Client portrait differentiation feature detection method and device, equipment, storage medium and program product

Abstract

The embodiment of the invention provides a customer portrait differentiation feature detection method and device, equipment, a storage medium and a program product. Belongs to the technical field of big data. Comprising the steps of obtaining at least one to-be-processed label corresponding to a target customer identifier; determining a to-be-input matrix according to the to-be-processed labels corresponding to the customer identifiers; determining a first abnormal score corresponding to a vector in the to-be-input matrix and a first influence value corresponding to the label value by adopting an isolated forest algorithm; determining a second abnormal score corresponding to the vector in the matrix to be input and a second influence value corresponding to the tag value by adopting a local outlier factor algorithm; and according to the first abnormal score, the second abnormal score, a first influence value corresponding to the target label value under the target vector and a second influence value corresponding to the target label value, determining differentiation features of the customer portrait corresponding to the target customer identifier. The method solves the problem that the features in the customer portrait are difficult to reflect the differentiated features of the customer.

Description

Technical Field

[0001] This application relates to the field of big data technology, and in particular to a method, apparatus, equipment, storage medium, and program product for detecting differentiated features of customer profiles. Background Technology

[0002] Currently, in various business operations, in order to better serve customers, it is necessary to mark, analyze, or process customer characteristics.

[0003] Currently, in order to more accurately reflect user characteristics, customer profiles are usually created to reflect customer features.

[0004] However, customer profiles in related technologies typically only reflect users' general characteristics and fail to highlight the significant differences between users. Summary of the Invention

[0005] This application provides a method, apparatus, device, storage medium, and program product for detecting differentiated features of customer profiles, in order to solve the problem that customer profiles can only reflect the general characteristics of users and are difficult to highlight the obvious differences between users.

[0006] In a first aspect, embodiments of this application provide a method for detecting differentiated features of a customer profile, comprising: acquiring at least one unprocessed label corresponding to a target customer identifier; determining an input matrix based on the unprocessed labels corresponding to each customer identifier, wherein the input matrix consists of vectors corresponding to customer identifiers, and the vectors consist of label values; using an isolated forest algorithm to determine a first anomaly score corresponding to a vector in the input matrix and a first influence value corresponding to a label value; using a local outlier factor algorithm to determine a second anomaly score corresponding to a vector in the input matrix and a second influence value corresponding to a label value; and determining differentiated features of the customer profile corresponding to the target customer identifier based on the first anomaly score corresponding to the target vector, the second anomaly score corresponding to the target vector, the first influence value corresponding to the target label value under the target vector, and the second influence value corresponding to the target label value, wherein the target customer identifier corresponds to the target vector.

[0007] In one possible implementation, the differential features of the customer profile corresponding to the target customer identifier are determined based on the first abnormal score corresponding to the target vector, the second abnormal score corresponding to the target vector, the first influence value corresponding to the target label value under the target vector, and the second influence value corresponding to the target label value. This includes: if the first influence value or the second influence value corresponding to the target label value is greater than the influence value threshold, then the target label value is determined as an abnormal label value; the probability of occurrence of the abnormal label value or combination of abnormal label values ​​in the input matrix is ​​determined; based on the probability of occurrence, the sparsity coefficient corresponding to the abnormal label value or combination of abnormal label values ​​is determined; the first abnormal score of the target vector is used as the weight of the first influence value corresponding to the abnormal label value, and the second abnormal score of the target vector is used as the weight of the second influence value corresponding to the abnormal label value; the first influence value and the second influence value are weighted and summed to obtain the comprehensive influence value of the abnormal label value; the comprehensive influence value of the abnormal label value is multiplied by the sparsity coefficient of the abnormal label value to obtain the differential score of the abnormal label value; and the differential features of the customer profile corresponding to the target customer identifier are determined based on the differential score of the abnormal label value.

[0008] In one possible implementation, the differential features of the customer profile corresponding to the target customer identifier are determined based on the differential scores of the abnormal label values. This includes: if the differential score corresponding to the abnormal label value of the target customer identifier is greater than or equal to a preset differential score threshold, then the label to be processed corresponding to the abnormal label value is determined as a differential feature of the customer profile corresponding to the target customer identifier; or, the number of label values ​​of abnormal label values ​​under the target customer identifier whose differential scores are greater than or equal to the differential score threshold is determined; if the number of label values ​​is greater than or equal to a preset number threshold, then the labels to be processed corresponding to the M abnormal label values ​​with the largest differential scores are determined as differential features of the customer profile corresponding to the target customer identifier, where M is a positive integer; or, if the differential scores corresponding to all abnormal label values ​​under the target customer identifier are less than the differential score threshold, then the labels to be processed corresponding to the N abnormal label values ​​with the largest differential scores corresponding to the target customer identifier are determined as differential features of the customer profile corresponding to the target customer identifier, where N is a positive integer.

[0009] In one possible implementation, after determining the differentiated features of the customer profile corresponding to the target customer identifier based on the first abnormal score corresponding to the target vector, the second abnormal score corresponding to the target vector, the first influence value corresponding to the target label value under the target vector, and the second influence value corresponding to the target label value, the method further includes: obtaining the business processing result corresponding to the target customer identifier; and changing the differentiated score threshold based on the business processing result.

[0010] In one possible implementation, the sparsity coefficient corresponding to the abnormal label value or combination of abnormal label values ​​is determined based on the occurrence probability, including: subtracting the occurrence probability from a preset value to obtain the sparsity coefficient corresponding to the abnormal label value or combination of abnormal label values; or, finding the correspondence between a preset occurrence probability interval and the sparsity coefficient based on the occurrence probability to obtain the sparsity coefficient corresponding to the abnormal label value or combination of abnormal label values.

[0011] In one possible implementation, the Isolation Forest algorithm is used to determine the first anomaly score corresponding to a vector in the input matrix and the first influence value corresponding to the label value. This includes: inputting the input matrix into the Isolation Forest algorithm to obtain multiple binary trees based on the label values, where each binary tree consists of vectors as nodes; if the target label type corresponding to the target label value is used, the target vector is divided into leaf nodes of the binary trees, and the number of partitions corresponding to the target label type under the target vector is increased by K, where K is a positive integer; the number of partitions of the target label type under the target vector in each binary tree is added together to obtain the total number of partitions of the target label type under the target vector; the first influence value corresponding to the target label type is determined based on the total number of partitions of the target label type; the average number of edges from the target vector to the root node of each binary tree is determined as the target edge number; and the first anomaly score is determined based on the target edge number.

[0012] In one possible implementation, a local outlier factor algorithm is used to determine the second outlier score corresponding to the vector in the input matrix and the second influence value corresponding to the label value. This includes: inputting the input matrix into the local outlier factor algorithm to obtain the second outlier score corresponding to each vector in the input matrix output by the algorithm; identifying vectors with second outlier scores greater than an outlier score threshold as outlier vectors; replacing the target label value in the outlier vector with a preset label value, and then re-executing the step of inputting the input matrix into the local outlier factor algorithm to obtain a new outlier factor corresponding to the outlier vector, wherein the preset label value corresponds to the label type of the target label value; and determining the second influence value of the target label value based on the second outlier score corresponding to the outlier vector and the new outlier factor.

[0013] Secondly, embodiments of this application provide a customer profile differential feature detection device, comprising:

[0014] The module includes a tag acquisition module for acquiring at least one tag to be processed corresponding to a target customer identifier; a matrix determination module for determining an input matrix based on the tags to be processed corresponding to each customer identifier, wherein the input matrix consists of vectors corresponding to customer identifiers, and each vector consists of tag values; a first determination module for determining the first anomaly score corresponding to a vector in the input matrix and the first influence value corresponding to a tag value using an isolated forest algorithm; a second determination module for determining the second anomaly score corresponding to a vector in the input matrix and the second influence value corresponding to a tag value using a local outlier factor algorithm; and a feature determination module for determining the differentiated features of the customer profile corresponding to the target customer identifier based on the first anomaly score corresponding to the target vector, the second anomaly score corresponding to the target vector, the first influence value corresponding to the target tag value under the target vector, and the second influence value corresponding to the target tag value, wherein the target customer identifier corresponds to the target vector.

[0015] Thirdly, embodiments of this application provide an electronic device, including: a memory and a processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0016] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0017] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0018] The customer profile differentiation feature detection method, apparatus, device, storage medium, and program product provided in this application embodiment obtain the first and second anomaly scores of the vector corresponding to the customer, as well as the first and second influence values ​​of the customer's tag, by adopting the isolated forest algorithm and the local outlier factor algorithm. Based on the first and second influence values, the differentiation features of the customer profile corresponding to the target customer identifier are obtained, thereby highlighting the characteristics corresponding to the customer identifier. Attached Figure Description

[0019] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0020] Figure 1 A schematic diagram illustrating a scenario for the differential feature detection method for customer profiles provided in this application;

[0021] Figure 2 A flowchart illustrating the differential feature detection method for customer profiles provided in this application embodiment;

[0022] Figure 3 This is a schematic diagram illustrating the usage process of the isolated forest algorithm provided in the embodiments of this application;

[0023] Figure 4 A schematic diagram of the differential feature detection device for customer profiles provided in this application embodiment;

[0024] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0025] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0026] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0027] In today's era of massive digital business processing, customer characteristics are the foundation for providing high-quality services.

[0028] Current customer profiles are typically constructed using only fixed dimensions, making it difficult to reflect the unique characteristics of customers.

[0029] To address the aforementioned technical problems, the inventors propose the following technical concept: By constructing vectors corresponding to customers using user tags, and then forming an input matrix from these vectors corresponding to different customers, the Isolation Forest algorithm is used to determine the first outlier score and the first influence value corresponding to the tag in the input matrix. The Local Outlier Factor algorithm is used to determine the second outlier score and the second influence value corresponding to the tag in the input matrix. Combining the first outlier score, the first influence value, the second outlier score, and the second influence value of the target vector, the differentiated features corresponding to the target customer identifier are determined.

[0030] This application is applied to scenarios involving the detection of differentiated features in customer profiles. It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with relevant laws, regulations and standards, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0031] Figure 1 A schematic diagram illustrating a scenario for the differential feature detection method for customer profiles provided in this application. (Example) Figure 1 This scenario includes: server 101 and data storage unit 102.

[0032] In the specific implementation process, server 101 can be implemented using a cluster of one or more servers with more powerful processing capabilities and higher security. Where possible, it can also be replaced by computers, laptops, etc. with strong computing power.

[0033] The data storage unit 102 may include a single database or a combination of multiple databases. For example, it may include one or more databases such as Oracle, MySQL, DRDS (Distributed Relational Database Service), and ES (Elasticsearch). This application does not impose any special restrictions on this.

[0034] The connection between server 101 and data storage unit 102 can be either wired or wireless.

[0035] Server 101 is used to read the tags to be processed corresponding to the customer identifier from the data storage unit 102, generate the input matrix according to the tags to be processed, use the isolated forest algorithm to determine the first outlier score corresponding to the vector in the input matrix and the first influence value corresponding to the tag, use the local outlier factor algorithm to determine the second outlier score corresponding to the vector in the input matrix and the second influence value corresponding to the tag value, and combine the first outlier score, the first influence value, the second outlier score, and the second influence value of the target vector to determine the differential features corresponding to the target customer identifier.

[0036] It is understood that the scenarios illustrated in the embodiments of this application do not constitute a specific limitation on the method for detecting differentiated features of customer profiles. In other feasible embodiments of this application, the above scenarios may include more or fewer components than illustrated, or combine some components, or split some components, or arrange different components, which can be determined according to the actual application scenario and are not limited here. Figure 1 The scenario shown can be implemented by hardware, software, or a combination of both.

[0037] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will be described below with reference to the accompanying drawings.

[0038] Figure 2 This is a flowchart illustrating the differential feature detection method for customer profiles provided in this application embodiment. The execution entity of this application embodiment may be... Figure 1 The server 101 in this embodiment can also be a computer and / or a mobile phone, etc., and this embodiment does not impose any special restrictions on it. Figure 2 As shown, the method includes:

[0039] S201: Obtain at least one tag to be processed corresponding to the target customer identifier.

[0040] In this step, the target customer identifier can be anonymized customer identifiers, and the tags to be processed include, for example, address, age, occupation type, etc.

[0041] S202: Determine the input matrix based on the labels to be processed corresponding to each customer identifier. The input matrix consists of vectors corresponding to the customer identifiers, and each vector consists of label values.

[0042] This step includes converting the tags to be processed into numerical values ​​to obtain tag values. The tag values ​​corresponding to the target customer identifier are combined to obtain the target vector corresponding to the target customer identifier. The vectors corresponding to each customer identifier are combined to obtain the input matrix.

[0043] The process can involve converting the tags to be processed into numerical values ​​using methods such as One-Hot encoding and Embedding, or by using a pre-defined mapping between tags and numerical values. This mapping can be pre-set by staff based on experimental data or empirical parameters.

[0044] S203: Using the isolated forest algorithm, determine the first anomaly score corresponding to the vector in the input matrix and the first influence value corresponding to the label value.

[0045] This step involves randomly selecting any label type and a random label value to divide the vectors in the input matrix into a left subtree and a right subtree. Then, randomly selecting any label type and a random label value to divide the vectors in the left and right subtrees again, until no further division is possible. This process yields the number of edges from each vector to the root node. Based on the variables from each vector to the root node, the first anomaly score of the vector is determined. Finally, based on the number of times the vector is divided into leaf nodes according to the label value, the first influence value corresponding to the label value is determined.

[0046] Figure 3 This is a schematic diagram illustrating the usage process of the isolated forest algorithm provided in an embodiment of this application. Figure 3 As shown, there are currently 5 vectors forming the input matrix. Based on the label value A, 3 of these 5 vectors are divided into the right subtree, and 2 are divided into the left subtree. The left subtree is further divided using label B, resulting in two leaf nodes. The 3 vectors in the right subtree are further divided using label B, resulting in 2 being divided into the left subtree and 1 into the right subtree (leaf node). Label C is used to divide 1 into the left subtree (leaf node) and 1 into the right subtree (leaf node).

[0047] S204: The local outlier factor algorithm is used to determine the second outlier score corresponding to the vector in the input matrix and the second influence value corresponding to the label value.

[0048] In this step, the input matrix is ​​fed into the Local Outlier Factor Algorithm. The Local Outlier Factor Algorithm treats the vectors in the input matrix as points and calculates the outlier score of each vector, i.e., the second outlier score. By changing the label value in the vector to a preset value or setting it to zero, the outlier score is calculated again. The two outlier scores are subtracted to obtain the second influence value.

[0049] S205: Based on the first abnormal score corresponding to the target vector, the second abnormal score corresponding to the target vector, the first influence value corresponding to the target label value under the target vector, and the second influence value corresponding to the target label value, determine the differentiated features of the customer profile corresponding to the target customer identifier, wherein the target customer identifier corresponds to the target vector.

[0050] This step includes multiplying the first anomaly score by the first impact value to obtain a first product, multiplying the second anomaly score by the second impact value to obtain a second product, and adding the first product and the second product to obtain the total anomaly value. If the total anomaly value is greater than a preset anomaly value threshold, the label to be processed corresponding to the target label value is determined as a differential feature.

[0051] As can be seen from the description of the above embodiments, the embodiments of this disclosure obtain the first anomaly score and the second anomaly score of the vector corresponding to the customer, as well as the first influence value and the second influence value corresponding to the customer's tag by using the isolated forest algorithm and the local outlier factor algorithm. Based on the first influence value and the second influence value, the differentiated features of the customer profile corresponding to the target customer identifier are obtained, thereby highlighting the characteristics corresponding to the customer identifier.

[0052] In one possible implementation, after step S205 above, the method further includes: generating a customer profile corresponding to the target customer identifier based on preset tags and differentiated features.

[0053] In one possible implementation, step S205 above, based on the first anomaly score corresponding to the target vector, the second anomaly score corresponding to the target vector, the first influence value corresponding to the target label value under the target vector, and the second influence value corresponding to the target label value, determines the differentiated features of the customer profile corresponding to the target customer identifier, including:

[0054] S2051: If the first or second influence value corresponding to the target label value is greater than the influence value threshold, the target label value is determined to be an abnormal label value.

[0055] In this step, the threshold value for influence can be preset by the staff based on experimental data or empirical parameters.

[0056] S2052: Determine the probability of an outlier label value or combination of outlier label values ​​appearing in the input matrix.

[0057] In this step, the occurrence probability is obtained by counting the number of times the abnormal label value appears in all vectors and then dividing by the total number of vectors. If at least two abnormal label values ​​appear in the target vector, these at least two abnormal label values ​​are treated as a combination of abnormal label values. The occurrence probability is then obtained by counting the number of times the combination of abnormal label values ​​appears in all vectors and then dividing the number of occurrences by the total number of vectors.

[0058] S2053: Determine the sparsity coefficient corresponding to the outlier label value or combination of outlier label values ​​based on the probability of occurrence.

[0059] This step may include subtracting the occurrence probability from a preset value to obtain the sparsity coefficient; or it may include finding the correspondence between a preset occurrence probability range and the sparsity coefficient based on the occurrence probability to obtain the sparsity coefficient.

[0060] S2054: The first anomaly score of the target vector is used as the weight of the first influence value corresponding to the anomaly label value, and the second anomaly score of the target vector is used as the weight of the second influence value corresponding to the anomaly label value. The first influence value and the second influence value are weighted and summed to obtain the comprehensive influence value of the anomaly label value.

[0061] This step includes multiplying the first abnormal score by the first impact value to obtain a first product, multiplying the second abnormal score by the second impact value to obtain a second product, and adding the first product and the second product to obtain the comprehensive impact value.

[0062] S2055: Multiply the overall impact value of the abnormal label value by the sparsity coefficient of the abnormal label value to obtain the differential score of the abnormal label value.

[0063] In this step, for example, the overall impact value of the outlier label is 0.8, the sparsity of the outlier label is 0.9, and the differentiation score is 0.72. Or, for another example, if the overall impact value of the outlier label is 0.6, the sparsity is 0.5, and the differentiation score is 0.3.

[0064] S2056: Based on the differential scores of abnormal label values, determine the differential features of the customer profile corresponding to the target customer identifier.

[0065] This step includes determining the label to be processed corresponding to the target anomaly label value as a differential feature if the differential score of the target anomaly label is greater than the preset differential score threshold.

[0066] As can be seen from the description of the above embodiments, the embodiments of this disclosure determine the abnormal label values ​​in the label values ​​according to the influence value, determine the sparsity coefficient of the abnormal label or the combination of abnormal labels according to the occurrence probability of the abnormal label value or the abnormal label value in the input matrix, calculate the weighted sum of the first influence value and the second influence value corresponding to the abnormal label value using the first abnormal score and the second abnormal score as weights to obtain the comprehensive influence value, multiply the comprehensive influence value and the sparsity coefficient to obtain the differentiation score, and determine which features are the customer's differentiation features by the differentiation score, thereby combining the rarity and influence of the abnormal labels to select more accurate differentiation features.

[0067] In one possible implementation, step S2056 above, determining the differentiated features of the customer profile corresponding to the target customer identifier based on the differentiated score of the abnormal label value, includes: step S2056A, step S2056B or S2056C.

[0068] S2056A: If the differential score corresponding to the abnormal label value of the target customer identifier is greater than or equal to the preset differential score threshold, then the label to be processed corresponding to the abnormal label value is determined to be the differential feature of the customer profile corresponding to the target customer identifier.

[0069] In this step, for example, under customer identifier A, the differential score corresponding to the abnormal label value 1 is 0.8, and the differential score threshold is 0.6. Therefore, the label to be processed corresponding to the abnormal label value 1 is a differential feature. As another example, under customer identifier B, the differential score corresponding to the abnormal label value 2 is 0.85, and the differential score threshold is 0.7. Therefore, the label to be processed corresponding to the abnormal label value 2 is a differential feature.

[0070] S2056B: Determine the number of abnormal tag values ​​under the target customer identifier whose differential score is greater than or equal to the differential score threshold; if the number of tag values ​​is greater than or equal to the preset number threshold, then determine the tags to be processed corresponding to the M abnormal tag values ​​with the largest differential scores as the differential features of the customer profile corresponding to the target customer identifier, where M is a positive integer.

[0071] In this step, for example, under the current customer identifier A, if the differential scores corresponding to 8 abnormal label values ​​are all greater than the differential score threshold, then the labels to be processed corresponding to the 4 abnormal label values ​​with the largest differential scores (M = 4) are determined as the differential features of the customer profile corresponding to the target customer identifier. As another example, under the current customer identifier B, if the differential scores corresponding to 6 abnormal label values ​​are all greater than the differential score threshold, then the labels to be processed corresponding to the 3 abnormal label values ​​with the largest differential scores (M = 3) are determined as the differential features of the customer profile corresponding to the target customer identifier.

[0072] S2056C: If the differential scores corresponding to all abnormal label values ​​under the target customer identifier are less than the differential score threshold, then the labels to be processed corresponding to the N abnormal label values ​​with the largest differential scores corresponding to the target customer identifier are determined as the differential features of the customer profile corresponding to the target customer identifier, where N is a positive integer.

[0073] In this step, N, such as 1, 2, 3, can be preset by the staff.

[0074] As can be seen from the description of the above embodiments, the embodiments of this disclosure select all or the largest M abnormal label values ​​corresponding to the labels to be processed as differential features when there are many abnormal label values ​​with large differential scores, and use the N abnormal label values ​​with the largest differential scores as labels to be processed when there are no abnormal label values ​​greater than the differential score threshold.

[0075] After determining the differentiated features of the customer profile corresponding to the target customer identifier based on the first abnormal score corresponding to the target vector, the second abnormal score corresponding to the target vector, the first influence value corresponding to the target label value under the target vector, and the second influence value corresponding to the target label value in step S205, the process further includes steps S220 and S221.

[0076] S220: Obtain the business processing result corresponding to the target customer identifier.

[0077] In this step, the business processing result can include whether the business was successfully processed or failed.

[0078] S221: Change the differential score threshold based on the business processing results.

[0079] This step includes increasing the differential score threshold if the business transaction fails, and decreasing the differential score threshold if the business transaction is successful.

[0080] As can be seen from the description of the above embodiments, the embodiments of this disclosure increase the accuracy of subsequent differentiated labels by dynamically changing the differentiated score threshold according to the business processing results.

[0081] In one possible implementation, step S2053 above determines the sparsity coefficient corresponding to the abnormal label value or combination of abnormal label values ​​based on the occurrence probability, including step S2053A or step S2053B.

[0082] S2053A: Subtract the occurrence probability from the preset value to obtain the sparsity coefficient corresponding to the abnormal label value or the combination of abnormal label values.

[0083] In this step, preset values ​​are, for example, 1, 0.98, 0.9, etc.

[0084] For example, if the preset value is 1 and the probability of occurrence is 0.2, then the coefficient of the abnormal label or combination of abnormal label values ​​is 0.8. As another example, if the preset value is 0.9 and the probability of occurrence is 0.3, then the coefficient of the abnormal label or combination of abnormal label values ​​is 0.6.

[0085] S2053B: Based on the occurrence probability, find the correspondence between the preset occurrence probability interval and the sparsity coefficient to obtain the sparsity coefficient corresponding to the abnormal label value or the combination of abnormal label values.

[0086] In this step, the correspondence between the probability interval and the sparsity coefficient can be preset by the staff based on experimental data or empirical parameters.

[0087] For example, if the probability of occurrence is 0.73, the corresponding probability interval is "[0.7, 0.8]". Using "[0.7, 0.8]" to find the correspondence between the probability interval and the sparsity coefficient, the corresponding sparsity coefficient is 0.2. As another example, if the probability of occurrence is 0.553, the corresponding probability interval is "[0.5, 0.6]". Using "[0.5, 0.6]" to find the correspondence between the probability interval and the sparsity coefficient, the corresponding sparsity coefficient is 0.4.

[0088] As can be seen from the description of the above embodiments, the embodiments of this disclosure determine a suitable sparsity coefficient by the occurrence probability, so that the subsequently obtained differential features are more accurate.

[0089] In one possible implementation, step S203 above employs the isolated forest algorithm to determine the first anomaly score corresponding to the vector in the input matrix and the first influence value corresponding to the label value, including:

[0090] S2031: Input the matrix to be input into the isolated forest algorithm to obtain multiple binary trees based on the label values, where the binary trees are composed of vectors as nodes.

[0091] In this step, the Isolation Forest algorithm randomly selects a label type to be processed and randomly generates label values. These label values ​​are then used to partition the vectors in the input matrix into different nodes, resulting in a binary tree. By partitioning the input matrix multiple times, multiple binary trees are obtained.

[0092] S2032: If the target label type corresponding to the target label value is used, and the target vector is divided into leaf nodes of a binary tree, then the number of divisions corresponding to the target label type under the target vector is increased by K, where K is a positive integer.

[0093] In this step, for example, when partitioning a vector using label type A corresponding to label value A, if vector 1 is partitioned to a leaf node, then the partition count corresponding to label type A under vector 1 is increased by 1 (K is 1). Similarly, when partitioning a vector using label type B corresponding to label value B, if vector 2 is partitioned to a leaf node, then the partition count corresponding to label type B under vector 2 is increased by 2 (K is 2).

[0094] S2033: Add up the number of times the target label type is divided under the target vector in each binary tree to get the total number of times the target label type is divided under the target vector.

[0095] In this step, for example, if vector A is partitioned into target label types 2, 1, 3, 1, and 1 times in each binary tree, then the total number of partitions for the target label type under the target vector is 8. Similarly, if vector B is partitioned into target label types 1, 1, 2, 3, and 2 times in each binary tree, then the total number of partitions for the target label type under the target vector is 9.

[0096] S2034: Determine the first influence value corresponding to the target tag type based on the total number of divisions of the target tag type.

[0097] In this step, the total number of partitions can be directly used as the first influence value corresponding to the target label type. Alternatively, the total number of partitions can be divided by the number of binary trees to obtain the first influence value corresponding to the target label type. Furthermore, the total number of partitions can be mapped to a preset interval to obtain the first influence value.

[0098] For example, vector A is divided into leaf nodes of a binary tree 3 times by the target label value, and 3 is determined as the first influence value; as another example, vector B is divided into leaf nodes of a binary tree 5 times by the target label value, and 5 is mapped to a preset interval to obtain the first influence value.

[0099] The threshold for the number of divisions can be preset or calculated based on the average or mode of the total number of divisions for each tag type.

[0100] S2035: The average number of edges from the target vector to the root node of each binary tree is used to determine the target edge number.

[0101] For example Figure 3 In the vector "(A4, B4, C4, D4, E4)", the number of edges from the root node is 3, and the number of edges from the vector "(A1, B1, C1, D1, E1)" to the root node is 2.

[0102] S2036: Determine the first anomaly score based on the target number of edges.

[0103] This step may include mapping the target edge count to the interval between 0 and 1 to obtain the first anomaly score.

[0104] For example, the largest number of edges is mapped to 1, and the smallest number of edges is mapped to 0, thus mapping the target number of edges to the first abnormal score.

[0105] As can be seen from the description of the above embodiments, the embodiments of this disclosure use the isolated forest algorithm to divide the input matrix into multiple binary trees. Based on the number of times the target vector is divided into leaf nodes according to the target label type corresponding to the target label value, the influence value corresponding to that target label type is determined. The average value of the variables from the target vector to the root node of the binary tree is determined as the target edge number. The first anomaly score is determined by the target edge number. This realizes the determination of the anomaly degree corresponding to the vector and the influence degree corresponding to the label value by the isolated forest algorithm, which is convenient for subsequent determination of the user's differentiated characteristics.

[0106] In one possible implementation, step S204 above uses the local outlier factor algorithm to determine the second outlier score corresponding to the vector in the input matrix and the second influence value corresponding to the label value, including steps S2041 to S2044.

[0107] S2041: Input the matrix to be input into the local outlier factor algorithm to obtain the second outlier score corresponding to each vector in the matrix to be input output by the local outlier factor algorithm.

[0108] In this step, the local outlier factor algorithm uses each vector in the input matrix as a point to determine the second outlier score corresponding to each vector in the input matrix.

[0109] S2042: The vector with the second outlier score greater than the outlier score threshold is identified as an outlier vector.

[0110] In this step, the anomaly score threshold can be preset by staff based on experimental data or empirical parameters.

[0111] S2043: Replace the target label value in the outlier vector with the preset label value, and execute the step of inputting the matrix to be input into the local outlier factor algorithm again to obtain the new outlier factor corresponding to the outlier vector, wherein the preset label value corresponds to the label type of the target label value.

[0112] In this step, the preset label value can be a regular numerical value corresponding to the label value type of the target label value, or it can be 0 or a null value, etc. The step of inputting the matrix to be input into the local outlier factor algorithm can be step S2041 as described above.

[0113] S2044: Determine the second influence value of the target label value based on the second outlier score corresponding to the outlier vector and the new outlier factor.

[0114] This step includes calculating the difference between the second outlier score corresponding to the outlier vector and the new outlier factor to obtain the second influence value of the target label value. Alternatively, it may include calculating the difference between the second outlier score and the new outlier factor to obtain the score difference, and then multiplying this score difference by a preset value to obtain the second influence value of the target label value.

[0115] The second influence value of each target label value can be obtained by repeating the above steps.

[0116] As can be seen from the description of the above embodiments, the embodiments of this disclosure obtain the second anomaly score corresponding to each vector by using the local outlier factor algorithm, and recalculate the second anomaly score of the vector by changing the value of the target label value. Based on the degree of change of the second anomaly score calculated twice, the second influence value corresponding to the target label value is determined, thereby facilitating the subsequent determination of the user's differentiated characteristics.

[0117] Figure 4 This is a schematic diagram of the structure of the customer profile differentiation feature detection device provided in the embodiments of this application. Figure 4As shown, the customer profile differential feature detection device 400 includes: a tag acquisition module 401, a matrix determination module 402, a first determination module 403, a second determination module 404, and a feature determination module 405.

[0118] The tag acquisition module 401 is used to acquire at least one tag to be processed corresponding to the target customer identifier;

[0119] The matrix determination module 402 is used to determine the input matrix based on the labels to be processed corresponding to each customer identifier, wherein the input matrix is ​​composed of vectors corresponding to the customer identifiers, and the vectors are composed of label values;

[0120] The first determination module 403 is used to determine the first anomaly score and the first influence value corresponding to the label value of the vector in the input matrix by using the isolated forest algorithm.

[0121] The second determination module 404 is used to determine the second anomaly score and the second influence value corresponding to the label value of the vector in the input matrix by using the local outlier factor algorithm.

[0122] The feature determination module 405 is used to determine the differentiated features of the customer profile corresponding to the target customer identifier based on the first abnormal score corresponding to the target vector, the second abnormal score corresponding to the target vector, the first influence value corresponding to the target label value under the target vector, and the second influence value corresponding to the target label value, wherein the target customer identifier corresponds to the target vector.

[0123] The apparatus provided in this embodiment can be used to execute the technical solutions of the above method embodiments. Its implementation principle and technical effects are similar, and will not be described again here.

[0124] In one possible implementation, the feature determination module 405 is used to: determine the target label value as an anomalous label value if the first influence value or the second influence value corresponding to the target label value is greater than the influence value threshold; determine the probability of occurrence of the anomalous label value or combination of anomalous label values ​​in the input matrix; determine the sparsity coefficient corresponding to the anomalous label value or combination of anomalous label values ​​based on the probability of occurrence; use the first anomalous score of the target vector as the weight of the first influence value corresponding to the anomalous label value, and the second anomalous score of the target vector as the weight of the second influence value corresponding to the anomalous label value, and sum the first influence value and the second influence value by weight to obtain the comprehensive influence value of the anomalous label value; multiply the comprehensive influence value of the anomalous label value by the sparsity coefficient of the anomalous label value to obtain the differential score of the anomalous label value; and determine the differential features of the customer profile corresponding to the target customer identifier based on the differential score of the anomalous label value.

[0125] In one possible implementation, the feature determination module 405 is used to determine the label to be processed corresponding to the abnormal label value as a differentiated feature of the customer profile corresponding to the target customer identifier if the differentiated score corresponding to the abnormal label value of the target customer identifier is greater than or equal to a preset differentiated score threshold; or, determine the number of abnormal label values ​​under the target customer identifier whose differentiated score is greater than or equal to the differentiated score threshold; if the number of label values ​​is greater than or equal to a preset number threshold, determine the label to be processed corresponding to the M abnormal label values ​​with the largest differentiated scores as a differentiated feature of the customer profile corresponding to the target customer identifier, where M is a positive integer; or, if the differentiated scores corresponding to all abnormal label values ​​under the target customer identifier are less than the differentiated score threshold, determine the label to be processed corresponding to the N abnormal label values ​​with the largest differentiated scores corresponding to the target customer identifier as a differentiated feature of the customer profile corresponding to the target customer identifier, where N is a positive integer.

[0126] In one possible implementation, the feature determination module 405 is used to subtract the occurrence probability from a preset value to obtain the sparsity coefficient corresponding to the abnormal label value or the combination of abnormal label values; or, based on the occurrence probability, to find the correspondence between a preset occurrence probability interval and the sparsity coefficient to obtain the sparsity coefficient corresponding to the abnormal label value or the combination of abnormal label values.

[0127] In one possible implementation, the first determining module 403 is specifically used to input the matrix to be input into the isolated forest algorithm to obtain multiple binary trees divided according to the label values, wherein the binary trees are composed of vectors as nodes; if the target label type corresponding to the target label value is adopted, the target vector is divided into leaf nodes of the binary trees, and the number of divisions corresponding to the target label type under the target vector is increased by K, where K is a positive integer; the number of divisions of the target label type under the target vector in each binary tree is added together to obtain the total number of divisions of the target label type under the target vector; the first influence value corresponding to the target label type is determined according to the total number of divisions of the target label type; the average number of edges from the target vector to the root node of the binary tree in each binary tree is determined as the target edge number; and the first anomaly score is determined according to the target edge number.

[0128] In one possible implementation, the second determining module 404 is specifically used to input the input matrix to the local outlier factor algorithm to obtain the second outlier score corresponding to each vector in the input matrix output by the local outlier factor algorithm; determine the vectors whose second outlier score is greater than the outlier score threshold as outlier vectors; replace the target label value in the outlier vector with a preset label value, and execute the step of inputting the input matrix to the local outlier factor algorithm again to obtain the new outlier factor corresponding to the outlier vector, wherein the preset label value corresponds to the label type of the target label value; and determine the second influence value of the target label value based on the second outlier score corresponding to the outlier vector and the new outlier factor.

[0129] The apparatus provided in this embodiment can be used to execute the technical solutions of the above method embodiments. Its implementation principle and technical effects are similar, and will not be described again here.

[0130] To implement the above embodiments, this application also provides an electronic device.

[0131] refer to Figure 5 The diagram illustrates a structural schematic of an electronic device 500 suitable for implementing embodiments of this application. The electronic device 500 can be a terminal device or a server. The terminal device can include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, personal digital assistants (PDAs), portable Android devices (PADs), portable media players (PMPs), and in-vehicle terminals (e.g., in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. Figure 5 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0132] like Figure 5 As shown, the electronic device 500 may include a processor (e.g., a central processing unit, a graphics processing unit, etc.) 501 and a memory 502 communicatively connected to the processor. The processor can perform various appropriate actions and processes based on programs stored in the memory 502, computer-executed instructions, or programs loaded from storage device 508 into random access memory (RAM) 503, thereby implementing the customer profile differentiation feature detection method in any of the above embodiments. The memory may be a read-only memory (ROM). The RAM 503 also stores various programs and data required for the operation of the electronic device 500. The processing device 501, the memory 502, and the RAM 503 are interconnected via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0133] Typically, the following devices can be connected to I / O interface 505: input devices 506 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 507 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 508 including, for example, magnetic tapes, hard disks, etc.; and communication devices 509. Communication device 509 allows electronic device 500 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 5 An electronic device 500 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively.

[0134] Specifically, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable storage medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 509, or installed from a storage device 508, or installed from a memory 502. When the computer program is executed by the processing device 501, it performs the functions defined in the methods of the embodiments of this application.

[0135] It should be noted that the computer-readable storage medium described above in this application can be a computer-readable signal medium, a computer storage medium, or any combination of the two. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable storage medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable storage medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0136] The aforementioned computer-readable storage medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0137] The aforementioned computer-readable storage medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform the method shown in the above embodiments.

[0138] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a Local Area Network (LAN) or a Wide Area Network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0139] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0140] The modules described in the embodiments of this application can be implemented in software or in hardware. The names of the units do not necessarily limit the module itself; for example, the first determining module can also be described as a "first influence value determining module".

[0141] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), and so on.

[0142] This application also provides a computer-readable storage medium storing computer-executable instructions. When a processor executes the computer-executable instructions, it implements the technical solution of the customer profile differential feature detection method in any of the above embodiments. Its implementation principle and beneficial effects are similar to those of the customer profile differential feature detection method. Please refer to the implementation principle and beneficial effects of the customer profile differential feature detection method, which will not be repeated here.

[0143] In the context of this application, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0144] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the technical solution of the customer profile differential feature detection method in any of the above embodiments. Its implementation principle and beneficial effects are similar to those of the customer profile differential feature detection method, and can be found in the implementation principle and beneficial effects of the customer profile differential feature detection method, which will not be repeated here.

[0145] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this application.

[0146] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0147] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A method for detecting differentiated features in customer profiles, characterized in that, include: Obtain at least one tag to be processed corresponding to the target customer identifier; Based on the unprocessed tags corresponding to each customer identifier, a matrix to be input is determined, wherein the matrix to be input consists of vectors corresponding to customer identifiers, and the vectors consist of tag values. The isolated forest algorithm is used to determine the first anomaly score corresponding to the vector in the input matrix and the first influence value corresponding to the label value. The local outlier factor algorithm is used to determine the second anomaly score corresponding to the vector in the input matrix and the second influence value corresponding to the label value. Based on the first anomaly score corresponding to the target vector, the second anomaly score corresponding to the target vector, the first influence value corresponding to the target label value under the target vector, and the second influence value corresponding to the target label value, the differentiated features of the customer profile corresponding to the target customer identifier are determined, wherein the target customer identifier corresponds to the target vector.

2. The method according to claim 1, characterized in that, The step of determining the differentiated features of the customer profile corresponding to the target customer identifier based on the first anomaly score corresponding to the target vector, the second anomaly score corresponding to the target vector, the first influence value corresponding to the target label value under the target vector, and the second influence value corresponding to the target label value includes: If the first influence value or the second influence value corresponding to the target label value is greater than the influence value threshold, then the target label value is determined to be an abnormal label value; Determine the probability of the occurrence of the abnormal label value or a combination of the abnormal label values ​​in the input matrix; Based on the occurrence probability, determine the sparsity coefficient corresponding to the abnormal label value or the combination of abnormal label values; The first anomaly score of the target vector is used as the weight of the first influence value corresponding to the anomaly label value, and the second anomaly score of the target vector is used as the weight of the second influence value corresponding to the anomaly label value. The first influence value and the second influence value are weighted and summed to obtain the comprehensive influence value of the anomaly label value. The differential score of the abnormal label value is obtained by multiplying the comprehensive impact value of the abnormal label value by the sparsity coefficient of the abnormal label value. Based on the differential score of the abnormal label value, the differential features of the customer profile corresponding to the target customer identifier are determined.

3. The method according to claim 2, characterized in that, The step of determining the differentiated features of the customer profile corresponding to the target customer identifier based on the differentiated score of the abnormal label value includes: If the differential score corresponding to the abnormal tag value of the target customer identifier is greater than or equal to a preset differential score threshold, then the tag to be processed corresponding to the abnormal tag value is determined to be a differential feature of the customer profile corresponding to the target customer identifier; or, Determine the number of abnormal tag values ​​under the target customer identifier whose differential score is greater than or equal to the differential score threshold; if the number of tag values ​​is greater than or equal to a preset number threshold, then determine the tags to be processed corresponding to the M abnormal tag values ​​with the largest differential scores as the differential features of the customer profile corresponding to the target customer identifier, where M is a positive integer; or, If the differential scores corresponding to all abnormal tag values ​​under the target customer identifier are less than the differential score threshold, then the tags to be processed corresponding to the N abnormal tag values ​​with the largest differential scores corresponding to the target customer identifier are determined as the differential features of the customer profile corresponding to the target customer identifier, where N is a positive integer.

4. The method according to claim 2, characterized in that, The step of determining the sparsity coefficient corresponding to the anomaly label value or a combination of anomaly label values ​​based on the occurrence probability includes: The sparsity coefficient corresponding to the anomaly label value or a combination of anomaly label values ​​is obtained by subtracting the occurrence probability from a preset value; or, Based on the occurrence probability, the correspondence between the preset occurrence probability interval and the sparsity coefficient is found to obtain the sparsity coefficient corresponding to the abnormal label value or the combination of the abnormal label values.

5. The method according to any one of claims 1 to 4, characterized in that, The method employing the isolated forest algorithm to determine the first anomaly score corresponding to the vector in the input matrix and the first influence value corresponding to the label value includes: The input matrix is ​​input into the isolated forest algorithm to obtain multiple binary trees divided according to the label values, wherein the binary trees are composed of the vectors as nodes; If the target label type corresponding to the target label value is used, and the target vector is divided into the leaf nodes of the binary tree, then the number of divisions corresponding to the target label type under the target vector is increased by K, where K is a positive integer; The total number of partitions of the target label type under the target vector in each binary tree is obtained by summing the number of partitions of the target label type under the target vector. Based on the total number of divisions for the target tag type, determine the first influence value corresponding to the target tag type; The average number of edges from the target vector to the root node of each binary tree is taken as the target number of edges. The first anomaly score is determined based on the target number of edges.

6. The method according to any one of claims 1 to 4, characterized in that, The step of employing the local outlier factor algorithm to determine the second anomaly score corresponding to the vector in the input matrix and the second influence value corresponding to the label value includes: The input matrix is ​​input into the local outlier factor algorithm to obtain the second outlier score corresponding to each vector in the input matrix output by the local outlier factor algorithm. Vectors whose second anomaly score is greater than the anomaly score threshold are identified as outlier vectors; The target label value in the outlier vector is replaced with a preset label value, and the step of inputting the matrix to be input into the local outlier factor algorithm is executed again to obtain a new outlier factor corresponding to the outlier vector, wherein the preset label value corresponds to the label type of the target label value; The second influence value of the target label value is determined based on the second anomaly score corresponding to the outlier vector and the new anomaly factor.

7. A device for detecting differentiated features in customer profiles, characterized in that, include: The tag acquisition module is used to acquire at least one tag to be processed corresponding to the target customer identifier; The matrix determination module is used to determine the input matrix based on the labels to be processed corresponding to each customer identifier, wherein the input matrix is ​​composed of vectors corresponding to the customer identifiers, and the vectors are composed of label values; The first determination module is used to determine the first anomaly score corresponding to the vector in the input matrix and the first influence value corresponding to the label value using the isolated forest algorithm. The second determination module is used to determine the second anomaly score corresponding to the vector in the input matrix and the second influence value corresponding to the label value by using the local outlier factor algorithm. The feature determination module is used to determine the differentiated features of the customer profile corresponding to the target customer identifier based on the first abnormal score corresponding to the target vector, the second abnormal score corresponding to the target vector, the first influence value corresponding to the target label value under the target vector, and the second influence value corresponding to the target label value, wherein the target customer identifier corresponds to the target vector.

8. An electronic device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method described in any one of claims 1 to 6.

Images