User portrait scoring method, device, computer equipment and readable storage medium

Abstract

The present application provides a user portrait scoring method, apparatus, computer equipment and computer-readable storage medium, which belong to the field of data analysis technology. In order to solve the problem of user portrait accuracy, by obtaining multi-dimensional indicator information of the user, the similarity of information features between different indicator information is counted, and according to the similarity of information features, redundant indicator information is removed to obtain target indicator information, and then the target indicator information is normalized, and the sample proportion of the target indicator information is counted. According to the sample proportion and the coefficient of variation, the indicator weight of the target indicator information in the user portrait scoring process is obtained, and the normalized indicator feature value and the indicator weight corresponding to each target indicator information are multiplied, and the product results are added to obtain the user portrait score for the user portrait. The user's multi-dimensional features can be accurately and comprehensively measured, redundant indicator information is removed, and the accuracy of the user portrait is improved.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a user portrait scoring method, apparatus, computer equipment, and computer-readable storage medium. Background Art

[0002] Users are the foundation for the scalable development of IoT services. Comprehensive user quality assessments based on user-level data insights and analysis are crucial for improving IoT customer acquisition and achieving high-quality, scalable growth. With the development of the IoT industry, operators at all levels are increasingly demanding comprehensive user quality assessments. However, traditional data support systems suffer from incomplete basic user information, a lack of a comprehensive user information view, a limited understanding of existing users, a lack of a comprehensive user evaluation system, a lack of channels for acquiring external users, insufficient user needs exploration, a lack of user lifecycle management tools, and inefficient user management. Furthermore, management lacks a comprehensive operational view that reaches down to the front lines, business and operational analysts lack data granularity down to district and county branches, and lack self-service data acquisition tools. Marketing personnel lack data support for their own development and user profiles. Consequently, traditional data support systems are no longer able to meet or adequately address the data needs of all levels of operations for comprehensive user quality assessments.

[0003] User profile scoring models emerged in response to these circumstances, used to assess users' comprehensive qualities. The most critical and technically challenging aspect of user profile scoring models is the design of the scoring algorithm. Ensuring objective and accurate scoring results is central to algorithm design. However, traditional user profile scoring methods often rely heavily on the subjective experience of business personnel, resulting in a lack of objectivity and accuracy, potentially leading to adverse consequences for business operations. Summary of the Invention

[0004] The present application provides a user portrait scoring method, apparatus, computer equipment and computer-readable storage medium, which can solve the technical problem of low accuracy of user portrait scoring in traditional technologies.

[0005] In a first aspect, the present application provides a user portrait scoring method, comprising: obtaining original data information corresponding to preset initial indicator information of different dimensions of the same user, and digitizing the original data information to obtain indicator values ​​corresponding to the original data information; based on the indicator values, statistically analyzing the similarities of information features between different preset initial indicator information, and based on the information feature similarities, removing redundant preset initial indicator information to obtain target indicator information; normalizing the target indicator values ​​corresponding to the target indicator information to obtain normalized indicator characteristic values; based on the normalized indicator characteristic values, obtaining the target indicator values ​​of other users that belong to the same target as the normalized indicator characteristic values; The normalized indicator characteristic value of the same attribute of the indicator information, and based on the normalized indicator characteristic value of the same attribute and the normalized indicator characteristic value, the sample proportion of the normalized indicator characteristic value in the sum of the normalized indicator characteristic value of the same attribute and the normalized indicator characteristic value is counted; according to the sample proportion, the coefficient of variation generated by the indicator value in the normalization process is counted, and according to the coefficient of variation, the indicator weight of the target indicator information in the user portrait scoring process is obtained; the normalized indicator characteristic value corresponding to each item of the target indicator information of the user and the indicator weight are multiplied, and the multiplication results are added to obtain the user portrait score for the user.

[0006] On the second aspect, the present application also provides a user portrait scoring device, including: a digitization unit, used to obtain the original data information corresponding to the preset initial indicator information of different dimensions of the same user, and digitize the original data information to obtain the indicator value corresponding to the original data information; a first statistical unit, used to count the information feature similarities between different preset initial indicator information according to the indicator value, and remove redundant preset initial indicator information according to the information feature similarity to obtain target indicator information; a normalization unit, used to normalize the target indicator value corresponding to the target indicator information to obtain a normalized indicator characteristic value; a second statistical unit, used to obtain the information feature similarities of other users corresponding to the normalized indicator according to the normalized indicator characteristic value. The standard characteristic value belongs to the normalized indicator characteristic value of the same attribute of the same target indicator information, and according to the normalized indicator characteristic value of the same attribute and the normalized indicator characteristic value, the sample proportion of the normalized indicator characteristic value in the sum of the normalized indicator characteristic value of the same attribute and the normalized indicator characteristic value is counted; the first acquisition unit is used to count the coefficient of variation generated by the indicator value in the normalization process according to the sample proportion, and according to the coefficient of variation, obtain the indicator weight of the target indicator information in the user portrait scoring process; the calculation unit is used to multiply the normalized indicator characteristic value corresponding to each item of the target indicator information of the user and the indicator weight, and add the multiplication results to obtain the user portrait score for the user.

[0007] In a third aspect, the present application also provides a computer device comprising a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps of the user portrait scoring method when executing the computer program.

[0008] In a fourth aspect, the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the processor executes the steps of the user portrait scoring method.

[0009] The present application provides a user portrait scoring method, apparatus, computer device and computer-readable storage medium, which obtains multi-dimensional indicator information of the user and digitizes the indicator information to obtain indicator values. According to the indicator values, the similarity of information features between different indicator information is statistically analyzed. According to the similarity of information features, redundant indicator information is removed to obtain target indicator information. The target indicator information is then normalized to remove the influence of different dimensions, and the sample proportion of the target indicator information is statistically analyzed. According to the sample proportion and the coefficient of variation, the indicator weight of the target indicator information in the user portrait scoring process is obtained. Finally, The normalized indicator characteristic value and indicator weight corresponding to each target indicator information of the user are multiplied, and the product results are added to obtain the user portrait score for the user. Therefore, in the process of user portrait, the characteristics of multiple dimensions of the user can be accurately and comprehensively measured in a quantitative form, and redundant indicator information is removed, fully ensuring the independence of the target indicator information. Then, the information contained in the user data of multiple dimensions can be objectively and fully comprehensively mined to obtain a more accurate, objective and comprehensive portrait score for the user, thereby improving the accuracy of the user portrait and avoiding the bias caused by personal subjective tendencies in user evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0011] Figure 1 A flowchart of the user portrait scoring method provided in an embodiment of the present application;

[0012] Figure 2 This is a schematic diagram of the first sub-process of the user portrait scoring method provided in an embodiment of the present application;

[0013] Figure 3This is a schematic diagram of the second sub-process of the user portrait scoring method provided in an embodiment of the present application;

[0014] Figure 4 This is a schematic diagram of the third sub-process of the user portrait scoring method provided in an embodiment of the present application;

[0015] Figure 5 This is a schematic diagram of the fourth sub-process of the user portrait scoring method provided in an embodiment of the present application;

[0016] Figure 6 A schematic block diagram of a user portrait scoring device provided in an embodiment of the present application;

[0017] Figure 7 A schematic block diagram of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0019] It will be understood that when used in this specification and the appended claims, the terms “comprises” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.

[0020] See also Figure 1 , Figure 1 This is a flow chart of the user portrait scoring method provided in the embodiment of the present application. Figure 1 As shown, the method includes the following steps S11-S16:

[0021] S11. Obtain original data information corresponding to preset initial indicator information of different dimensions of the same user, and digitize the original data information to obtain indicator values ​​corresponding to the original data information.

[0022] Specifically, pre-stored user data information of different dimensions can be obtained from the HIVE data warehouse. The user data information may include user data information of dimensions such as user profile dimension information, business dimension information, value dimension information, risk dimension information, behavior dimension information and cooperation dimension information. The user data information of each dimension may also include sub-dimension information. Each sub-dimension information may include several preset initial indicator information. Each preset initial indicator information corresponds to the original data information. The sub-dimension information included in the profile dimension information may involve basic information, asset information, risk information and operating conditions. Among them, the preset initial indicator information included in the basic information involves indicator information such as the number of online users, the industry to which they belong, the user category, the establishment time and the number of employees. The indicator value corresponding to the original data information is the value of the number of online users, the value corresponding to the industry to which they belong, the value corresponding to the user category, the value corresponding to the establishment time and the number of employees. The values ​​corresponding to the scale of work, etc., and so on, please refer to the following Table 1. As shown in Table 1, in the field of Internet of Things, for Internet of Things users, based on the HIVE data warehouse, original user data information can be collected from multiple tables such as call bill data table, user data table, customer data table, etc., and the original user data can be aggregated and connected to the first wide table, and then the first wide table is exported to a CSV file, and then the data in the CSV file is read in using R language, and the data in the CSV file is preprocessed such as outlier detection and deletion, missing value filling, etc. to obtain the original data information corresponding to the preset initial indicator information of different dimensions of the same user, and the original data information is digitized to obtain the indicator value corresponding to the original data information. The index value can be stored in the intermediate wide table, so as to subsequently score the user portrait based on the data in the intermediate wide table. The user dimension can specifically include the content shown in Table 1.

[0023] Table 1

[0024]

[0025]

[0026]

[0027] Among them, for IoT users, the overview dimension is used to quantitatively evaluate the customer's overall situation by evaluating the customer's basic information, risk information, asset information, and operating conditions; the business dimension is used to quantitatively evaluate the customer's business usage by evaluating the relevant circumstances of the customer's use of IoT services, such as the number of users of subordinate industry applications, network standards, etc.; the value dimension is used to conduct statistical analysis of the customer's breach of contract and violation of regulations based on the data provided by the risk monitoring system, and to quantitatively evaluate the risk of breach of contract or violation; the behavior dimension is used to quantitatively evaluate the behavior pattern of the customer through the Internet behavior characteristics of the users under the name of the customer; the cooperation dimension is used to quantitatively evaluate the value and possibility of commercial cooperation by evaluating the customer type, business model, and cooperation support characteristics.

[0028] After obtaining the original data information corresponding to the preset initial indicator information of different dimensions of the same user, the original data information is digitized. For the preset initial indicator information that is a numerical value, the numerical value corresponding to the preset initial indicator information is directly extracted. For example, for numerical preset initial indicator information such as the number of online users and the scale of employees, the numerical value corresponding to the preset initial indicator information can be directly extracted to obtain the indicator value of the preset initial indicator information. For the preset initial indicator information that is non-numerical, for example, for preset initial indicator information such as the industry and user category, the importance of the non-numerical preset initial indicator information such as the industry and user category in scoring the user portrait can be used to pre-assign corresponding numerical values ​​to the non-numerical preset initial indicator information such as the industry and user category, thereby digitizing the original data information corresponding to the non-numerical preset initial indicator information to obtain the indicator value corresponding to the preset initial indicator information. For example, for users in industries closely related to the Internet of Things industry, when scoring the Internet of Things user portrait, users in industries closely related to the Internet of Things industry can be given larger numerical values, etc.

[0029] Further, see Figure 2 , Figure 2 This is a schematic diagram of the first sub-process of the user portrait scoring method provided in the embodiment of the present application, such as Figure 2 As shown, in this embodiment, the raw data information is digitized to obtain the indicator value corresponding to the preset initial indicator information, including:

[0030] S111, digitizing the original data information to obtain initial indicator values;

[0031] S112: Normalize the initial indicator value to obtain the indicator value corresponding to the preset initial indicator information.

[0032] Specifically, in order to solve the problem that the raw data information corresponding to different preset initial indicator information has a skewed distribution, the raw data information is first digitized to obtain an initial indicator value to reduce the skewness of the preset initial indicator information, and then the initial indicator value is standardized to obtain an indicator value corresponding to the preset initial indicator information to remove the dimension of the preset initial indicator information. For example, for the raw data information corresponding to a preset initial indicator information, the raw data information is digitized and a logarithmic transformation log(x+1) can be used to obtain an initial indicator value x, and then the initial indicator value x is standardized, wherein for high-quality preset initial indicator information, the standardized result r can be calculated by the following formula (1), and for low-quality preset initial indicator information, the standardized result r can be calculated by the following formula (2), thereby obtaining the indicator value corresponding to each of the preset initial indicator information.

[0033]

[0034]

[0035] Among them, r ij is the result of standardization, that is, the index value corresponding to the preset initial index information. If in a matrix, r ij is the element at the i-th row and j-th column position, min{x j} is the minimum value of all initial indicator values ​​corresponding to the preset initial indicator information x of different users. If in a matrix, min{x j} is the smallest value in the jth column, man{x j} is the maximum value of all initial indicator values ​​corresponding to the preset initial indicator information x of different users. If in a matrix, man{x j} is the largest value in the jth column.

[0036] S12. Counting the similarities of information features between different preset initial indicator information based on the indicator values, and removing redundant preset initial indicator information based on the information feature similarities to obtain target indicator information.

[0037] Specifically, in order to prevent redundant information from increasing the variance of the user portrait scoring results and thus reducing the accuracy of the user portrait scoring, the information feature similarity between different preset initial indicator information is analyzed to remove redundant preset initial indicator information through information feature similarity analysis, and redundant preset initial indicator information with high similarity is removed, thereby ensuring that the retained different target indicator information has high independence. First, the information feature similarity between different preset initial indicator information is calculated. If the information feature similarity between two different preset initial indicator information meets the preset similarity threshold, the two different preset initial indicator information are regarded as similar preset initial indicator information. If the information feature similarity between two different preset initial indicator information does not meet the preset similarity threshold, the two different preset initial indicator information are regarded as dissimilar preset initial indicator information, that is, the two different preset initial indicator information have independence. If two different preset initial indicator information are regarded as similarity preset initial indicator information, one of which is the preset first initial indicator information and the other is the preset second initial indicator information, then based on the information feature similarity between the preset first initial indicator information and all other preset initial indicator information of the same category, the average similarity of the first information feature corresponding to the preset first initial indicator information is obtained, and based on the information feature similarity between the preset second initial indicator information and all other preset initial indicator information of the same category, the average similarity of the second information feature corresponding to the preset second initial indicator information is obtained, and then based on the first information feature average similarity and the second information feature average similarity, the preset initial indicator information with the highest average information feature similarity is removed, and the retained one is the target indicator information.

[0038] Further, see Figure 3 , Figure 3 This is a schematic diagram of the second sub-process of the user portrait scoring method provided in the embodiment of the present application, such as Figure 3 As shown, in this embodiment, based on the indicator values, the information feature similarities between different preset initial indicator information are counted, and based on the information feature similarities, redundant preset initial indicator information is removed to obtain target indicator information, including:

[0039] S121. Based on the indicator value, arbitrarily obtain two preset initial indicator information, if one of them is preset first initial indicator information and the other is preset second initial indicator information, calculate the information feature similarity between the preset first initial indicator information and the preset second initial indicator information, and obtain an information feature similarity value corresponding to the information feature similarity;

[0040] S122: If the information feature similarity value is greater than or equal to a preset similarity threshold, use the preset first initial indicator information and the preset second initial indicator information as similarity indicator information;

[0041] S123. Obtaining other information feature similarity values ​​corresponding to the first initial indicator information and all other preset initial indicator information, and averaging all of the other information feature similarity values ​​to obtain a first information feature average similarity value corresponding to the first initial indicator information, and obtaining a second information feature average similarity value corresponding to the second initial indicator information;

[0042] S124. Compare the average similarity value of the first information feature with the average similarity value of the second information feature, obtain initial indicator information of the smaller average similarity value of the first information feature and the second information feature, and obtain target indicator information.

[0043] Specifically, according to the indicator value, two preset initial indicator information are arbitrarily obtained as an indicator information feature pair. If one of the preset initial indicator information of the indicator information feature pair is the preset first initial indicator information and the other preset initial indicator information is the preset second initial indicator information, the information feature similarity between the preset first initial indicator information and the preset second initial indicator information is counted to obtain the information feature similarity value corresponding to the information feature similarity. If x is set i is the information characteristic value of the preset first initial indicator information of the i-th user, y i The information feature value of the preset second initial indicator information of the i-th user can be used to calculate the value of the information feature similarity value r corresponding to the information feature similarity, and the similarity between the two features can be judged by the size of the r value:

[0044]

[0045] in, is the average value of the sum of the respective indicator values ​​of different users under the preset first initial indicator information. If in a matrix, is x i The average value of the sum of the respective indicator values ​​under the initial indicator information of the column, is the average value of the sum of the respective indicator values ​​of different users under the preset second initial indicator information, y i The average value of the sum of the respective indicator values ​​under the initial indicator information of the column. The closer the absolute value of r is to 1, the higher the feature similarity between the two preset initial indicator information is, and the closer it is to 0, the lower the feature similarity between the two preset initial indicator information is.

[0046] Then, based on the information feature similarity value, if the information feature similarity value is greater than or equal to a preset similarity threshold, for example, the preset similarity threshold may be 0.8, that is, |r|≥0.8, the preset first initial indicator information and the preset second initial indicator information are used as similar indicator information. If the information feature similarity value is less than the preset similarity threshold, the preset first initial indicator information and the preset second initial indicator information are dissimilar indicator information. Then, other information feature similarity values ​​corresponding to the first initial indicator information and all other preset initial indicator information are obtained, and all the other information feature similarity values ​​are averaged to obtain a first information feature average similarity value corresponding to the first initial indicator information, and a second information feature average similarity value corresponding to the second initial indicator information is obtained. The first information feature average similarity value is compared with the second information feature average similarity value, and the initial indicator information with the highest information feature average similarity value is removed. The initial indicator information with the smaller information feature average similarity value between the first information feature average similarity value and the second information feature average similarity value can be obtained to obtain the target indicator information. All constructed indicator information feature pairs are screened through the above process, and ultimately all target indicator information have good independence.

[0047] S13. Normalize the target indicator value corresponding to the target indicator information to obtain a normalized indicator characteristic value.

[0048] Specifically, since the original data information corresponds to an indicator value, the value corresponding to the target indicator information can be called a target indicator value, and the target indicator value is normalized to obtain a normalized indicator characteristic value, thereby converting the preset initial indicator information of different dimensions from a dimensional expression to a dimensionless expression to solve the problem of different dimensions of the preset initial indicator information.

[0049] Further, see Figure 4 , Figure 4 This is a schematic diagram of the third sub-process of the user portrait scoring method provided in the embodiment of the present application, such as Figure 4 As shown, in this embodiment, the target indicator value corresponding to the target indicator information is normalized to obtain a normalized indicator characteristic value, including:

[0050] S131, constructing a target indicator information matrix using target indicator values ​​of target indicator information of different dimensions for the same user as rows and target indicator values ​​of target indicator information of the same dimension for different users as columns;

[0051] S132. Based on the maximum value and the minimum value contained in each column of the target indicator information matrix, the column values ​​in each column are normalized to obtain a normalized indicator characteristic value.

[0052] Specifically, the raw data information corresponding to the preset initial indicator information of different dimensions for different users can be obtained, and the raw data information can be digitized to obtain the indicator values ​​corresponding to the raw data information. The indicator values ​​of the preset initial indicator information of different dimensions for the same user are used as rows, and the indicator values ​​of the preset initial indicator information of the same dimension for different users are used as columns to construct an initial indicator information matrix, thereby constructing the initial indicator information matrix from the indicator values ​​of different dimensions for different users. Based on the initial indicator information matrix, the information feature similarity between the different preset initial indicator information is analyzed to remove redundant preset initial indicator information through information feature similarity analysis. The columns containing redundant preset initial indicator information with high similarity are removed, thereby retaining the columns containing different target indicator information. The matrix composed of the retained columns is the target indicator information matrix, and the target indicator values ​​of the target indicator information of different dimensions for the same user are used as rows, and the target indicator values ​​of the target indicator information of the same dimension for different users are used as columns. Then, based on the maximum and minimum values ​​contained in each column of the target indicator information matrix, the column values ​​in each column are normalized to obtain normalized indicator feature values. For example, the following n*m matrix A can be constructed based on target indicator information of different dimensions for different users:

[0053]

[0054] Among them, each row describes the target indicator value corresponding to all the target indicator information of a user, which can be called a user row. For example, in the field of Internet of Things, each row can describe the user data information of various dimensions such as the profile dimension information, business dimension information, value dimension information, risk dimension information, behavior dimension information and cooperation dimension information of the same user. n rows describe the target indicator values ​​of n users, and each column describes the target indicator information of the same feature type, which can be called a feature column. The feature column is the target indicator value of the same attribute corresponding to the target indicator information of the same column. m columns describe target indicator information of m feature types. For example, in the field of Internet of Things, each column can describe user data information of the same dimension of different users. For example, the first column describes the profile dimension information of different users, the second column describes the business dimension information of different users, the third column describes the value dimension information of different users, and the fourth column describes the risk dimension information of different users, etc. ij The element in the i-th row and j-th column of the description matrix A is the target indicator value corresponding to the j-th target indicator information of the i-th user.

[0055] Then, the target indicator information of the same feature type of different users is obtained, that is, the target indicator values ​​of the same attribute in the same column of the A matrix are obtained. The target indicator information of the same column can be described by the target indicator values ​​of the same attribute. Use the following formula (4) to convert x ij Perform normalization processing to obtain the corresponding normalized indicator characteristic value.

[0056]

[0057] Among them, x ij Describes the element in the i-th row and j-th column of matrix A, min(x j ) is the minimum value of the target indicator information in column j, max(x j ) is the maximum value in the target indicator information of column j, x' ij is the normalized indicator eigenvalue.

[0058] S14. According to the normalized indicator characteristic value, obtain the normalized indicator characteristic values ​​of other users with the same attribute that belong to the same target indicator information as the normalized indicator characteristic value, and according to the normalized indicator characteristic value with the same attribute and the normalized indicator characteristic value, count the sample proportion of the normalized indicator characteristic value in the sum of the normalized indicator characteristic value with the same attribute and the normalized indicator characteristic value.

[0059] Specifically, the embodiment of the present application is based on the same type of data of multiple users to count the sample proportion of each indicator information of a certain user, and then to create a user portrait for each user. Therefore, in order to measure the importance of a certain user's target indicator information in the user portrait, and then to judge the importance of the user among all users, the normalized indicator characteristic value is obtained according to the normalized indicator characteristic value, and the normalized indicator characteristic value of the same attribute of other users belonging to the same target indicator information as the normalized indicator characteristic value is obtained, and according to the normalized indicator characteristic value of the same attribute and the normalized indicator characteristic value, the proportion of the normalized indicator characteristic value in the sum of the normalized indicator characteristic value of the same attribute and the normalized indicator characteristic value is counted to obtain the target indicator information. The importance of information in the same feature category data, wherein the indicator information of other users can be obtained together with the indicator information of the user, such as the case shown in the example matrix A, the indicator information of multiple users is obtained at the same time, and then assembled into matrix A, and then a user profile is made for each user based on the indicator information of multiple users. The indicator information of other users can also be pre-stored. This situation can be a situation where a new user is profiled based on the indicator information of other user portraits that have been made before. For example, the indicator information of a certain user can be obtained, and then the indicator information of the user is added to the original matrix to construct matrix A, and then a user profile is made for the user based on the indicator information of matrix A. For example, the normalized indicator characteristic value A1 corresponding to the number of online users of A, the normalized indicator characteristic value of the same attribute corresponding to the number of online users of B is B1, and the normalized indicator characteristic value of the same attribute corresponding to the number of online users of C is C1. The importance of A1 in the process of profiling A can be obtained by counting the sample proportions of A1 in A1, B1 and C1. For the matrix A in the above example, for the element x in the i-th row and j-th column of matrix A, ij , element x (i-1)j With element x (i+1)j That is the element x ij The normalized indicator characteristic value of the same attribute of the same target indicator information, that is, the value of the target indicator information in the same column, that is, the sample proportion of the i-th record in the j-th column under the j-th target indicator information, can be calculated by the following formula (4):

[0060]

[0061] Among them, P ij The element x' in row i and column j is ij The proportion of samples in the jth column, and the proportion of samples of each target indicator information is counted in turn.

[0062] S15. According to the sample proportion, the coefficient of variation generated by the indicator value during the normalization process is calculated, and according to the coefficient of variation, the indicator weight of the target indicator information in the user portrait scoring process is obtained.

[0063] Specifically, since the normalization process is a data variation process, the indicator values ​​produce data variation during the normalization process. Therefore, according to the sample proportion, the coefficient of variation of the indicator values ​​produced during the normalization process is calculated, and according to the coefficient of variation, the indicator weight of the target indicator information in the user portrait scoring statistics process is obtained. For example, for the matrix A in the above example, for the element x in the i-th row and j-th column, ij Converted into indicator eigenvalue x' ij The variation generated in the process, according to x' ij The proportion of samples in the jth column P ij , x' can be calculated by the following formula (5) ij The coefficient of variation g j ,

[0064]

[0065] Then, calculate x' using the following formula (6): ij The indicator weight Wj:

[0066]

[0067] And calculate the X' of each target indicator information in different dimensions ij The indicator weight W j .

[0068] S16. Multiply the normalized indicator feature value corresponding to each target indicator information of the user and the indicator weight, and add the product results to obtain a user portrait score for the user.

[0069] Specifically, the normalized indicator feature value corresponding to each target indicator information of the user and the indicator weight corresponding to each normalized indicator feature value are multiplied, and the product results are added to obtain the user profile score for the user. For example, when scoring the user profile of an IoT user, the scoring result is as follows:

[0070] Comprehensive score = 0.40 * value score + 0.23 * business score - 0.16 * risk score + 0.10 * behavior score + 0.07 * overview score + 0.04 * cooperation score;

[0071] Among them, overview score = 0.11 * basic information + 0.2 * risk information + 0.45 * asset information + 0.24 * operating conditions;

[0072] Basic information = 0.5*number of online users + 0.33*industry + 0.15*customer category + 0.04*establishment time + 0.04*employee size;

[0073] Asset information = 0.65*registered capital + 0.283*number of patents + 0.07*number of trademarks;

[0074] Risk information = 0.41 * information on dishonest individuals + 0.38 * number of serious violations + 0.15 * number of administrative penalties + 0.06 * number of abnormal operations;

[0075] Business status = 0.1 * latest tax rating + 0.638 * number of products + 0.25 * number of qualification certificates;

[0076] The same process can be applied to obtain a user portrait score, and the scores of each dimension involved in the above user portrait score and the comprehensive score can be displayed in a graphical form, for example, as a bar chart, so that users can understand the user portrait more intuitively.

[0077] In an embodiment of the present application, multi-dimensional indicator information of a user is obtained and digitized to obtain indicator values. Based on the indicator values, the similarity of information features between different indicator information is statistically analyzed. Based on the similarity of information features, redundant indicator information is removed to obtain target indicator information. The target indicator information is then normalized to remove the influence of different dimensions, and the sample proportion of the target indicator information is statistically analyzed. Based on the sample proportion and the coefficient of variation, the indicator weight of the target indicator information in the user portrait scoring process is obtained. Finally, the normalized indicator feature value corresponding to each target indicator information of the user and the indicator weight are multiplied, and the product results are added to obtain a user portrait score for the user. In this way, in the process of user portrait, the characteristics of multiple dimensions of the user can be accurately and comprehensively measured in a quantitative manner, and redundant indicator information is removed, fully ensuring the independence of the target indicator information. Furthermore, the information contained in the user data of multiple dimensions can be objectively and fully comprehensively mined to obtain a more accurate, objective, and comprehensive portrait score for the user, thereby improving the accuracy of the user portrait and avoiding the bias caused by personal subjective tendencies in user evaluation.

[0078] In one embodiment, see Figure 5 , Figure 5 This is a schematic diagram of the fourth sub-process of the user portrait scoring method provided in the embodiment of the present application, as shown in FIG. Figure 5As shown, in this embodiment, other users include multiple users, and after multiplying the normalized indicator feature value corresponding to each target indicator information of the user and the indicator weight, and adding the product results to obtain a user portrait score for the user, it also includes:

[0079] S17. Obtain the user profile score corresponding to each other user's respective user profile;

[0080] S18. Display the user portrait scores of the user and other users.

[0081] Specifically, when scoring a user portrait based on the user portrait scoring method of the present application, the original data information corresponding to the preset initial indicator information of different dimensions of a user can be obtained. However, since the sample proportion of the statistical indicator information involves the same attribute indicator information of other users, the original data information corresponding to the preset initial indicator information of different dimensions of multiple users can also be obtained, and a matrix A similar to the above example can be constructed based on the indicator information of multiple users. In this way, the portraits of multiple users can be scored in parallel. After obtaining the user portrait score of the user portrait, the user portrait score of each other user can also be obtained. Then, the user portrait scores of all users are displayed at the same time to achieve parallel scoring of multiple user portraits. Compared with traditional technologies, when a user is evaluated based on a large number of original indicator information, due to the large number of original indicators, it can only be performed on a single user when observing the original indicators, and it may take a long time from observing the indicators to drawing a user evaluation conclusion. The present application provides a comprehensive user portrait score for the user portrait, which can realize the user portrait evaluation results of multiple users at a time, improve the efficiency of user evaluation, and can simply and clearly understand the strengths and weaknesses of each user by understanding the scores of multiple user portraits at the same time.

[0082] Furthermore, the user portrait scores of the user and other users are displayed, including:

[0083] The user portrait scores of the user and other users are sorted and displayed in a preset scoring order.

[0084] Specifically, the user portrait scores of the user and other users are sorted in a preset scoring order, for example, they can be sorted in order from high to low, and then the sorted user portrait scores are displayed, which can achieve the ranking of user portraits. Compared with the traditional technology that uses original indicators to make it difficult to achieve direct and quantitative comparison of the quality of users, the comprehensive score of the user portrait provided by the embodiment of the present application enables relevant personnel to directly compare users based on the sorting results of the user portraits, and guide or adjust the focus of work based on the comparison results of the user portraits, which can improve work efficiency and save work costs. Furthermore, according to actual business needs, the user portrait scores of the user and other users can be displayed in other ways, such as highlighting users and user portrait scores that meet the preset display conditions, or displaying users and user portrait scores that meet the preset display conditions, while other users and user portrait scores that do not meet the preset display conditions are not displayed, or users and user portrait scores that meet different preset display conditions are classified and displayed, etc.

[0085] It should be noted that the user portrait scoring method described in the above embodiments can recombine the technical features contained in different embodiments as needed to obtain a combined implementation plan, but all of them are within the scope of protection required by this application.

[0086] See also Figure 6 , Figure 6 This is a schematic block diagram of a user portrait scoring device provided in an embodiment of the present application. Corresponding to the above-mentioned user portrait scoring method, an embodiment of the present application also provides a user portrait scoring device. Figure 6 As shown, the user portrait scoring device includes a unit for executing the above-mentioned user portrait scoring method. The user portrait scoring device can be configured in a computer device and can be applied to user portraits in the field of the Internet of Things. Figure 6 The user portrait scoring device 60 includes a digitization unit 61, a first statistical unit 62, a normalization unit 63, a second statistical unit 64, a first acquisition unit 65 and a calculation unit 66.

[0087] The digitization unit 61 is used to obtain raw data information corresponding to preset initial indicator information of different dimensions of the same user, and digitize the raw data information to obtain indicator values ​​corresponding to the raw data information;

[0088] A first statistical unit 62 is configured to calculate information feature similarities between different preset initial indicator information based on the indicator values, and remove redundant preset initial indicator information based on the information feature similarities to obtain target indicator information;

[0089] A normalization unit 63 is configured to normalize the target indicator value corresponding to the target indicator information to obtain a normalized indicator characteristic value;

[0090] A second statistical unit 64 is configured to obtain, based on the normalized indicator characteristic value, normalized indicator characteristic values ​​of the same attribute of other users belonging to the same target indicator information as the normalized indicator characteristic value, and to calculate, based on the normalized indicator characteristic value of the same attribute and the normalized indicator characteristic value, a sample proportion of the normalized indicator characteristic value in the sum of the normalized indicator characteristic value of the same attribute and the normalized indicator characteristic value;

[0091] A first obtaining unit 65 is configured to calculate a coefficient of variation of the indicator value generated during the normalization process based on the sample proportion, and obtain the indicator weight of the target indicator information in the user profile scoring process based on the coefficient of variation;

[0092] The calculation unit 66 is used to multiply the normalized indicator feature value corresponding to each target indicator information of the user and the indicator weight, and add the product results to obtain a user portrait score for the user.

[0093] In one embodiment, the user dimensions include profile dimension information, business dimension information, value dimension information, risk dimension information, behavior dimension information, and cooperation dimension information.

[0094] In one embodiment, the digitization unit 61 includes:

[0095] A digitization subunit, configured to digitize the raw data information to obtain an initial indicator value;

[0096] The standardization subunit is used to standardize the initial indicator value to obtain the indicator value corresponding to the preset initial indicator information.

[0097] In one embodiment, the first statistical unit 62 includes:

[0098] a first acquisition subunit, configured to obtain, according to the indicator value, arbitrarily two pieces of preset initial indicator information; if one of the pieces is the preset first initial indicator information and the other is the preset second initial indicator information, calculate information feature similarity between the preset first initial indicator information and the preset second initial indicator information, and obtain an information feature similarity value corresponding to the information feature similarity;

[0099] a determination subunit, configured to use the preset first initial indicator information and the preset second initial indicator information as similarity indicator information if the information feature similarity value is greater than or equal to a preset similarity threshold;

[0100] a second obtaining subunit, configured to obtain other information feature similarity values ​​corresponding to the first initial indicator information and all other preset initial indicator information, and average all of the other information feature similarity values ​​to obtain an average first information feature similarity value corresponding to the first initial indicator information, and obtain an average second information feature similarity value corresponding to the second initial indicator information;

[0101] The comparison subunit is used to compare the average similarity value of the first information feature with the average similarity value of the second information feature, obtain the initial indicator information of the information feature with the smaller average similarity value between the first information feature and the second information feature, and obtain the target indicator information.

[0102] In one embodiment, the normalization unit 63 includes:

[0103] A construction subunit is used to construct a target indicator information matrix using target indicator values ​​of target indicator information of different dimensions of the same user as rows and target indicator values ​​of target indicator information of the same dimension of different users as columns;

[0104] The normalization subunit is used to perform normalization calculation on the column values ​​in each column based on the maximum value and the minimum value contained in each column of the target indicator information matrix to obtain a normalized indicator characteristic value.

[0105] In one embodiment, the other users include multiple users, and the user portrait scoring device 60 further includes:

[0106] The second obtaining unit is used to obtain the user portrait score corresponding to each user portrait of other users;

[0107] The display unit is used to display the user portrait scores of the user and other users.

[0108] In one embodiment, the display unit is specifically configured to sort and display the user portrait scores of the user and other users in a preset scoring order.

[0109] It should be noted that those skilled in the art can clearly understand that the specific implementation process of the above-mentioned user portrait scoring device and each unit can refer to the corresponding description in the aforementioned method embodiment. For the convenience and brevity of the description, it will not be repeated here.

[0110] At the same time, the division and connection methods of the various units in the above-mentioned user portrait scoring device are only used for illustration. In other embodiments, the user portrait scoring device can be divided into different units as needed, and the various units in the user portrait scoring device can also be connected in different orders and methods to complete all or part of the functions of the above-mentioned user portrait scoring device.

[0111] The above user portrait scoring device can be implemented in the form of a computer program. The computer program can be used in Figure 7 Runs on the computer equipment shown.

[0112] See also Figure 7 , Figure 7 1 is a schematic block diagram of a computer device provided in an embodiment of the present application. The computer device 500 may be a computer device such as a terminal, or a component or part in other devices.

[0113] See Figure 7 The computer device 500 includes a processor 502, a memory and a network interface 505 connected through a system bus 501, wherein the memory may include a non-volatile storage medium 503 and an internal memory 504, and the memory may also be a volatile storage medium.

[0114] The non-volatile storage medium 503 can store an operating system 5031 and a computer program 5032. When the computer program 5032 is executed, the processor 502 can execute the above-mentioned user portrait scoring method.

[0115] The processor 502 is used to provide computing and control capabilities to support the operation of the entire computer device 500.

[0116] The internal memory 504 provides an environment for the operation of the computer program 5032 in the non-volatile storage medium 503. When the computer program 5032 is executed by the processor 502, the processor 502 can execute the above-mentioned user portrait scoring method.

[0117] The network interface 505 is used to communicate with other devices through the network. Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device 500 to which the solution of the present application is applied. The specific computer device 500 may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components. For example, in some embodiments, the computer device may only include a memory and a processor. In such an embodiment, the structure and function of the memory and processor are the same as those in the embodiment of the present application. Figure 7 The embodiments shown are consistent and will not be described again here.

[0118] The processor 502 is configured to execute a computer program 5032 stored in a memory to implement the following steps: obtaining raw data information corresponding to preset initial indicator information of different dimensions of the same user, and digitizing the raw data information to obtain indicator values ​​corresponding to the raw data information; based on the indicator values, statistically analyzing the similarities in information features between different preset initial indicator information, and based on the similarities in information features, removing redundant preset initial indicator information to obtain target indicator information; normalizing the target indicator values ​​corresponding to the target indicator information to obtain normalized indicator feature values; and based on the normalized indicator feature values, obtaining the information features of other users that are similar to the normalized indicator feature values. The eigenvalue belongs to the normalized indicator characteristic value of the same attribute of the same target indicator information, and according to the normalized indicator characteristic value of the same attribute and the normalized indicator characteristic value, the sample proportion of the normalized indicator characteristic value in the sum of the normalized indicator characteristic value of the same attribute and the normalized indicator characteristic value is counted; according to the sample proportion, the coefficient of variation generated by the indicator value in the normalization process is counted, and according to the coefficient of variation, the indicator weight of the target indicator information in the user portrait scoring process is obtained; the normalized indicator characteristic value corresponding to each item of the target indicator information of the user and the indicator weight are multiplied, and the multiplication results are added to obtain the user portrait score for the user.

[0119] In one embodiment, when the processor 502 obtains the original data information corresponding to the preset initial indicator information of different dimensions of the same user, the dimensions of the user include profile dimension information, business dimension information, value dimension information, risk dimension information, behavior dimension information and cooperation dimension information.

[0120] In one embodiment, when the processor 502 converts the raw data information into numerical values ​​to obtain the indicator value corresponding to the preset initial indicator information, the processor 502 specifically implements:

[0121] Numerizing the raw data information to obtain initial indicator values;

[0122] The initial indicator value is normalized to obtain the indicator value corresponding to the preset initial indicator information.

[0123] In one embodiment, when the processor 502 implements counting the information feature similarities between different preset initial indicator information based on the indicator value, and removing redundant preset initial indicator information based on the information feature similarities to obtain the target indicator information, it specifically implements:

[0124] According to the indicator value, arbitrarily obtain two preset initial indicator information, if one of them is the preset first initial indicator information and the other is the preset second initial indicator information, calculate the information feature similarity between the preset first initial indicator information and the preset second initial indicator information, and obtain the information feature similarity value corresponding to the information feature similarity;

[0125] If the information feature similarity value is greater than or equal to a preset similarity threshold, the preset first initial indicator information and the preset second initial indicator information are used as similarity indicator information;

[0126] Obtaining other information feature similarity values ​​corresponding to the first initial indicator information and all other preset initial indicator information, and averaging all the other information feature similarity values ​​to obtain a first information feature average similarity value corresponding to the first initial indicator information, and obtaining a second information feature average similarity value corresponding to the second initial indicator information;

[0127] Compare the first information feature average similarity value and the second information feature average similarity value, obtain initial indicator information of the information feature average similarity value with the smaller one between the first information feature average similarity value and the second information feature average similarity value, and obtain target indicator information.

[0128] In one embodiment, when the processor 502 normalizes the target indicator value corresponding to the target indicator information to obtain the normalized indicator characteristic value, the processor 502 specifically implements:

[0129] The target indicator information matrix is ​​constructed by taking the target indicator values ​​of target indicator information of different dimensions of the same user as rows and taking the target indicator values ​​of target indicator information of the same dimension of different users as columns;

[0130] Based on the maximum value and the minimum value contained in each column of the target indicator information matrix, the column values ​​in each column are normalized to obtain the normalized indicator characteristic value.

[0131] In one embodiment, the other users include multiple users. After multiplying the normalized indicator feature value corresponding to each of the target indicator information of the user and the indicator weight and adding the product results to obtain a user portrait score for the user, the processor 502 further implements:

[0132] Get the user profile score corresponding to each other user's respective user profile;

[0133] The user portrait scores of the user and other users are displayed.

[0134] In one embodiment, when the processor 502 displays the user portrait scores of the user and other users, it specifically implements:

[0135] The user portrait scores of the user and other users are sorted and displayed in a preset scoring order.

[0136] It should be understood that in the embodiment of the present application, the processor 502 may be a central processing unit (CPU), and the processor 502 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0137] Those skilled in the art will appreciate that all or part of the steps in the method of the above-described embodiment can be implemented by a computer program, which can be stored in a computer-readable storage medium. The computer program is executed by at least one processor in the computer system to implement the steps in the method of the above-described embodiment.

[0138] Therefore, the present application also provides a computer-readable storage medium. The computer-readable storage medium may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, and the computer-readable storage medium stores a computer program that, when executed by a processor, causes the processor to perform the following steps:

[0139] A computer program product, when running on a computer, enables the computer to execute the steps of the user portrait scoring method described in the above embodiments.

[0140] The computer-readable storage medium may be an internal storage unit of the aforementioned device, such as a hard disk or memory of the device. The computer-readable storage medium may also be an external storage device of the device, such as a plug-in hard disk equipped on the device, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. Furthermore, the computer-readable storage medium may include both an internal storage unit of the device and an external storage device.

[0141] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0142] The storage medium is a physical, non-transient storage medium, for example, it can be a USB flash drive, a mobile hard disk, a read-only memory (ROM), a magnetic disk or an optical disk, or other physical storage medium that can store computer programs.

[0143] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0144] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of each unit is merely a logical functional division, and other division methods may be used in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not implemented.

[0145] The steps in the method of the embodiment of the present application can be adjusted in order, combined, and deleted according to actual needs. The units in the device of the embodiment of the present application can be combined, divided, and deleted according to actual needs. In addition, the functional units in the various embodiments of the present application can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into a single unit.

[0146] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the existing technology, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling an electronic device (which can be a personal computer, terminal, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application.

[0147] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present application, and such modifications or substitutions should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A user portrait scoring method, characterized in that: The method comprises: Obtaining raw data information corresponding to preset initial indicator information of different dimensions for the same user, and digitizing the raw data information to obtain indicator values ​​corresponding to the raw data information; According to the indicator values, statistically analyzing the similarities of information features between different preset initial indicator information, and removing redundant preset initial indicator information based on the similarities of the information features to obtain target indicator information; Normalizing the target indicator value corresponding to the target indicator information to obtain a normalized indicator characteristic value; According to the normalized indicator characteristic value, obtain normalized indicator characteristic values ​​of the same attribute of other users that belong to the same target indicator information as the normalized indicator characteristic value, and according to the normalized indicator characteristic value of the same attribute and the normalized indicator characteristic value, calculate the sample proportion of the normalized indicator characteristic value in the sum of the normalized indicator characteristic value of the same attribute and the normalized indicator characteristic value; According to the sample proportion, the coefficient of variation of the indicator value generated during the normalization process is calculated, and according to the coefficient of variation, the indicator weight of the target indicator information in the user portrait scoring process is obtained; Multiplying the normalized indicator feature value corresponding to each target indicator information of the user and the indicator weight, and adding the product results to obtain a user portrait score for the user; According to the indicator values, statistically analyzing the similarities of information features between different preset initial indicator information, and removing redundant preset initial indicator information based on the similarities of the information features to obtain target indicator information, including: According to the indicator value, arbitrarily obtain two preset initial indicator information, if one of them is the preset first initial indicator information and the other is the preset second initial indicator information, calculate the information feature similarity between the preset first initial indicator information and the preset second initial indicator information, and obtain the information feature similarity value corresponding to the information feature similarity; If the information feature similarity value is greater than or equal to a preset similarity threshold, the preset first initial indicator information and the preset second initial indicator information are used as similarity indicator information; Obtaining other information feature similarity values ​​corresponding to the first initial indicator information and all other preset initial indicator information, and averaging all the other information feature similarity values ​​to obtain a first information feature average similarity value corresponding to the first initial indicator information, and obtaining a second information feature average similarity value corresponding to the second initial indicator information; Compare the first information feature average similarity value and the second information feature average similarity value, obtain initial indicator information of the information feature average similarity value with the smaller one between the first information feature average similarity value and the second information feature average similarity value, and obtain target indicator information.

2. The user portrait scoring method according to claim 1, characterized in that: The user dimensions include profile dimension information, business dimension information, value dimension information, risk dimension information, behavior dimension information and cooperation dimension information.

3. The user portrait scoring method according to claim 1, characterized in that: The raw data information is digitized to obtain the indicator value corresponding to the preset initial indicator information, including: Numerizing the raw data information to obtain initial indicator values; The initial indicator value is normalized to obtain the indicator value corresponding to the preset initial indicator information.

4. The user portrait scoring method according to claim 1, characterized in that: Normalizing the target indicator value corresponding to the target indicator information to obtain a normalized indicator characteristic value includes: The target indicator information matrix is ​​constructed by taking the target indicator values ​​of target indicator information of different dimensions of the same user as rows and taking the target indicator values ​​of target indicator information of the same dimension of different users as columns; Based on the maximum value and the minimum value contained in each column of the target indicator information matrix, the column values ​​in each column are normalized to obtain the normalized indicator characteristic value.

5. The user portrait scoring method according to claim 1, characterized in that: Other users include multiple users. After multiplying the normalized indicator feature value corresponding to each target indicator information of the user and the indicator weight, and adding the product results to obtain a user portrait score for the user, the following is also included: Get the user profile score corresponding to each other user's respective user profile; The user portrait scores of the user and other users are displayed.

6. The user portrait scoring method according to claim 5, characterized in that: Display the user profile scores of the user and other users, including: The user portrait scores of the user and other users are sorted and displayed in a preset scoring order.

7. A user portrait scoring device, characterized in that: The device comprises: A digitization unit is used to obtain raw data information corresponding to preset initial indicator information of different dimensions of the same user, and digitize the raw data information to obtain indicator values ​​corresponding to the raw data information; a first statistical unit, configured to calculate information feature similarities between different preset initial indicator information based on the indicator values, and remove redundant preset initial indicator information based on the information feature similarities to obtain target indicator information; a normalization unit, configured to normalize the target indicator value corresponding to the target indicator information to obtain a normalized indicator characteristic value; A second statistical unit is configured to obtain, based on the normalized indicator characteristic value, normalized indicator characteristic values ​​of the same attribute of other users that belong to the same target indicator information as the normalized indicator characteristic value, and, based on the normalized indicator characteristic value of the same attribute and the normalized indicator characteristic value, calculate a sample proportion of the normalized indicator characteristic value in the sum of the normalized indicator characteristic value of the same attribute and the normalized indicator characteristic value; A first acquisition unit is configured to calculate a coefficient of variation of the indicator value generated during the normalization process based on the sample proportion, and obtain, based on the coefficient of variation, an indicator weight of the target indicator information in the user profile scoring process; A calculation unit, configured to multiply the normalized indicator feature value corresponding to each target indicator information of the user and the indicator weight, and add the product results to obtain a user profile score for the user; The first statistical unit includes: a first acquisition subunit, configured to obtain, according to the indicator value, arbitrarily two pieces of preset initial indicator information; if one of the pieces is the preset first initial indicator information and the other is the preset second initial indicator information, calculate information feature similarity between the preset first initial indicator information and the preset second initial indicator information, and obtain an information feature similarity value corresponding to the information feature similarity; a determination subunit, configured to use the preset first initial indicator information and the preset second initial indicator information as similarity indicator information if the information feature similarity value is greater than or equal to a preset similarity threshold; a second obtaining subunit, configured to obtain other information feature similarity values ​​corresponding to the first initial indicator information and all other preset initial indicator information, and average all of the other information feature similarity values ​​to obtain an average first information feature similarity value corresponding to the first initial indicator information, and obtain an average second information feature similarity value corresponding to the second initial indicator information; The comparison subunit is used to compare the average similarity value of the first information feature with the average similarity value of the second information feature, obtain the initial indicator information of the information feature with the smaller average similarity value between the first information feature and the second information feature, and obtain the target indicator information.

8. A computer device, characterized in that: The computer device includes a memory and a processor connected to the memory; the memory is used to store a computer program; and the processor is used to run the computer program to perform the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that The storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 can be implemented.

Images