Image generation method, device, computer device, and storage medium for target object

Abstract

The present disclosure relates to a portrait generation method and device of a target object, a computer device and a storage medium. The method comprises: obtaining sample data of at least one target object, determining first feature index data in the sample data; performing clustering calculation at least by using the first feature index data, determining a center feature value of a clustering center; calculating a deviation value between the center feature value and a model feature value of a preset target model, determining label data of the target object according to the deviation value, the first feature index data and the model feature value, the target model being determined according to third feature index data of a target position, and the model feature value being determined according to the third feature index data; and generating a portrait of the target object according to the label data of the target object. The present method can make full use of various data of the target object and accurately generate a portrait.

Description

Technical Field

[0001] This disclosure relates to the field of data processing technology, and in particular to a method, apparatus, computer device, and storage medium for generating a portrait of a target object. Background Technology

[0002] With the development of information technology, many companies now store various employee information, such as work experience, performance, job position, and salary standards, in their business systems. As companies grow, employee competency metrics have become extremely important, especially in personnel management and recruitment.

[0003] Currently, the common way to quantify the competency indicators of each employee is to use talent profiling to match employees to their positions or to conduct targeted management of individual employees.

[0004] However, most current talent profiling methods rely on the experience of human resource managers and are compiled based on employee information obtained according to their requirements. This often involves a certain degree of subjectivity, and because talent profiles are only based on employee information obtained according to their requirements and do not make full use of various employee data, the resulting talent profiles are inaccurate. Summary of the Invention

[0005] Therefore, it is necessary to provide a method, apparatus, computer equipment, and storage medium for generating portraits of target objects that can make full use of various employee data to accurately generate portraits, addressing the aforementioned technical problems.

[0006] Firstly, this disclosure provides a method for generating a profile of a target object, the method comprising:

[0007] Obtain sample data of at least one target object, and determine the first feature index data in the sample data;

[0008] At least the first feature index data is used to perform clustering calculations to determine the central feature values ​​of the cluster centers;

[0009] Calculate the deviation between the central feature value and the model feature value of the preset target model. Based on the deviation value, the first feature index data, and the model feature value, determine the label data of the target object. The target model is determined based on the third feature index data of the target location, and the model feature value is determined based on the third feature index data.

[0010] A profile of the target object is generated based on the target object's tag data.

[0011] In one embodiment, the step of performing clustering calculations using at least the first feature index data to determine the central feature values ​​of the cluster centers includes:

[0012] Clustering calculations are performed on the first feature index data to obtain the feature values ​​of the cluster centers corresponding to the first feature index data.

[0013] Based on the feature values ​​of the cluster centers corresponding to the first feature index data, the central feature values ​​of the cluster centers are determined.

[0014] In one embodiment, the step of performing clustering calculations using at least the first feature index data to determine the central feature values ​​of the cluster centers includes:

[0015] Determine second feature index data from the first feature index data, and determine the remaining feature index data in the first feature index data excluding the second feature index data, wherein the number of second feature index data is less than the number of first feature index data.

[0016] Clustering calculations are performed on the second feature index data to determine the initial feature values ​​of the initial cluster centers;

[0017] Based on the initial feature values ​​and the remaining feature index data, the center feature values ​​of the cluster centers are determined.

[0018] In one embodiment, determining the center feature value of the cluster center based on the initial feature value and the remaining feature index data includes:

[0019] Calculate the similarity between the initial feature values ​​and the remaining feature index data;

[0020] In response to the similarity being less than a preset similarity threshold, the remaining feature index data is allocated to the data corresponding to the initial cluster center to obtain the first cluster;

[0021] Calculate the first feature value of the first cluster center of the first cluster; in response to the fact that the similarity between the first feature value and the remaining feature index data is less than a preset similarity threshold, determine the first feature value as the center feature value of the cluster center.

[0022] In response to the fact that the similarity between the first feature value and the remaining feature index data is greater than or equal to a preset similarity threshold, the remaining feature index data is reassigned to the data corresponding to the initial cluster center until the similarity between the first feature value and the remaining feature index data is less than the preset similarity threshold.

[0023] In one embodiment, before calculating the deviation between the central feature value and the model feature value of the preset target model, the method further includes:

[0024] The first feature index data and the model feature values ​​are normalized using the logarithmic transformation method.

[0025] In one embodiment, determining the label data of the target object based on the deviation value, the first feature index data, and the model feature value includes:

[0026] In response to the deviation value being less than a preset deviation threshold, a second deviation value is calculated between the first feature index data and the central feature value, and the data score of the first feature index data is determined based on the second deviation value.

[0027] The label data of the target object is determined based on the data score of the first feature index data.

[0028] In one embodiment, the method further includes: adjusting the target model and / or reacquiring the sample data in response to the deviation value being greater than or equal to a preset deviation threshold.

[0029] In one embodiment, before generating the portrait of the target object based on the tag data of the target object, the method further includes:

[0030] Based on the target type corresponding to the label data of the target object and the data score of the first feature index data, the type score corresponding to the target type is determined;

[0031] Accordingly, generating a profile of the target object based on the target object's tag data includes:

[0032] A profile of the target object is generated based on the tag data of the target object and the type score corresponding to the target type.

[0033] In one embodiment, after generating the portrait of the target object based on the target object's tag data, the method further includes:

[0034] In response to the need to search for the profile of the target object, a distributed search and analysis engine is used to search for the profile of the target object.

[0035] Secondly, this disclosure also provides a portrait generation apparatus for a target object, the apparatus comprising:

[0036] The data acquisition module is used to acquire sample data of at least one target object and determine the first feature index data in the sample data;

[0037] The clustering calculation module is used to perform clustering calculations using at least the first feature index data to determine the central feature value of the cluster center;

[0038] The label data determination module is used to calculate the deviation value between the central feature value and the model feature value of the preset target model, and determine the label data of the target object based on the deviation value, the first feature index data and the model feature value. The target model is determined based on the third feature index data of the target location, and the model feature value is determined based on the third feature index data.

[0039] The image generation module is used to generate an image of the target object based on the tag data of the target object.

[0040] Thirdly, this disclosure also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of any of the above-described method embodiments.

[0041] Fourthly, this disclosure also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps of any of the above-described method embodiments.

[0042] Fifthly, this disclosure also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the steps of any of the above-described method embodiments.

[0043] In the above embodiments, by acquiring sample data of at least one target object, first feature index data is determined from the sample data. The sample data typically includes various data related to the target object, thus enabling the determination of multi-faceted first feature index data. Furthermore, clustering calculations can be performed on the multi-faceted first feature index data to determine the central feature values ​​of the cluster centers, identifying the most accurate or standard central feature value for each type. The deviation between the central feature values ​​and the model feature values ​​of a preset target model is calculated. Based on the deviation value, the first feature index data, and the model feature values, the label data of the target object is determined. Multiple types of first feature index data can be used to determine the label data of the target object. Because the label data is obtained using the various data and deviation values ​​of the target object, it accurately reflects various aspects of the target object. Therefore, a portrait of the target object can be accurately generated based on the label data. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in the specific embodiments of this disclosure or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0045] Figure 1 This is a schematic diagram illustrating the application environment of a target object image generation method in one embodiment.

[0046] Figure 2 This is a flowchart illustrating a method for generating a portrait of a target object in one embodiment.

[0047] Figure 3 This is a flowchart illustrating step S204 in one embodiment;

[0048] Figure 4 This is a flowchart illustrating step S204 in one embodiment;

[0049] Figure 5 This is a flowchart illustrating step S406 in one embodiment;

[0050] Figure 6 This is a flowchart illustrating step S206 in one embodiment;

[0051] Figure 7 This is a schematic diagram of the image of the target object in one embodiment;

[0052] Figure 8 This is a schematic block diagram of a target object image generation device in one embodiment;

[0053] Figure 9 This is a schematic diagram of the internal structure of a computer device in one embodiment. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of this disclosure clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this disclosure.

[0055] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings herein are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, apparatus, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0056] In this article, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0057] This disclosure provides a method for generating a portrait of a target object, which can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated on server 104 or placed on a cloud or other network server. Terminal 102 can obtain sample data of at least one target object from server 104. Terminal 102 can determine the first feature index data in the sample data. Terminal 102 uses at least the first feature index data to perform clustering calculations to determine the central feature value of the cluster center. Terminal 102 can calculate the deviation value between the central feature value and the model feature value of a preset target model, and determine the label data of the target object based on the deviation value and the first feature index data. The target model can be determined by server 104 or terminal 102 based on the third feature index data of the target location. The model feature value is determined based on the third feature index data. Terminal 102 can generate a portrait of the target object based on the label data of the target object. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, etc. Server 104 can be implemented using a standalone server or a server cluster composed of multiple servers.

[0058] In one embodiment, such as Figure 2 As shown, a method for generating a portrait of a target object is provided, which can be applied to... Figure 1 Taking terminal 102 as an example, the explanation includes the following steps:

[0059] S202, acquire sample data of at least one target object, and determine the first feature index data in the sample data.

[0060] The target audience typically includes employees for whom profiles need to be generated. In some embodiments of this disclosure, the profile is often a job profile. Establishing clear talent standards is a crucial part of talent planning in the process of talent selection within an enterprise. Clear talent standards, known as job profiles, are precise descriptions of employees capable of high performance in key positions. These profiles include directly observable explicit characteristics (such as gender, age, knowledge, experience, etc.) and implicit characteristics (such as personality, learning ability, motivation, etc.). Sample data typically includes various information about the target audience, such as work performance, years of service, salary, daily working hours, and so on. The first characteristic indicator data typically includes characteristic indicators from the sample data, such as behavioral characteristic indicators, performance characteristic indicators, ability characteristic indicators, experience levels, and other characteristic indicators. For example, if the sample data is work performance, then the first characteristic indicator data could be performance indicators. The first characteristic indicator data varies depending on the sample data. Furthermore, the first characteristic indicator data can include multiple different characteristic indicators. In some embodiments of this disclosure, the number and type of the first characteristic indicators are not absolutely limited.

[0061] Specifically, the terminal can determine the data source and data interface based on different indicator extraction requirements, and then obtain sample data of the corresponding target object based on the data source and data interface. This allows it to determine the first feature indicator data within the sample data.

[0062] In some exemplary embodiments, after obtaining the sample data, the sample data can be cleaned to determine the first feature index data in the cleaned sample data, thereby removing the influence of invalid data.

[0063] S204, at least the first feature index data is used to perform clustering calculations to determine the central feature value of the cluster center.

[0064] Clustering calculations typically involve cluster analysis. Cluster analysis is based on similarity; patterns within a cluster are more similar than patterns in different clusters. Cluster centers are the centers of the clusters obtained after clustering calculations. The eigenvalues ​​of these centers often reflect the characteristics of the target object. For example, center eigenvalues ​​might indicate a tendency towards performance or personnel management.

[0065] Specifically, clustering algorithms can be used to cluster the first feature index data to obtain cluster centers and their central feature values. Clustering algorithms may include K-MEANS, K-MEDOIDS, Clara, and Clarans. In some embodiments of this disclosure, no absolute limitation is placed on the clustering algorithm used. If the first feature index data consists of multiple different types, then typically in step S206, the deviation between the central feature value and the model feature value of a preset target model is calculated. Based on the deviation value, the first feature index data, and the model feature value, the label data of the target object is determined.

[0066] The target model is determined based on the third characteristic indicator data of the target location, and the model feature values ​​are determined based on the third characteristic indicator data. The target location is typically a job position to be matched, such as a sales position. The corresponding third characteristic indicator data can typically be performance data, client data, etc. Each job position to be matched has a corresponding target model. The target model can usually include multiple model feature values. A model feature value can typically be composed of one or more third characteristic indicator data. For example, a combination of multiple third characteristic indicator data can yield a single model feature value. A single third characteristic indicator data can also yield a single model feature value. The model feature values ​​of the target model can be used to characterize whether a target object is competent for the position, ensuring that the personnel can successfully complete the job requirements. For example, if the position (target location) is a sales position, its corresponding model feature indicator could be a monthly sales performance of 100,000. Generally, only by achieving a monthly sales performance of 100,000 can one be competent for the sales position. In addition, the type of the model feature value usually needs to be the same as the type of the central feature value. For example, if the model feature value is of the performance type, then the type of the central feature value usually also needs to be of the performance type.

[0067] Specifically, the deviation between the central feature value and each model feature value of the target model can be calculated. When the deviation is relatively small, the label data can be determined based on the first feature index data and the model feature values. Each first feature index data usually corresponds to a different label. When the deviation between the first feature index data and the model feature values ​​is small, the label data of the target object can be determined based on the index corresponding to the first feature index data.

[0068] For example, each primary characteristic indicator typically belongs to a larger indicator type. For instance, primary characteristic indicators under performance indicators could include performance data, customer data, etc. If the deviation between the central characteristic value of the performance data and the model characteristic value is small, and the deviation between the central characteristic value of the customer data and another model characteristic value is small, then outstanding performance capabilities can be identified as label data for the target object.

[0069] It is understood that the method of calculating the deviation value is not absolutely limited in some embodiments of this disclosure. For example, the deviation value can be obtained by subtracting the model feature value from the central feature value. Other methods can also be used to calculate the deviation value between the central feature value and the model feature value. For example, the deviation value can be obtained by subtracting the average value of the central feature value from the central feature value and dividing it by the average value of the model feature value.

[0070] S208, Generate a portrait of the target object based on the tag data of the target object.

[0071] Specifically, the label data of the target object identified above can be used to generate a profile of the target object.

[0072] In the aforementioned method for generating a profile of a target object, sample data of at least one target object is acquired, and first feature index data is determined from the sample data. The sample data typically includes various data related to the target object, thus enabling the determination of multi-faceted first feature index data. Furthermore, clustering calculations can be performed on the multi-faceted first feature index data to determine the central feature values ​​of the cluster centers, identifying the most accurate or standard central feature value for each type. The deviation between the central feature values ​​and the model feature values ​​of a preset target model is calculated. Based on the deviation value, the first feature index data, and the model feature values, the label data of the target object is determined. This method utilizes multiple types of first feature index data to determine the label data of the target object. Because the label data is obtained using the various data and deviation values ​​of the target object, it accurately reflects various aspects of the target object. Therefore, a profile of the target object can be accurately generated based on the target object's label data.

[0073] In one embodiment, such as Figure 3 As shown, the step of performing clustering calculations using at least the first feature index data to determine the central feature values ​​of the cluster centers includes:

[0074] S302, perform clustering calculation on the first feature index data to obtain the feature value of the cluster center corresponding to the first feature index data;

[0075] S304. Determine the center feature value of the cluster center based on the feature value of the cluster center corresponding to the first feature index data.

[0076] Specifically, clustering calculations can be performed on the first characteristic index data of each type to obtain the feature values ​​of the cluster centers for each type of first characteristic index data. Then, based on the feature values ​​of each type obtained above, the central feature values ​​of the cluster centers for each type can be determined.

[0077] In another case, such as Figure 4 As shown, the step of performing clustering calculations using at least the first feature index data to determine the central feature values ​​of the cluster centers includes:

[0078] S402, determine the second feature index data in the first feature index data, and determine the remaining feature index data in the first feature index data excluding the second feature index data, wherein the number of the second feature index data is less than the number of the first feature index data.

[0079] S404, perform clustering calculations on the second feature index data to determine the initial feature values ​​of the initial cluster centers.

[0080] S406, Determine the center feature value of the cluster center based on the initial feature value and the remaining feature index data.

[0081] Specifically, a certain number of first characteristic indicator data can be selected from the first characteristic indicator data, and this certain number of first characteristic indicator data can be second characteristic indicator data. In some embodiments of this disclosure, the certain number of data is not limited, and the certain number can be determined based on the quantity of first characteristic indicator data. Then, the remaining characteristic indicator data other than the second characteristic indicator data in the first characteristic indicator data is determined. Then, clustering calculation is performed on the second characteristic indicator data to determine the central characteristic value of the initial cluster centers obtained after the clustering calculation. Then, the central characteristic value of the cluster centers can be determined based on the initial characteristic value and the remaining characteristic indicator data obtained above. For example, the characteristic value of the cluster centers of the remaining characteristic indicator data can be calculated, and the central characteristic value of the cluster centers can be determined based on the relationship between the initial characteristic value and the characteristic value of the cluster centers of the remaining characteristic indicator data. For example, if the difference between the initial characteristic value and the characteristic value of the cluster centers of the remaining characteristic indicator data is greater than a preset difference threshold, then it can usually be proven that the initial characteristic value and / or the remaining characteristic indicator data are unreasonable, and it is necessary to reselect the second characteristic indicator data and then recalculate the initial characteristic value and / or the remaining characteristic indicator data. If the difference is less than a preset difference threshold, then the cluster center of the initial feature value and / or the remaining feature index data can be determined as the center feature value. The initial feature value typically corresponds to the type of the second feature index data. For example, if the second feature index data consists of two types of data, then each type of second feature index data can typically obtain a corresponding initial feature value.

[0082] In some exemplary embodiments, for example, if the first feature index data is A, B, C and D, C and D can be selected as the second feature index data, and the remaining feature index data can be A and B. The initial feature values ​​of C and D can be calculated respectively, and then the center feature value of the cluster center can be determined based on the initial feature value and the remaining feature index data.

[0083] In this embodiment, by using two different methods to determine the central feature value of the cluster center, different methods can be selected under different circumstances, thereby improving the computational efficiency or the accuracy of the central feature value, making the subsequently calculated label data more accurate.

[0084] In one embodiment, such as Figure 5 As shown, determining the central feature value of the cluster center based on the initial feature value and the remaining feature index data includes:

[0085] S502, calculate the similarity between the initial feature values ​​and the remaining feature index data;

[0086] S504, determine whether the similarity is less than a preset similarity threshold.

[0087] S506, in response to the similarity being less than a preset similarity threshold, the remaining feature index data is allocated to the data corresponding to the initial cluster center to obtain the first cluster;

[0088] S508, calculate the first feature value of the first cluster center of the first cluster, and in response to the fact that the similarity between the first feature value and the remaining feature index data is less than a preset similarity threshold, determine the first feature value as the center feature value of the cluster center;

[0089] S510, in response to the fact that the similarity between the first feature value and the remaining feature index data is greater than or equal to a preset similarity threshold, the remaining feature index data is reassigned to the data corresponding to the initial cluster center until the similarity between the first feature value and the remaining feature index data is less than the preset similarity threshold.

[0090] The similarity can be calculated using methods such as Euclidean distance, cosine similarity, and Pearson correlation coefficient.

[0091] Specifically, the similarity between the initial feature value and each remaining feature index data can be calculated. Then, it is determined whether the calculated similarity is less than a preset similarity threshold. Those skilled in the art can set the similarity threshold according to actual circumstances; this disclosure does not impose any restrictions on the similarity threshold. When the calculated similarity is less than the preset similarity threshold, it can be determined that the initial feature value and the remaining feature index data are relatively correlated. Therefore, the remaining feature index data can be allocated to the second index data that obtained the initial feature value, resulting in a first cluster. Then, the first cluster center and the first feature value of the first cluster center can be calculated using a clustering algorithm. The similarity between the first feature value and the aforementioned remaining feature index is calculated again. When this similarity is less than the preset similarity threshold, it can be determined that the first cluster obtained after allocating the remaining feature index data is relatively reasonable. At this point, it can be determined that the first feature value of the first cluster center can be the central feature value of the cluster center. When the similarity between the first feature value and the remaining feature indicators is greater than or equal to the preset similarity threshold, it can be determined that the first cluster obtained after allocating the remaining feature indicator data is unreasonable. At this time, the data deviation in the first cluster may be relatively large. Therefore, it is necessary to reallocate the remaining feature indicator data, obtain the first cluster again after reallocation, and recalculate the first feature value of the first cluster until the similarity between the first feature value and the remaining feature indicator data is less than the preset similarity threshold, so as to obtain the center feature value.

[0092] In some exemplary embodiments, for example, the initial feature value of the initial cluster center is S1, the remaining feature index data can be A and B, and the data corresponding to the initial cluster center (second feature index data) can be C and D. The similarity between S1 and A, and between S1 and B can be calculated respectively. If the obtained similarity scores are 5 and 6 respectively, and the similarity threshold is 7, then the similarity between S1 and A is 5, and the similarity between S1 and B is 6, both less than the similarity threshold. Therefore, the remaining feature index data A and B can be assigned to the second feature index data to obtain the first cluster. The first cluster can include A, B, C, and D. Then, the first feature value S2 of the first cluster center of the first cluster can be calculated. The similarity between S2 and A, and between S2 and B, can be calculated again. The obtained similarity scores are 3 and 4 respectively, both less than the similarity threshold. Therefore, the first feature value S2 can be determined as the center feature value of the cluster center.

[0093] If the initial feature values ​​of the initial cluster centers are S1 and S2, the second feature index data corresponding to the initial feature value S1 can be C, and the second feature index data corresponding to the initial feature value S2 can be D. The remaining feature index data can be A and B. The similarity between S1 and A, S1 and B, S2 and A, and S2 and B can be calculated respectively. If the similarity between S1 and A is 5, the similarity between S1 and B is 10, the similarity between S2 and A is 6, and the similarity between S2 and B is 6, when the similarity threshold is 7, the remaining index data A can be assigned to C, and the remaining index data B can be assigned to D. A and C can be in the first cluster, and B and D can also be in the first cluster. Then, the first feature value of the first cluster center of the first cluster of A and C can be calculated, and the similarity between the first feature value and the remaining index data A and B can be calculated. When the similarity between the first feature value and the remaining index data A is 7, the remaining index data A needs to be reassigned. It can be assigned to D. Then, A, B, and D can form the first cluster. Next, the first feature value of the cluster centers of the first cluster A, B, and D is calculated. Then, the similarity between this first feature value and the remaining index data A is calculated. When the similarity is 5, the first feature value can be determined as the central feature value of the cluster center.

[0094] In this embodiment, the remaining feature index data is assigned to the second feature index data using similarity. If the similarity threshold is not met, the remaining feature index data can be reallocated, thereby enabling accurate clustering of various feature index data and accurately obtaining the central feature value of the cluster center.

[0095] In one embodiment, before calculating the deviation between the central feature value and the model feature value of the preset target model, the method further includes:

[0096] The first feature index data and the model feature values ​​are normalized using the logarithmic transformation method.

[0097] Logarithmic transformation is a common method for data transformation, specifically a logarithmic transformation. This special transformation can convert a theoretically unsolved model problem into a solved one. The reason for taking the logarithm is that the logarithmic function is monotonically increasing within its domain. Taking the logarithm does not change the relative relationships of the data. Its main function is to help stabilize variance, keeping the distribution close to a normal distribution and making the data independent of the distribution's mean.

[0098] Specifically, in order to eliminate the long-tail effect that may be caused by extreme values ​​in the first feature index data and the model feature values, the logarithmic transformation method can be used to normalize the first feature index data and the model feature values.

[0099] In this embodiment, using the logarithmic transformation method can reduce the absolute value of the data, facilitating calculation. Furthermore, in some cases, the differences between different intervals across the entire value range of the data have varying impacts. Besides, taking the logarithm does not change the properties or correlations of the data.

[0100] In one embodiment, such as Figure 6 As shown, determining the label data of the target object based on the deviation value, the first feature index data, and the model feature value includes:

[0101] S602, determine whether the deviation value is greater than the preset deviation threshold.

[0102] S604, in response to the deviation value being less than a preset deviation threshold, calculate a second deviation value between the first feature index data and the central feature value, and determine the data score of the first feature index data based on the second deviation value.

[0103] S606, determine the tag data of the target object based on the data score of the first feature index data.

[0104] Specifically, the relationship between the deviation value and a pre-set deviation threshold can be determined. When the deviation value is less than the pre-set deviation threshold, it can be determined that there is a strong correlation between the first feature index data and the model feature value. That is, the first feature index data is more consistent with the target position. Therefore, a second deviation value can be calculated between the first feature index data and its corresponding central feature value. Then, the data score of the first feature index data is determined based on the second deviation value. For example, if the second deviation value is relatively high, the corresponding data score of the first feature index data is usually lower. If the second deviation value is relatively low, the corresponding data score of the first feature index data is usually higher.

[0105] Additionally, the target object's tag data can be obtained by summing the scores of all primary characteristic indicators for each type. Tag data could be, for example, "strong ability in xxx" or "rich experience in xxx". For instance, if the primary characteristic indicators are performance and customer metrics, and the scores for performance and customer metrics are both relatively high (80 and 85 respectively), the sum would be 165. This 165 score can be compared to a preset standard score threshold of 120. If it is greater than 120, the target object's tag data could be: "Excellent performance ability, high customer satisfaction". It should be understood that the above is only for illustrative purposes.

[0106] S608, in response to the deviation value being greater than or equal to a preset deviation threshold, adjust the target model and / or reacquire the sample data.

[0107] Specifically, when the deviation value is greater than or equal to a pre-set deviation threshold, it can be determined that the deviation value is large. There are two possibilities for a large deviation: one is a problem with the sampling data, meaning the sample data obtained for the target object is incorrect, in which case the sample data can be re-obtained. The other possibility is that the model feature values ​​of the target model no longer meet the current data requirements, in which case the appropriate model features should be adjusted based on the deviation. However, in most cases, companies are very clear about the model feature values ​​of the target model for their target location. Therefore, in most cases, it is necessary to re-obtain the sample data.

[0108] In this embodiment, when the deviation value is relatively small, it can be proven that the correlation between the first feature index data and the model feature value is relatively strong. Therefore, it can be determined that the first feature index data can be used to determine the label data. The deviation between the first feature index data and the central feature value can be calculated to accurately determine the deviation from the central feature value. Based on the second deviation value, the label data of the target object can be accurately determined.

[0109] In one embodiment, before generating the portrait of the target object based on the target object's tag data, the method further includes:

[0110] Based on the target type corresponding to the label data of the target object and the data score of the first feature index data, the type score corresponding to the target type is determined;

[0111] Accordingly, generating a profile of the target object based on the target object's tag data includes:

[0112] A profile of the target object is generated based on the tag data of the target object and the type score corresponding to the target type.

[0113] Among them, the target type can usually be various different types of abilities of the target object, such as experience, contribution, innovation, professional skills and so on.

[0114] Specifically, each tag data of the target object usually belongs to a target type (experience, expertise, innovation, and contribution). Each tag data has a corresponding score, which can be accumulated to obtain the type score of each target type. Alternatively, each tag data can be multiplied by its corresponding weight and added together to obtain the type score of each target type.

[0115] like Figure 7 As shown, a profile of the target object can be generated based on the tag data and type scores. During the generation of the target object profile, a scatter plot component can be introduced to visually present the tag data in a word cloud format, and random font colors can be used to enhance the display effect. Additionally, a multi-dimensional radar chart can be generated based on the type scores of the target type.

[0116] In this embodiment, by determining the type score corresponding to the target column of the target object through the target object's tag data, it is possible to more accurately determine the data of the target object in various aspects, and the generated target object profile can more accurately reflect the various indicator capabilities of the target object.

[0117] In one embodiment, after generating the portrait of the target object based on the target object's tag data, the method further includes:

[0118] In response to the need to search for the profile of the target object, a distributed search and analysis engine is used to search for the profile of the target object.

[0119] The distributed search and analysis engine is typically an Elasticsearch intelligent search engine. Elasticsearch is a search server based on Lucene that provides distributed full-text search capabilities.

[0120] Specifically, when there are many target objects and a large sample size, resulting in a large amount of tag data in the target object profiles, the continuous accumulation of tag data makes it impossible to meet the search requirements of large-scale dynamic combinations using only traditional relational databases. Therefore, the Elasticsearch distributed search and analysis engine is introduced. When it is necessary to search the profile of a target object, all tag data is indexed and made searchable. Then, by connecting to the database server through a RESTful web interface, the user's requirements for 100% customization and high-speed, stable search can be met.

[0121] In this embodiment, the Elasticsearch distributed search and analysis engine enables real-time searching and can stably, reliably, and quickly obtain the profile of the corresponding target object using tag data.

[0122] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0123] Based on the same inventive concept, this disclosure also provides a target object portrait generation apparatus for implementing the above-described target object portrait generation method. The solution provided by this apparatus is similar to the implementation described in the above-described method; therefore, the specific limitations in one or more target object portrait generation apparatus embodiments provided below can be found in the limitations of the target object portrait generation method described above, and will not be repeated here.

[0124] In one embodiment, such as Figure 8 As shown, a target object profile generation device 800 is provided, including: a data acquisition module 802, a clustering calculation module 804, a tag data determination module 806, and a profile generation module 808, wherein:

[0125] The data acquisition module 802 is used to acquire sample data of at least one target object and determine the first feature index data in the sample data.

[0126] The clustering calculation module 804 is used to perform clustering calculations using at least the first feature index data to determine the central feature value of the cluster center.

[0127] The label data determination module 806 is used to calculate the deviation value between the central feature value and the model feature value of the preset target model, and determine the label data of the target object based on the deviation value, the first feature index data and the model feature value. The target model is determined based on the third feature index data of the target location, and the model feature value is determined based on the third feature index data.

[0128] The portrait generation module 808 is used to generate a portrait of the target object based on the tag data of the target object.

[0129] In one embodiment of the device, the clustering calculation module 804 is further configured to perform clustering calculations on the first feature index data to obtain feature values ​​of the cluster centers corresponding to the first feature index data; and determine the center feature values ​​of the cluster centers based on the feature values ​​of the cluster centers corresponding to the first feature index data.

[0130] In one embodiment of the device, the clustering calculation module 804 includes:

[0131] The indicator data determination module is used to determine the second characteristic indicator data in the first characteristic indicator data, and to determine the remaining characteristic indicator data in the first characteristic indicator data excluding the second characteristic indicator data, wherein the number of the second characteristic indicator data is less than the number of the first characteristic indicator data.

[0132] The clustering calculation submodule is used to perform clustering calculations on the second feature index data to determine the initial feature values ​​of the initial cluster centers.

[0133] The central feature value determination module is used to determine the central feature value of the cluster center based on the initial feature value and the remaining feature index data.

[0134] In one embodiment of the device, the center feature value determination module includes:

[0135] The similarity calculation module is used to calculate the similarity between the initial feature value and the remaining feature index data.

[0136] The first clustering module is used to allocate the remaining feature index data to the data corresponding to the initial cluster center in response to the similarity being less than a preset similarity threshold, thereby obtaining the first cluster.

[0137] The feature value determination module is used to calculate the first feature value of the first cluster center of the first cluster, and in response to the fact that the similarity between the first feature value and the remaining feature index data is less than a preset similarity threshold, the first feature value is determined to be the center feature value of the cluster center.

[0138] The secondary allocation module is used to reassign the remaining feature index data to the data corresponding to the initial cluster center in response to the similarity between the first feature value and the remaining feature index data being greater than or equal to a preset similarity threshold, until the similarity between the first feature value and the remaining feature index data is less than the preset similarity threshold.

[0139] In one embodiment of the device, the device further includes a normalization processing module, used to normalize the first feature index data and the model feature values ​​using a logarithmic transformation method.

[0140] In one embodiment of the device, the tag data determination module 806 is further configured to, in response to the deviation value being less than a preset deviation threshold, calculate a second deviation value between the first feature index data and the central feature value, determine a data score of the first feature index data based on the second deviation value, and determine the tag data of the target object based on the data score of the first feature index data.

[0141] In one embodiment of the device, the device further includes: a type score determination module, configured to determine a type score corresponding to the target type based on the target type corresponding to the tag data of the target object and the data score of the first feature index data.

[0142] The portrait generation module 808 is further configured to generate a portrait of the target object based on the tag data of the target object and the type score corresponding to the target type.

[0143] In one embodiment of the device, the device further includes: a search module, configured to search for the image of the target object using a distributed search and analysis engine in response to the need to search for the image of the target object.

[0144] Each module in the aforementioned target object image generation device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0145] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 9As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a method for generating a portrait of a target object. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0146] Those skilled in the art will understand that Figure 9 The structure shown is merely a block diagram of a portion of the structure related to the present disclosure and does not constitute a limitation on the computer device to which the present disclosure is applied. A specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0147] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in any of the above method embodiments.

[0148] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in any of the above method embodiments.

[0149] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in any of the above method embodiments.

[0150] It should be noted that the sample data of the target objects involved in this disclosure are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data shall comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0151] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this disclosure can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this disclosure may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this disclosure may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0152] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0153] The embodiments described above are merely illustrative of several implementations of this disclosure, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent disclosure. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this disclosure, and these all fall within the protection scope of this disclosure. Therefore, the protection scope of this disclosure should be determined by the appended claims.

Claims

1. A method for generating a portrait of a target object, characterized in that, The method includes: Obtain sample data of at least one target object, and determine the first feature index data in the sample data; At least the first feature index data is used to perform clustering calculations to determine the central feature values ​​of the cluster centers; Calculate the deviation between the central feature value and the model feature value of the preset target model. Based on the deviation value, the first feature index data, and the model feature value, determine the label data of the target object. The target model is determined based on the third feature index data of the target location, and the model feature value is determined based on the third feature index data. Generate a profile of the target object based on the target object's tag data; Wherein, the step of performing clustering calculations using at least the first feature index data to determine the central feature values ​​of the cluster centers includes: Determine second feature index data from the first feature index data, and determine the remaining feature index data in the first feature index data excluding the second feature index data, wherein the number of second feature index data is less than the number of first feature index data. Clustering calculations are performed on the second feature index data to determine the initial feature values ​​of the initial cluster centers; Determining the center feature value of a cluster center based on the initial feature value and the remaining feature index data includes: calculating the similarity between the initial feature value and the remaining feature index data; in response to the similarity being less than a preset similarity threshold, assigning the remaining feature index data to the data corresponding to the initial cluster center to obtain a first cluster; calculating the first feature value of the first cluster center of the first cluster; in response to the similarity between the first feature value and the remaining feature index data being less than a preset similarity threshold, determining the first feature value as the center feature value of the cluster center; in response to the similarity between the first feature value and the remaining feature index data being greater than or equal to a preset similarity threshold, reassigning the remaining feature index data to the data corresponding to the initial cluster center until the similarity between the first feature value and the remaining feature index data is less than a preset similarity threshold; The step of determining the label data of the target object based on the deviation value, the first feature index data, and the model feature value includes: In response to the deviation value being less than a preset deviation threshold, a second deviation value is calculated between the first feature index data and the central feature value, and the data score of the first feature index data is determined based on the second deviation value. The label data of the target object is determined based on the data score of the first feature index data; Before generating the profile of the target object based on the tag data of the target object, the method further includes: Based on the target type corresponding to the label data of the target object and the data score of the first feature index data, the type score corresponding to the target type is determined; Accordingly, generating a profile of the target object based on the target object's tag data includes: A profile of the target object is generated based on the tag data of the target object and the type score corresponding to the target type.

2. The method according to claim 1, characterized in that, The step of performing clustering calculations using at least the first feature index data to determine the central feature values ​​of the cluster centers includes: Clustering calculations are performed on the first feature index data to obtain the feature values ​​of the cluster centers corresponding to the first feature index data. Based on the feature values ​​of the cluster centers corresponding to the first feature index data, the central feature values ​​of the cluster centers are determined.

3. The method according to claim 1, characterized in that, Before calculating the deviation between the central feature value and the model feature value of the preset target model, the method further includes: The first feature index data and the model feature values ​​are normalized using the logarithmic transformation method.

4. The method according to claim 1, characterized in that, The method further includes: adjusting the target model and / or reacquiring the sample data in response to the deviation value being greater than or equal to a preset deviation threshold.

5. The method according to claim 1, characterized in that, After generating the portrait of the target object based on the tag data of the target object, the method further includes: In response to the need to search for the profile of the target object, a distributed search and analysis engine is used to search for the profile of the target object.

6. A portrait generation device for a target object, characterized in that, The device includes: The data acquisition module is used to acquire sample data of at least one target object and determine the first feature index data in the sample data; The clustering calculation module is used to perform clustering calculations using at least the first feature index data to determine the central feature value of the cluster center; The label data determination module is used to calculate the deviation value between the central feature value and the model feature value of the preset target model, and determine the label data of the target object based on the deviation value, the first feature index data and the model feature value. The target model is determined based on the third feature index data of the target location, and the model feature value is determined based on the third feature index data. The profile generation module is used to generate a profile of the target object based on the tag data of the target object. The clustering calculation module includes: The indicator data determination module is used to determine the second characteristic indicator data in the first characteristic indicator data, and to determine the remaining characteristic indicator data in the first characteristic indicator data excluding the second characteristic indicator data, wherein the number of the second characteristic indicator data is less than the number of the first characteristic indicator data. The clustering calculation submodule is used to perform clustering calculations on the second feature index data to determine the initial feature values ​​of the initial cluster centers. The center feature value determination module is used to determine the center feature value of the cluster center based on the initial feature value and the remaining feature index data; The central feature value determination module includes: The similarity calculation module is used to calculate the similarity between the initial feature value and the remaining feature index data; The first clustering module is used to allocate the remaining feature index data to the data corresponding to the initial cluster center in response to the similarity being less than a preset similarity threshold, thereby obtaining the first cluster; The feature value determination module is used to calculate the first feature value of the first cluster center of the first cluster, and in response to the fact that the similarity between the first feature value and the remaining feature index data is less than a preset similarity threshold, the first feature value is determined to be the center feature value of the cluster center. The secondary allocation module is used to reassign the remaining feature index data to the data corresponding to the initial cluster center in response to the similarity between the first feature value and the remaining feature index data being greater than or equal to a preset similarity threshold, until the similarity between the first feature value and the remaining feature index data is less than the preset similarity threshold. The label data determination module is further configured to, in response to the deviation value being less than a preset deviation threshold, calculate a second deviation value between the first feature index data and the central feature value, determine the data score of the first feature index data based on the second deviation value, and determine the label data of the target object based on the data score of the first feature index data; The type score determination module is used to determine the type score corresponding to the target type based on the target type corresponding to the tag data of the target object and the data score of the first feature index data; Accordingly, the portrait generation module is also used to generate a portrait of the target object based on the tag data of the target object and the type score corresponding to the target type.

7. The apparatus according to claim 6, characterized in that, The clustering calculation module is further configured to perform clustering calculations on the first feature index data to obtain the feature values ​​of the cluster centers corresponding to the first feature index data; and to determine the central feature values ​​of the cluster centers based on the feature values ​​of the cluster centers corresponding to the first feature index data.

8. The apparatus according to claim 6, characterized in that, The device further includes: The normalization module is used to perform normalization processing on the first feature index data and the model feature values ​​using the logarithmic transformation method.

9. The apparatus according to claim 6, characterized in that, The device further includes: The search module is used to search for the profile of the target object using a distributed search and analysis engine in response to the need to search for the profile of the target object.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

12. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images