comparison matrix.

In one embodiment, the normalizing the feature comparison matrix to obtain a normalized comparison matrix includes: acquiring current characteristics from the account characteristics; the current feature is any one of the account features; based on the feature comparison matrix, acquiring a plurality of first importance coefficients corresponding to the current feature, and acquiring a normalization result of each first importance coefficient; each first importance coefficient is used for representing the importance degree of each account feature to the current feature; and obtaining the normalized comparison matrix based on the normalized results of the first importance coefficients corresponding to each current feature.

In one embodiment, the obtaining the weight value corresponding to each account feature based on the normalized comparison matrix includes: summing all normalization results contained in the normalization comparison matrix based on the normalization comparison matrix to obtain a first summation result; obtaining normalization results of a plurality of second importance coefficients corresponding to the current features from the normalization comparison matrix, and summing the normalization results of the second importance coefficients to obtain second summation results; each second importance coefficient is used for representing the importance degree of the current feature to each account feature; and taking the ratio of the second summation result of the current feature to the first summation result as a weight value corresponding to the current feature.

In one embodiment, the obtaining a target historical account from the plurality of historical accounts includes: acquiring a current account and other accounts from the plurality of historical accounts; the current account is any one of the plurality of historical accounts; the rest accounts are other historical accounts except the current account in the historical accounts; acquiring a sum of weighted Euclidean distances of the current account and each of the rest accounts; and determining the current account with the smallest sum of the weighted Euclidean distances with each of the rest accounts as the target historical account.

In one embodiment, after obtaining the target account for the potential preset user behavior of the target object from the accounts to be screened according to the similarity degree, the method further includes: pushing information to be recommended aiming at the target object to the target account.

In a second aspect, the present application provides an account screening apparatus. The device comprises:

the system comprises a first acquisition module, a second acquisition module and a storage module, wherein the first acquisition module is used for acquiring an account to be screened aiming at a target object and a plurality of historical accounts for presetting user behaviors aiming at the existence of the target object;

the second acquisition module is used for acquiring a target historical account from the plurality of historical accounts; the target historical account is a historical account with the greatest similarity with other historical accounts in the plurality of historical accounts; the other historical accounts are historical accounts of the plurality of historical accounts except the target historical account;

the feature extraction module is used for acquiring a plurality of account features of the account to be screened and acquiring the similarity between the account to be screened and the target historical account according to the plurality of account features;

and the screening module is used for acquiring the target account aiming at the potential preset user behavior of the target object from the account to be screened according to the similarity.

In a third aspect, the present application also provides a computer device. The computer device comprises a memory storing a computer program and a processor which when executing the computer program performs the steps of:

In a fourth aspect, the present application also provides a computer-readable storage medium. The computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of:

In a fifth aspect, the present application also provides a computer program product. The computer program product comprises a computer program which, when executed by a processor, implements the steps of:

In the account screening method, the account screening device, the computer equipment and the storage medium, the account to be screened aiming at the target object and a plurality of historical accounts with preset user behaviors aiming at the existence of the target object can be obtained; and a target historical account can be obtained from the plurality of historical accounts; further, a plurality of account characteristics of the account to be screened are obtained, and the similarity degree between the account to be screened and the target historical account is obtained according to the plurality of account characteristics; finally, a target account aiming at the potential preset user behavior of the target object can be obtained from the accounts to be screened according to the similarity. In the account screening method provided by the embodiment of the application, the target account is obtained from the account to be screened by combining the analytic hierarchy process with the weighted Euclidean distance, so that the screening accuracy is improved, and the target account can be accurately positioned.

Drawings

Fig. 1 is a flow chart of an account screening method according to an embodiment of the present application;

fig. 2 is a schematic flow chart of obtaining a similarity degree between an account to be screened and a target historical account according to an embodiment of the present application;

fig. 3 is a schematic flow chart of obtaining weight values corresponding to the account features in the embodiment of the present application;

fig. 4 is a schematic flow chart of acquiring weight values corresponding to each account feature based on the normalized comparison matrix according to the embodiment of the present application;

fig. 5 is a schematic flow chart of obtaining a target history account according to an embodiment of the present application;

fig. 6 is a flowchart of another account screening method according to an embodiment of the present application;

fig. 7 is a block diagram of an account screening device according to an embodiment of the present application;

fig. 8 is an internal structure diagram of a computer device according to an embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application. In one embodiment, as shown in fig. 1, an account screening method is provided, where the method is applied to a server for illustration, it is understood that the method may also be applied to a terminal, and may also be applied to a system including the terminal and the server, and implemented through interaction between the terminal and the server. In this embodiment, the method includes the steps of:

Step S101, obtaining an account to be screened for a target object, and a plurality of historical accounts for the existence of the target object for presetting user behavior.

The target object may be a product to be promoted, which has popularization requirements. The product refers to anything that is used and consumed by a person and that meets a person's needs, including tangible items, intangible services, organizations, ideas, or a combination thereof. The preset user behavior may be a behavior generated by the user for the target object, and may include, but is not limited to, a degree of satisfaction of the user with the target object, a usage rate of the target object by the user, whether to purchase the target object, a requirement of the target object by the user, and the like. The account to be screened may be a population of users that have not used or consumed the target object. The plurality of history accounts may be a group of users that have used or consumed the target object and have high satisfaction with the target object, for example, may be high satisfaction users that have purchased a financial product.

Step S102, acquiring a target historical account from a plurality of historical accounts.

The target historical account can be a historical account with the greatest similarity with other historical accounts in the plurality of historical accounts, and the target historical account can also be called a centroid point historical account; the other history accounts are history accounts other than the target history account among the plurality of history accounts.

Step S103, obtaining a plurality of account characteristics of the account to be screened, and obtaining the similarity degree between the account to be screened and the target historical account according to the plurality of account characteristics.

The account feature may be a user feature that needs to be considered when screening the user population for the target object, such as deposit, age, and income information. The similarity degree can be used for measuring the probability of using or consuming the target object by the account to be screened so as to represent whether the account to be screened can be a target account with potential preset user behaviors.

Step S104, obtaining a target account aiming at the potential preset user behavior of the target object from the accounts to be screened according to the similarity.

Wherein, the target account of the potential preset user behavior may refer to a potential user of the target object.

In the account screening method, the account to be screened aiming at the target object and a plurality of historical accounts with preset user behaviors aiming at the existence of the target object can be obtained; and a target historical account can be obtained from the plurality of historical accounts; further, a plurality of account characteristics of the account to be screened are obtained, and the similarity degree between the account to be screened and the target historical account is obtained according to the plurality of account characteristics; finally, a target account aiming at the potential preset user behavior of the target object can be obtained from the accounts to be screened according to the similarity. In the account screening method provided by the embodiment of the application, the target account is obtained from the account to be screened by combining the analytic hierarchy process with the weighted Euclidean distance, so that the screening accuracy is improved, and the target account can be accurately positioned.

In some embodiments, as shown in fig. 2, step S103 may include:

step S201, according to the importance degree among the account features, weight values corresponding to the account features are obtained.

The relative importance among the plurality of account features is provided, and the importance degree of any two account features in the plurality of account features can be compared to obtain the relative importance of the two account features, and the importance degree can be described as the importance, and particularly, the relative importance degree can be described in a relative importance relation table of table 1.

The weight value can be the proportion of each account feature in all account features, and can be used for measuring the importance degree of the account features on the target object for screening the potential user group.

Importance comparison	Relative importance value	Description of the invention
			Feature 1 is more important than feature 2	3、5、7、9……	The higher the relative importance is, the larger the value is
Feature 1 is of equal importance as feature 2	1	Importance is equivalent, the value is 1
			Feature 1 is less important than feature 2	1/3、1/5、1/7、1/9…	The lower the relative importance, the smaller the value

TABLE 1 relative importance relationship table

Step S202, according to the weight values respectively corresponding to the account characteristics, the weighted Euclidean distance between each account to be screened and the target historical account is obtained and is used as the similarity degree between each account to be screened and the target historical account.

In the embodiment of the application, the similarity degree is characterized by weighting Euclidean distance. The smaller the weighted Euclidean distance between the account to be screened and the target historical account is, the higher the similarity degree between the account to be screened and the target historical account is. Euclidean distance, also known as euclidean distance, is the most common distance measure, which is the absolute distance between two points in a multidimensional space. It can also be understood that: the true distance between two points in m-dimensional space, or the natural length of the vector (i.e., the distance of the point from the origin). The euclidean distance in two and three dimensions is the actual distance between two points.

The calculation of the weighted euclidean distance between the account to be screened and the target history account can be found by equation (1):

wherein m is the number of account features; i is an ith account feature of the plurality of account features; k and p each represent an account; x is X _ki The value corresponding to the characteristic i of the account k; x is X _pi The value corresponding to the characteristic i of the account p; w (W) _i The weight value corresponding to the feature i; d represents the weighted euclidean distance between account k and account p.

The target history account may be a history account having the greatest degree of similarity to other history accounts of the plurality of history accounts, and thus, the target history account may be regarded as a most representative already used or consumed target object, and a user having high satisfaction with the target object. Therefore, if the similarity between the account to be screened and the target historical account is higher, the likelihood that the account to be screened becomes the target account of the potential preset user behavior is higher, namely the likelihood that the account to be screened becomes the potential user of the target object is higher.

In the method, the similarity degree between each account to be screened and the target historical account can be measured more accurately through the weighted Euclidean distance, so that potential users of the target objects can be screened accurately.

In some embodiments, as shown in fig. 3, step S201 may include:

step S301, a feature comparison matrix is constructed according to importance degrees among a plurality of account features.

The relative importance of any two account features in the plurality of account features can be obtained according to the importance degree among the plurality of account features, and as shown in table 1, a feature comparison matrix can be constructed according to the relative importance, that is, each element in the feature comparison matrix is a relative importance value of two account features. The following is an illustration of three account features, and the feature comparison matrix for three account features, feature 1, feature 2, and feature 3, is constructed as follows:

h is a feature comparison matrix of three constructed account features; h ₁₁ A relative importance value for feature 1 and feature 1; h ₁₂ A relative importance value for feature 1 and feature 2; h ₁₃ A relative importance value for feature 1 and feature 3; h ₂₁ A relative importance value for feature 2 and feature 1; h ₂₂ A relative importance value for feature 2 and feature 2; h ₂₃ A relative importance value for feature 2 and feature 3; h ₃₁ A relative importance value for feature 3 and feature 1; h ₃₂ A relative importance value for feature 3 and feature 2; h ₃₃ A relative importance value for feature 3 and feature 3; h ₁₁ 、H ₂₂ And H ₃₃ The three elements are 1, where the relative importance values may also include a first importance coefficient and a second importance coefficient, where the first importance coefficient is used to represent the importance of each account feature to the current feature, and the second importance coefficient is used to represent the importance of the current feature to each account feature.

Step S302, consistency verification is carried out on the feature comparison matrix, and weight values corresponding to the features of each account are obtained based on the feature comparison matrix when the verification result of the consistency verification is that the verification is passed.

In some possible implementation manners, the consistency check can be performed on the feature comparison matrix by calculating a consistency coefficient of the feature comparison matrix, the consistency check can be used for measuring rationality of relative importance value setting in the feature comparison matrix, when the consistency coefficient is smaller than a preset threshold (for example, 0.1), the setting of the relative importance value in the feature comparison matrix is reasonable, at this time, a check result of the consistency check on the feature comparison matrix is that the check is passed, and weight values corresponding to all account features respectively can be obtained based on the feature comparison matrix; when the consistency coefficient is larger than or equal to a preset threshold value, the setting of the relative importance value in the feature comparison matrix is unreasonable, and at the moment, the checking result of consistency checking of the feature comparison matrix is that the checking is not passed; in the case where the feature comparison matrix does not pass the check, it is necessary to reset each element in the feature comparison matrix. For calculation of the consistency coefficient, see the following formula (2):

Wherein CR is a consistency coefficient; lambda is the maximum feature root of the feature comparison matrix; m is the order of the feature comparison matrix; RI is an average random variable index, and RI values can be referred to in table 2RI value reference table.

m	1	2	3	4	5	6	7
								RI	0	0	0.52	0.89	1.12	1.26	1.36
m	8	9	10	11	12	13	14
								RI	1.41	1.46	1.49	1.52	1.54	1.56	1.58

Table 2RI numerical reference table

In the method, the rationality of the constructed feature comparison matrix can be checked in consistency, so that the accuracy of the weight value of the account feature obtained later is ensured.

In some embodiments, step S302 may include:

normalizing the feature comparison matrix to obtain a normalized comparison matrix; and acquiring weight values respectively corresponding to the account features based on the normalized comparison matrix.

The normalizing the feature comparison matrix to obtain the normalized comparison matrix may include:

acquiring current characteristics from the account characteristics; the current feature is any one of account features;

based on the feature comparison matrix, acquiring a plurality of first importance coefficients corresponding to the current feature, and acquiring a normalization result of each first importance coefficient; each first importance coefficient is used for representing the importance degree of each account feature to the current feature;

and obtaining a normalized comparison matrix based on the normalized results of the first importance coefficients corresponding to each current feature. Wherein the current characteristic is any one of the account characteristics.

Specifically, in some possible implementations, the feature comparison matrix H may be column summed to obtain a column summation matrix H _j To obtain the result of summing the importance of each account feature to the current feature, for example:

H _j ＝[H ₁₁ +H ₂₁ +H ₃₁ H ₁₂ +H ₂₂ +H ₃₂ H ₁₃ +H ₂₃ +H ₃₃ ]

in-column summing matrix H _j In (1) for element H ₁₁ +H ₂₁ +H ₃₁ The current feature is feature 1, H ₁₁ +H ₂₁ +H ₃₁ Namely, the result of summing the importance of the account features (feature 1, feature 2 and feature 3) to the current feature (feature 1); similarly, for element H ₁₂ +H ₂₂ +H ₃₂ The current feature is feature 2, H ₁₂ +H ₂₂ +H ₃₂ Summing the results for the importance of the account features (feature 1, feature 2, and feature 3) to the current feature (feature 2); h ₁₃ +H ₂₃ +H ₃₃ The results are summed for the importance of the account feature (feature 1, feature 2, and feature 3) to the current feature (feature 3).

After the importance summation result of each account feature on the current feature is obtained, the feature comparison matrix H and the column summation matrix H can be used for comparing the feature comparison matrix H with the column summation matrix H _j Is determined as normalized comparison matrix H ^* The normalized comparison matrix H ^* The element in (a) is the normalization result of each first importance coefficient. For example:

wherein, normalize the comparison matrix H ^* The element in (a) is the normalized result of each first importance coefficient, and can represent the proportion of the importance of each account feature to the current feature in the importance summation result. Taking the first column as an example, the comparison matrix H is normalized ^* The current features of the first column in (a) are feature 1,

it may be the specific gravity of the importance of feature 2 to feature 1 in the result of the summation of the importance of the account feature to the current feature (feature 1); />

It may be the specific gravity of the importance of feature 3 to feature 1 in the result of the summation of the importance of the account feature to the current feature (feature 1).

In some embodiments, as shown in fig. 4, obtaining weight values corresponding to the account features based on the normalized comparison matrix may include:

step S401, based on the normalized comparison matrix, summing the normalized results contained in the normalized comparison matrix to obtain a first summed result.

In some possible implementations, the implementation of step S401 is to normalize the comparison matrix H ^* Summing all elements in the list to obtain the sum of the weights of the importance of the account features to the current features in the importance summation result, namely a first summation result.

Step S402, obtaining normalization results of a plurality of second importance coefficients corresponding to the current features from the normalization comparison matrix, and summing the normalization results of the second importance coefficients to obtain a second summation result.

Wherein each second importance coefficient is used for representing the importance degree of the current feature to each account feature. In the above-described normalized comparison matrix,

And->

The normalization result of a plurality of corresponding second importance coefficients when the first feature is used as the current feature; similarly, the normalization results of the current feature, which is feature 2, and the plurality of second importance coefficients corresponding to the feature, may also be obtained.

In some possible implementations, the second summation result may be obtained by row summing the normalized comparison matrix.

Step S403, a ratio of the second summation result and the first summation result of the current feature is used as a weight value corresponding to the current feature.

Specifically, in a first step, the comparison matrix H is normalized ^* Performing row summation to obtain a row summation matrix H _i I.e. the second summation result; the method comprises the following steps:

second step, summing the rows in matrix H _i The sum of the specific gravity and the sum is the ratio of the second sum result to the first sum result of the current feature to obtain a weight matrix H _k The method comprises the following steps:

wherein the weight matrix H _k The elements in (a) may represent weight values corresponding to the account features, respectively.

The weight value corresponding to the feature 1;

the weight value corresponding to the feature 2;

the weight value corresponding to the feature 3. />

In the method, the weight value of each account characteristic can be accurately determined based on the characteristic comparison matrix, so that the similarity degree of the account to be screened and the target historical account can be obtained conveniently.

In some embodiments, as shown in fig. 5, step S102 may include:

step S501, obtaining a current account and other accounts from a plurality of historical accounts.

The current account is any one of a plurality of historical accounts; the remaining accounts are other historical accounts of the historical accounts except the current account.

Step S502, obtaining the sum of weighted Euclidean distances between the current account and each other account.

In the first step, the weighted euclidean distance between the current account and each of the remaining accounts may be obtained, and the calculation of the weighted euclidean distance may be referred to in the above formula (1) in step S202.

And secondly, obtaining the sum of the weighted Euclidean distances of the current account and each other account.

In step S503, the current account with the smallest sum of the weighted euclidean distances from the remaining accounts is determined as the target history account.

In the method, the target historical account can be accurately determined, so that the account to be screened can be screened based on the target historical account conveniently, and the target account aiming at the potential preset user behavior of the target object can be determined.

In some embodiments, as shown in fig. 6, another account screening method is provided, which may include:

step S601, obtaining an account to be screened for a target object, and a plurality of historical accounts for the existence of the target object for preset user behavior.

Step S602, obtaining a current account and other accounts from a plurality of historical accounts.

Step S603, obtaining the sum of weighted Euclidean distances of the current account and the rest accounts.

Step S604, the current account with the smallest sum of the weighted euclidean distances from the remaining accounts is determined as the target history account.

Step S605 obtains a plurality of account features of the account to be screened.

The account feature may be a user feature that needs to be considered when screening the user population for the target object, such as deposit, age, and income information.

Step S606, a feature comparison matrix is constructed according to the importance degree among the account features.

TABLE 1 relative importance relationship table

According to the importance degree among the plurality of account features, the relative importance of any two account features in the plurality of account features can be obtained, and as shown in table 1, a feature comparison matrix can be constructed according to the relative importance, namely, each element in the feature comparison matrix is the relative importance value of the two account features. The following is an illustration of three account features, and the feature comparison matrix for three account features, feature 1, feature 2, and feature 3, is constructed as follows:

H is a feature comparison matrix of three constructed account features; h ₁₁ A relative importance value for feature 1 and feature 1; h ₁₂ A relative importance value for feature 1 and feature 2; h ₁₃ A relative importance value for feature 1 and feature 3; h ₂₁ A relative importance value for feature 2 and feature 1; h ₂₂ A relative importance value for feature 2 and feature 2; h ₂₃ Is of relative importance for feature 2 and feature 3Sex values; h ₃₁ A relative importance value for feature 3 and feature 1; h ₃₂ A relative importance value for feature 3 and feature 2; h ₃₃ A relative importance value for feature 3 and feature 3; h ₁₁ 、H ₂₂ And H ₃₃ The three elements are 1, where the relative importance values may also include a first importance coefficient and a second importance coefficient, where the first importance coefficient is used to represent the importance of each account feature to the current feature, and the second importance coefficient is used to represent the importance of the current feature to each account feature.

Step S607, carrying out consistency check on the feature comparison matrix, and carrying out normalization processing on the feature comparison matrix to obtain a normalization comparison matrix when the check result of the consistency check is that the check result passes.

Table 2RI numerical reference table

and obtaining a normalized comparison matrix based on the normalized results of the first importance coefficients corresponding to each current feature.

Wherein the current characteristic is any one of the account characteristics.

Specifically, in some possible implementations, the feature comparison matrix H may be column summed to obtain a column summation matrix H _j To obtain a summation of the importance of the account feature to the current feature, such as:

it may be the specific gravity of the importance of feature 2 to feature 1 in the result of the summation of the importance of the account feature to the current feature (feature 1); / >

The result of summing the importance of feature 3 to feature 1 over the importance of the account feature to the current feature (feature 1) may beIs a specific gravity of (b).

Step S608, based on the normalized comparison matrix, performs summation processing on each normalized result included in the normalized comparison matrix, to obtain a first summation result. In some possible implementations, the implementation of step S401 is to normalize the comparison matrix H ^* Summing all elements in the list to obtain the sum of the weights of the importance of the account features to the current features in the importance summation result, namely a first summation result.

Step S609, obtaining normalization results of a plurality of second importance coefficients corresponding to the current features from the normalization comparison matrix, and summing the normalization results of the second importance coefficients to obtain a second summation result.

and->

In step S610, the ratio of the second summation result and the first summation result of the current feature is used as the weight value corresponding to the current feature.

The weight value corresponding to the feature 1;

the weight value corresponding to the feature 2;

the weight value corresponding to the feature 3.

Step S611, respectively carrying out forward normalization processing on the values corresponding to the account features of each account to be screened to obtain forward normalization values of the account features.

The forward normalization process may include a forward normalization process and a normalization process. The forward processing is performed on the values corresponding to the account features of each account to be screened to obtain forward values of each account feature, specifically, see the forward calculation mode table of table 3 below, where table 3 may include account feature data types and corresponding calculation methods.

TABLE 3 Forward calculation method Table

The normalized values corresponding to the account features of each account to be screened are normalized respectively to obtain the forward normalized values of each account feature, and specifically, the normalized calculation mode table of table 4 may be referred to below, and table 4 may include account feature data types and corresponding calculation methods.

Table 4 normalized calculation mode table

Step S612, according to the weight value and the forward normalization value respectively corresponding to the account characteristics, the weighted Euclidean distance between each account to be screened and the target historical account is obtained and is used as the similarity degree between each account to be screened and the target historical account.

Step S613, obtaining the target account of the potential preset user behavior for the target object from the accounts to be screened according to the similarity.

In some possible implementations, the accounts to be screened may be ranked according to the similarity between the accounts to be screened and the target historical accounts from large to small according to the similarity, so as to obtain a list of accounts to be screened with priority arrangement, and then, according to the preset coverage rate, determining the target account of the potential preset user behavior aiming at the target object in the list of accounts to be screened.

It should be understood that, although the steps in the flowcharts related to the embodiments described above are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.

Based on the same inventive concept, the embodiment of the application also provides an account screening device for realizing the account screening method. The implementation of the solution provided by the device is similar to the implementation described in the above method, so the specific limitation in the embodiment of one or more account screening devices provided below may be referred to the limitation of the account screening method hereinabove, and will not be repeated here.

In one embodiment, as shown in fig. 7, there is provided an account screening apparatus including: a first acquisition module 710, a second acquisition module 720, a feature extraction module 730, and a screening module 740, wherein:

the first obtaining module 710 is configured to obtain an account to be screened for a target object, and a plurality of historical accounts for existence of the target object for preset user behavior.

A second obtaining module 720, configured to obtain a target historical account from the plurality of historical accounts; the target historical account is a historical account with the greatest similarity with other historical accounts in the plurality of historical accounts; the other history accounts are history accounts other than the target history account among the plurality of history accounts.

The feature extraction module 730 is configured to obtain a plurality of account features of the account to be screened, and obtain a degree of similarity between the account to be screened and the target historical account according to the plurality of account features.

And the screening module 740 is used for acquiring the target account aiming at the potential preset user behavior of the target object from the accounts to be screened according to the similarity.

In one embodiment, the feature extraction module 730 is further configured to:

and acquiring weight values corresponding to the account features respectively according to the importance degrees among the account features.

And obtaining the weighted Euclidean distance between each account to be screened and the target historical account according to the weight value corresponding to each account characteristic, and taking the weighted Euclidean distance as the similarity degree between each account to be screened and the target historical account.

In one embodiment, the feature extraction module 730 is further configured to:

and constructing a feature comparison matrix according to the importance degrees among the account features.

And carrying out consistency check on the feature comparison matrix, and acquiring weight values respectively corresponding to the features of each account based on the feature comparison matrix under the condition that the check result of the consistency check is that the check is passed.

In one embodiment, the feature extraction module 730 is further configured to:

and carrying out normalization processing on the characteristic comparison matrix to obtain a normalized comparison matrix.

And acquiring weight values respectively corresponding to the account features based on the normalized comparison matrix.

In one embodiment, the feature extraction module 730 is further configured to:

Acquiring current characteristics from the account characteristics; the current characteristic is any one of the account characteristics.

Based on the feature comparison matrix, acquiring a plurality of first importance coefficients corresponding to the current feature, and acquiring a normalization result of each first importance coefficient; each first importance coefficient is used for representing the importance degree of each account feature to the current feature.

In one embodiment, the feature extraction module 730 is further configured to:

summing all normalization results contained in the normalization comparison matrix based on the normalization comparison matrix to obtain a first summation result;

obtaining normalization results of a plurality of second importance coefficients corresponding to the current features from the normalization comparison matrix, and summing the normalization results of the second importance coefficients to obtain second summation results; each second importance coefficient is used for representing the importance degree of the current feature to each account feature;

and taking the ratio of the second summation result of the current feature to the first summation result as the weight value corresponding to the current feature.

In one embodiment, the second obtaining module 720 is further configured to:

Acquiring a current account and other accounts from a plurality of historical accounts; the current account is any one of a plurality of historical accounts; the remaining accounts are other historical accounts of the historical accounts except the current account.

And obtaining the sum of the weighted Euclidean distances of the current account and each other account.

And determining the current account with the smallest sum of the weighted Euclidean distances with the rest accounts as a target historical account.

In one embodiment, the screening module 740 is further configured to:

pushing information to be recommended aiming at the target object to the target account.

The various modules in the account screening apparatus described above may be implemented in whole or in part in software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.

In one embodiment, a computer device is provided, which may be a server, and the internal structure of which may be as shown in fig. 8. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is for storing account screening data. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement an account screening method.

It will be appreciated by those skilled in the art that the structure shown in fig. 8 is merely a block diagram of some of the structures associated with the present application and is not limiting of the computer device to which the present application may be applied, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.

In an embodiment, there is also provided a computer device comprising a memory and a processor, the memory having stored therein a computer program, the processor implementing the steps of the method embodiments described above when the computer program is executed.

In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored which, when executed by a processor, carries out the steps of the method embodiments described above.

In an embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the steps of the method embodiments described above.

It should be noted that, user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, presented data, etc.) referred to in the present application are information and data authorized by the user or sufficiently authorized by each party.

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the various embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magnetic random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (Phase Change Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), and the like. The databases referred to in the various embodiments provided herein may include at least one of relational databases and non-relational databases. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic units, quantum computing-based data processing logic units, etc., without being limited thereto.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The above examples only represent a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the present application. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application shall be subject to the appended claims.

Claims

1. A method for screening accounts, the method comprising:

acquiring accounts to be screened aiming at a target object and a plurality of historical accounts for presetting user behaviors aiming at the existence of the target object;

2. The method of claim 1, wherein the obtaining the degree of similarity between the account to be screened and the target historical account according to the plurality of account features comprises:

according to the importance degrees among the account features, weight values corresponding to the account features are obtained;

and obtaining the weighted Euclidean distance between each account to be screened and the target historical account according to the weight value respectively corresponding to each account characteristic, and taking the weighted Euclidean distance as the similarity degree between each account to be screened and the target historical account.

3. The method according to claim 2, wherein the obtaining the weight value corresponding to each account feature according to the importance degree among the plurality of account features includes:

constructing a feature comparison matrix according to the importance degrees among the account features;

And carrying out consistency check on the feature comparison matrix, and acquiring weight values corresponding to the account features respectively based on the feature comparison matrix under the condition that the check result of the consistency check is that the check is passed.

4. The method of claim 3, wherein the obtaining, based on the feature comparison matrix, a weight value for each of the account features, comprises:

normalizing the characteristic comparison matrix to obtain a normalized comparison matrix;

5. The method of claim 4, wherein normalizing the feature comparison matrix to obtain a normalized comparison matrix comprises:

acquiring current characteristics from the account characteristics; the current feature is any one of the account features;

And obtaining the normalized comparison matrix based on the normalized results of the first importance coefficients corresponding to each current feature.

6. The method of claim 5, wherein the obtaining weight values for each of the account features based on the normalized comparison matrix comprises:

and taking the ratio of the second summation result of the current feature to the first summation result as a weight value corresponding to the current feature.

7. The method of claim 1, wherein the obtaining a target historical account from the plurality of historical accounts comprises:

acquiring a current account and other accounts from the plurality of historical accounts; the current account is any one of the plurality of historical accounts; the rest accounts are other historical accounts except the current account in the historical accounts;

Acquiring a sum of weighted Euclidean distances of the current account and each of the rest accounts;

and determining the current account with the smallest sum of the weighted Euclidean distances with each of the rest accounts as the target historical account.

8. The method according to claim 1, wherein after obtaining the target account for the potential preset user behavior of the target object from the accounts to be screened according to the similarity degree, the method further comprises:

9. An account screening apparatus, the apparatus comprising:

10. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1-8 when the computer program is executed.

11. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method according to any of claims 1-8.

12. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, implements the steps of the method according to any one of claims 1-8.