Standardized talent visual intelligent evaluation method and system

Abstract

The invention discloses a standardized talent visual intelligent assessment method and system. The method comprises the steps of collecting talent related data from different platforms and performing preprocessing; preprocessing the collected related data of the talents, and performing feature extraction through a feature extraction model to obtain semantic features and numerical features; constructing an evaluation model, inputting the semantic features and the numerical features into the evaluation model for training and testing, and obtaining a comprehensive ability score and a post matching degree; and setting a division standard, carrying out talent grade division according to the comprehensive ability score and the post matching degree, and carrying out visual display on a division result. According to the method, the user can conveniently screen and deeply know the talent information, the talent evaluation efficiency and quality can be improved, and comprehensive, scientific and reliable support is provided for enterprises in the human resource management decision aspects such as talent selection, training and post adaptation.

Description

Technical Field

[0001] The present invention relates to the field of data analysis technology, and in particular to a standardized talent visualization intelligent evaluation method and system. Background Art

[0002] In the digital age, the critical role of human resource management in corporate development is becoming increasingly prominent. As market competition intensifies, companies' demands for talent are becoming more diversified and specialized. Accurate and efficient talent assessment has become a crucial means for companies to gain a competitive advantage. Traditional talent assessment methods, such as interviews and leadership evaluations, rely primarily on subjective judgment. These methods are heavily influenced by individual subjective factors, lack objectivity and accuracy, and fail to meet companies' needs for comprehensive and scientific talent assessments. Furthermore, companies accumulate vast amounts of data during talent management. Unlocking the value of this data and achieving quantitative and visual talent evaluation has become a pressing challenge in the field of human resource management.

[0003] While some companies are beginning to leverage data analytics for talent assessment, numerous challenges remain. Data collection channels are fragmented, and data formats and standards vary significantly across platforms, making integration difficult. This leads to uneven data quality and impacts the accuracy of assessment results. Furthermore, existing feature extraction methods fail to fully tap the potential value of talent data, hindering the accuracy and generalization of assessment models. For example, when processing text-based talent descriptions, simple keyword matching approaches struggle to understand the semantics and context of the text and accurately extract key features. Furthermore, existing assessment models often overlook the varying talent requirements across industries and companies, lacking specificity and adaptability. Assessment results are often presented in simple data reports, lacking intuitiveness and interactivity, and thus failing to meet companies' needs for rapid and accurate access to talent information and decision-making. Summary of the Invention

[0004] The purpose of this section is to summarize some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the abstract and title of this application to avoid obscuring the purpose of this section, the abstract and the title of the invention, and such simplifications or omissions should not be used to limit the scope of the present invention.

[0005] In light of the above-mentioned existing problems, the present invention proposes a standardized visual intelligent talent assessment method to address existing talent assessment issues, such as the difficulty in data collection and integration, the lack of accuracy and generalization in data processing and model building, and the lack of intuitive and interactive presentation of assessment results.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a standardized talent visualization intelligent assessment method, comprising:

[0008] Collect talent-related data from different platforms and pre-process them;

[0009] After preprocessing the collected talent-related data, feature extraction is performed through a feature extraction model to obtain semantic features and numerical features;

[0010] Build an evaluation model, input semantic features and numerical features into the evaluation model for training and testing, and obtain comprehensive ability scores and job matching;

[0011] Set classification standards, classify talents according to comprehensive ability scores and job matching, and visualize the classification results.

[0012] As a preferred solution of the standardized talent visualization intelligent assessment method of the present invention, talent-related data is collected from different platforms and pre-processed, including:

[0013] Use data duplication checking algorithms to compare key information and remove duplicate data;

[0014] For missing values, if it is numerical data, use the mean or median to fill it according to the data distribution characteristics; if it is text data, fill it in or mark it according to the logical relationship between the data;

[0015] For outliers, statistical methods are used to identify and correct them.

[0016] As a preferred embodiment of the standardized talent visualization intelligent assessment method of the present invention, after preprocessing the collected talent-related data, feature extraction is performed using a feature extraction model to obtain semantic features and numerical features, including:

[0017] For text data, the text is input into the BERT model, which uses a multi-layer bidirectional Transformer encoder to capture the semantic information in the text and output a fixed-dimensional vector representation as a semantic feature.

[0018] For numerical data, the normalization method is used to obtain numerical features.

[0019] As a preferred solution of the standardized talent visualization intelligent assessment method of the present invention, it also includes:

[0020] The extracted semantic features and numerical features are selected by using the Pearson correlation coefficient method, which is expressed as:

[0021]

[0022] Where n is the number of samples, x i is the i-th observation value of feature X, is the mean of feature X, y i is the i-th observation value of the target variable Y, is the mean of the target variable Y, r xy is the Pearson correlation coefficient between feature X and target variable Y;

[0023] Set a correlation threshold, compare the calculated Pearson correlation coefficient with the correlation threshold, remove features whose absolute value of the correlation coefficient is less than the correlation threshold, and rank the remaining features according to the absolute value of the Pearson correlation coefficient.

[0024] As a preferred embodiment of the standardized talent visualization intelligent assessment method of the present invention, an assessment model is constructed, and semantic features and numerical features are input into the assessment model for training and testing to obtain a comprehensive ability score and job matching degree, including:

[0025] A multi-layer perceptron is used to build the evaluation model. The number of nodes in the input layer of the evaluation model is determined according to the dimensions of semantic features and numerical features, so that it is consistent with the number of features.

[0026] The processed semantic and numerical feature data are divided into training and test sets according to the set ratio; the evaluation model is trained on the training set, and the model parameters are updated using the stochastic gradient descent algorithm, with the mean square error selected as the loss function;

[0027] The forward propagation is performed based on the training set to calculate the predicted value, and the gradient of the loss function with respect to the weights and biases of each layer is calculated through the backpropagation algorithm. The parameters are adjusted according to the gradient to reduce the loss value, so that the model can learn the relationship between the features and the target value.

[0028] After training is completed, the test set is input into the evaluation model to obtain the accuracy of the comprehensive ability score and job matching.

[0029] As a preferred solution of the standardized talent visualization intelligent assessment method of the present invention, it also includes:

[0030] K-fold cross validation is used to evaluate the generalization ability of the model. The training data is evenly divided into K parts, and K-1 parts are selected as new training sets each time, and the remaining 1 part is used as the validation set.

[0031] The evaluation model is trained on different training sets in K cycles, verified on the corresponding validation sets, and the performance indicators of each verification are recorded. After completing K cycles, the average value of each indicator is calculated to obtain the model evaluation result.

[0032] As a preferred solution of the standardized talent visualization intelligent assessment method of the present invention, the following steps are performed: setting classification criteria, classifying talents according to comprehensive ability scores and job matching, and visually displaying the classification results, including:

[0033] Talents are divided into four different levels: A, B, C and D, based on the different intervals of comprehensive ability scores and job matching;

[0034] Based on the comprehensive ability scores and job matching obtained from the assessment model, all talents are graded according to the classification standards;

[0035] Synchronize the level classification results to the App and PC, bind the App's skill growth order data to the personal information database, and update the talent ability profile in real time;

[0036] The dynamic data of talents at all levels can be visualized on the PC webpage through the IP port, including the progress of skill ordering and ability improvement trajectory;

[0037] Access to the enterprise talent double-selection database is open to qualified talents. Talents that do not meet the requirements will automatically enter the elimination process and will not be included in the talent database display.

[0038] In a second aspect, the present invention provides a system for standardized visual intelligent assessment of talents, comprising:

[0039] Data acquisition module, used to collect talent-related data from different platforms and perform pre-processing;

[0040] The feature extraction module is used to pre-process the collected talent-related data and then extract features through the feature extraction model to obtain semantic features and numerical features;

[0041] The evaluation module is used to build an evaluation model, input semantic features and numerical features into the evaluation model for training and testing, and obtain comprehensive ability scores and job matching;

[0042] The visualization module is used to set classification criteria, classify talents according to comprehensive ability scores and job matching, and visualize the classification results.

[0043] In a third aspect, the present invention provides a computing device, comprising:

[0044] memory and processor;

[0045] The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the standardized talent visualization intelligent assessment method are implemented.

[0046] In a fourth aspect, the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the standardized talent visualization intelligent assessment method.

[0047] Compared with the existing technology, the present invention has the following beneficial effects: by collecting data from multiple platforms and performing feature extraction, the present invention obtains high-quality semantic and numerical features. The method based on the Pearson correlation coefficient effectively screens out key features, reduces dimensionality, avoids overfitting, and improves model training efficiency. A multi-layer perceptron evaluation model is used to accurately derive comprehensive ability scores and job matching degrees. Talent levels are divided based on the evaluation results, and the division results are visualized to facilitate user screening and in-depth understanding of talent information. This can improve the efficiency and quality of talent assessment and provide comprehensive, scientific, and reliable support for enterprises in human resource management decisions such as talent selection, training, and job matching. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort. Among them:

[0049] Figure 1 The figure is a schematic diagram of a general process of a standardized talent visualization intelligent assessment method according to one embodiment of the present invention. DETAILED DESCRIPTION

[0050] To make the above-mentioned objects, features, and advantages of the present invention more clearly understood, the following detailed description of the specific embodiments of the present invention is given in conjunction with the accompanying drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in this field without creative work should fall within the scope of protection of the present invention.

[0051] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0052] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0053] The present invention is described in detail with reference to schematic diagrams. For ease of illustration, cross-sectional views of device structures may be partially enlarged and not to scale when describing embodiments of the present invention. Furthermore, the schematic diagrams are merely illustrative and should not limit the scope of the present invention. Furthermore, in actual production, the three-dimensional dimensions of length, width, and depth should be included.

[0054] Furthermore, in the description of the present invention, it should be noted that the terms "upper, lower, inner, and outer" and other references to orientations or positional relationships are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely to facilitate and simplify the description of the present invention. They are not intended to indicate or imply that the systems or components referred to must have, be constructed, or operate in a specific orientation, and therefore should not be construed as limitations on the present invention. Furthermore, the terms "first, second, or third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0055] In this disclosure, unless otherwise specified or limited, the terms "mounted," "connected," and "connected" should be interpreted broadly. For example, they may refer to fixed, removable, or integral connections. They may also refer to mechanical, electrical, or direct connections, indirect connections through an intermediary, or internal communication between two components. Those skilled in the art will understand the specific meanings of these terms in this disclosure.

[0056] Example 1

[0057] Reference Figure 1 , which is an embodiment of the present invention, provides a standardized talent visualization intelligent assessment method, comprising:

[0058] S100: Collect talent-related data from different platforms and pre-process them;

[0059] Preferably, duplicate data is removed by comparing key information through a data duplication checking algorithm; for missing values, if they are numerical data, they are filled with the mean or median according to the data distribution characteristics; if they are text data, they are completed or marked according to the logical relationship between the data; for outliers, they are identified and corrected using statistical methods.

[0060] Specifically, when collecting talent-related data from different platforms, including the company's internal human resources management system, external online recruitment platforms, and databases of professional skills certification agencies, we obtain basic information about the talent, such as age, gender, education level, job position, years of work experience, job search intentions, expected salary, a detailed description of past work experience, and educational background.

[0061] During the data preprocessing stage, for duplicate data, a data duplication detection algorithm is used to carefully compare key information such as ID numbers, academic certificate numbers, and key items in work experience. Once a duplicate record is found, only one of them is retained to ensure the uniqueness of the data. For missing values ​​in numerical data, the distribution characteristics of the data are first analyzed. If the data presents a normal distribution, the mean filling method is used; if the data distribution is relatively discrete, the median is used for filling. For missing values ​​in text data, reasonable filling is performed based on the logical relationship between the data, such as the chronological order in work experience, the correlation between skill descriptions and job requirements, etc.; if accurate filling is not possible, it is marked for further processing. When dealing with outliers, statistical methods such as the 3σ principle are used to identify data that deviates from the normal range and correct or eliminate them according to the actual situation, thereby improving the quality and availability of the data.

[0062] S200: After pre-processing the collected talent-related data, feature extraction is performed using a feature extraction model to obtain semantic features and numerical features;

[0063] Preferably, for text-type data, the text is input into the BERT model, and the semantic information in the text is captured through a multi-layer bidirectional Transformer encoder, and a fixed-dimensional vector representation is output as a semantic feature; for numerical data, a normalization method is used to perform normalization processing to obtain numerical features.

[0064] Preferably, feature selection is performed on the extracted semantic features and numerical features by adopting a method based on the Pearson correlation coefficient, which is expressed as:

[0065]

[0066] Where n is the number of samples, x i is the i-th observation value of feature X, is the mean of feature X, y i is the i-th observation value of the target variable Y, is the mean of the target variable Y, r xy is the Pearson correlation coefficient between feature X and target variable Y;

[0067] Preferably, a correlation threshold is set, the calculated Pearson correlation coefficient is compared with the correlation threshold, features whose absolute value of the correlation coefficient is less than the correlation threshold are removed, and the importance of the remaining features is ranked according to the absolute value of the Pearson correlation coefficient.

[0068] It should be noted that the BERT model, with its multi-layer bidirectional Transformer encoder, deeply captures semantic information in text, effectively mining the underlying meaning of text data and outputting fixed-dimensional vector representations as semantic features. This significantly improves the accuracy and completeness of semantic features compared to traditional text feature extraction methods. For numerical data, normalization eliminates dimensionality differences between the data, aligning different numerical features with the same scale. This improves data comparability and usability, providing a sound data foundation for subsequent analysis and model training. Regarding feature selection, a method based on the Pearson correlation coefficient scientifically measures the linear correlation between each feature and the target variable. By setting a correlation threshold, it accurately removes features with low correlation, reducing feature dimensionality and model complexity, effectively avoiding overfitting. Furthermore, the remaining features are ranked by importance based on the absolute value of the Pearson correlation coefficient, providing a clear feature selection basis for the subsequent evaluation model construction. This allows the model to more efficiently learn data features, improves model training efficiency and predictive accuracy, and ultimately makes the final talent assessment results more reliable and accurate.

[0069] S300: Build an evaluation model, input semantic features and numerical features into the evaluation model for training and testing, and obtain a comprehensive ability score and job matching degree;

[0070] Preferably, a multi-layer perceptron is used to construct the evaluation model, and the number of nodes in the input layer of the evaluation model is determined according to the dimensions of the semantic features and the numerical features so that it is consistent with the number of features;

[0071] Preferably, the processed semantic and numerical feature data are divided into a training set and a test set according to a set ratio; the evaluation model is trained using the training set, and the model parameters are updated using a stochastic gradient descent algorithm, with the mean square error being selected as the loss function;

[0072] Preferably, forward propagation is performed based on the training set to calculate the predicted value, and the gradient of the loss function with respect to the weights and biases of each layer is calculated through the backpropagation algorithm. The parameters are adjusted according to the gradient to reduce the loss value so that the model can learn the relationship between the features and the target value. After the training is completed, the test set is input into the evaluation model to obtain the accuracy of the comprehensive ability score and job matching.

[0073] Specifically, after selecting the multi-layer perceptron to build the evaluation model, the number of input layer nodes is determined based on the dimensions of the semantic and numerical features, so that the input layer can fully receive all feature information. Assuming that after feature extraction and selection, the total number of semantic and numerical features is m, the number of input layer nodes is set to m.

[0074] The processed semantic and numerical feature data is divided into a training set and a test set in a certain ratio, such as 7:3. The training set is used to train the evaluation model. During the training process, the stochastic gradient descent algorithm is used to update the model parameters. The stochastic gradient descent algorithm randomly selects one or a small batch of samples from the training set, calculates the loss function value for these samples, and then adjusts the parameters based on the gradient of the loss function with respect to the weights and biases of each layer. The loss function is the mean squared error, which measures the average error between the model's predicted values ​​and the true values.

[0075] When performing forward propagation calculations based on the training set, the input layer receives feature data and passes it to the hidden layer. The neurons in the hidden layer perform weighted summation on the input data and perform nonlinear transformations through activation functions to learn the complex relationships in the data. After processing through multiple hidden layers, the output layer generates a predicted value, which is the preliminary prediction result of the comprehensive ability score and job matching degree.

[0076] The backpropagation algorithm calculates the gradient of the loss function with respect to the weights and biases of each layer. Starting from the output layer, the backpropagation algorithm propagates the gradient of the loss function with respect to the output layer back to the hidden and input layers, calculating the gradient of each parameter. Based on the calculated gradient, the weights and biases of each layer are adjusted at a certain learning rate to gradually reduce the loss value, allowing the model to continuously learn the relationship between features and the actual comprehensive ability score and job match.

[0077] After multiple iterations of training, model training is complete when the model's loss converges to a certain level or reaches a preset number of training iterations. The test set is then fed into the trained evaluation model, which then predicts the samples in the test set, yielding a comprehensive ability score and job match prediction.

[0078] Preferably, K-fold cross-validation is used to evaluate the generalization ability of the model, and the training data is evenly divided into K parts. K-1 parts are selected each time as new training sets, and the remaining 1 part is used as a validation set. The evaluation model is trained on different training sets in turn in K cycles, and validated on the corresponding validation sets. The performance indicators of each validation are recorded. After completing K cycles, the average value of each indicator is calculated to obtain the model evaluation result.

[0079] S400: Set classification criteria, classify talents based on comprehensive ability scores and job matching, and visualize the classification results;

[0080] Talents are divided into four different levels: A, B, C and D, based on the different intervals of comprehensive ability scores and job matching;

[0081] Based on the comprehensive ability scores and job matching obtained from the assessment model, all talents are graded according to the classification standards;

[0082] Synchronize the level classification results to the App and PC, bind the App's skill growth order data to the personal information database, and update the talent ability profile in real time;

[0083] The dynamic data of talents at all levels can be visualized on the PC webpage through the IP port, including the progress of skill ordering and ability improvement trajectory;

[0084] Access to the enterprise talent double-selection database is open to qualified talents. Talents that do not meet the requirements will automatically enter the elimination process and will not be included in the talent database display.

[0085] It should be noted that through the above-mentioned level division and dual-end linkage mechanism, on the one hand, the standardization and dynamic management of talent assessment results are achieved: the clear A / B / C / D level division combined with the comprehensive ability score and job matching provides companies with intuitive talent screening standards; the level results are synchronized to the App and PC sides, so that talents can view the ability portrait updates in real time through the App, and companies can visually track dynamic data such as skill order progress, ability improvement trajectory, etc. through the IP port on the PC side, solving the problems of terminal fragmentation and data lag in the traditional assessment system.

[0086] Furthermore, the application value of assessment results is enhanced: qualified candidates are automatically entered into the company's dual-selection database, streamlining the assessment and recruitment processes and improving job matching efficiency. Unqualified candidates are automatically eliminated and excluded from the database, ensuring the quality of the talent pool. Through dual-end data linkage and automated screening, the overall solution not only enhances the transparency and real-time nature of talent assessment, but also establishes a closed-loop ecosystem from assessment to application, providing efficient support for precise talent selection and targeted talent development.

[0087] The above is a schematic diagram of a standardized talent visualization intelligent assessment method according to this embodiment. It should be noted that the technical solution of this standardized talent visualization intelligent assessment system and the technical solution of the standardized talent visualization intelligent assessment method described above are based on the same concept. For details not described in detail in the technical solution of the standardized talent visualization intelligent assessment system according to this embodiment, please refer to the description of the technical solution of the standardized talent visualization intelligent assessment method described above.

[0088] The standardized talent visualization intelligent assessment system in this embodiment includes:

[0089] Data acquisition module, used to collect talent-related data from different platforms and perform pre-processing;

[0090] The feature extraction module is used to pre-process the collected talent-related data and then extract features through the feature extraction model to obtain semantic features and numerical features;

[0091] The evaluation module is used to build an evaluation model, input semantic features and numerical features into the evaluation model for training and testing, and obtain comprehensive ability scores and job matching;

[0092] The visualization module is used to set classification criteria, classify talents according to comprehensive ability scores and job matching, and visualize the classification results.

[0093] This embodiment further provides a computing device suitable for standardized visual intelligent assessment of talents, including:

[0094] Memory and processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the standardized talent visualization intelligent assessment method proposed in the above embodiment.

[0095] This embodiment also provides a storage medium having a computer program stored thereon. When the program is executed by a processor, the method for realizing standardized visual intelligent assessment of talents proposed in the above embodiment is implemented.

[0096] The storage medium proposed in this embodiment and the method for realizing standardized talent visualization intelligent assessment proposed in the above embodiment belong to the same inventive concept. For technical details not fully described in this embodiment, please refer to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0097] Through the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented with the help of software and necessary general-purpose hardware, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as a computer's floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods of various embodiments of the present invention.

[0098] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A standardized talent visualization intelligent assessment method, characterized by: include: Collect talent-related data from different platforms and pre-process them; After preprocessing the collected talent-related data, feature extraction is performed through a feature extraction model to obtain semantic features and numerical features; Build an evaluation model, input semantic features and numerical features into the evaluation model for training and testing, and obtain comprehensive ability scores and job matching; Set classification standards, classify talents according to comprehensive ability scores and job matching, and visualize the classification results.

2. The standardized talent visualization intelligent assessment method according to claim 1, characterized in that: Collect talent-related data from various platforms and pre-process it, including: Use data duplication checking algorithms to compare key information and remove duplicate data; For missing values, if it is numerical data, use the mean or median to fill it according to the data distribution characteristics; if it is text data, fill it in or mark it according to the logical relationship between the data; For outliers, statistical methods are used to identify and correct them.

3. The standardized talent visualization intelligent assessment method according to claim 1 or 2, characterized in that: After preprocessing the collected talent-related data, feature extraction is performed through the feature extraction model to obtain semantic features and numerical features, including: For text data, the text is input into the BERT model, which uses a multi-layer bidirectional Transformer encoder to capture the semantic information in the text and output a fixed-dimensional vector representation as a semantic feature. For numerical data, the normalization method is used to obtain numerical features.

4. The standardized talent visualization intelligent assessment method according to claim 3, characterized in that: Also includes: The extracted semantic features and numerical features are selected by using the Pearson correlation coefficient method, which is expressed as: Where n is the number of samples, x i is the i-th observation value of feature X, is the mean of feature X, y i is the i-th observation value of the target variable Y, is the mean of the target variable Y, r xy is the Pearson correlation coefficient between feature X and target variable Y; Set a correlation threshold, compare the calculated Pearson correlation coefficient with the correlation threshold, remove features whose absolute value of the correlation coefficient is less than the correlation threshold, and rank the remaining features according to the absolute value of the Pearson correlation coefficient.

5. The standardized talent visualization intelligent assessment method according to claim 4, characterized in that: Build an evaluation model, input semantic features and numerical features into the evaluation model for training and testing, and obtain a comprehensive ability score and job matching degree, including: A multi-layer perceptron is used to build the evaluation model. The number of nodes in the input layer of the evaluation model is determined according to the dimensions of semantic features and numerical features, so that it is consistent with the number of features. The processed semantic and numerical feature data are divided into training and test sets according to the set ratio; the evaluation model is trained on the training set, and the model parameters are updated using the stochastic gradient descent algorithm, with the mean square error selected as the loss function; The forward propagation is performed based on the training set to calculate the predicted value, and the gradient of the loss function with respect to the weights and biases of each layer is calculated through the backpropagation algorithm. The parameters are adjusted according to the gradient to reduce the loss value, so that the model can learn the relationship between the features and the target value. After training is completed, the test set is input into the evaluation model to obtain the accuracy of the comprehensive ability score and job matching.

6. The standardized talent visualization intelligent assessment method according to claim 5, characterized in that: Also includes: K-fold cross validation is used to evaluate the generalization ability of the model. The training data is evenly divided into K parts, and K-1 parts are selected as new training sets each time, and the remaining 1 part is used as the validation set. The evaluation model is trained on different training sets in K cycles, verified on the corresponding validation sets, and the performance indicators of each verification are recorded. After completing K cycles, the average value of each indicator is calculated to obtain the model evaluation result.

7. The standardized talent visualization intelligent assessment method according to claim 6, characterized in that: Set classification criteria, classify talents based on comprehensive ability scores and job matching, and visualize the classification results, including: Talents are divided into four different levels: A, B, C and D, based on the different intervals of comprehensive ability scores and job matching; Based on the comprehensive ability scores and job matching obtained from the assessment model, all talents are graded according to the classification standards; Synchronize the level classification results to the App and PC, bind the App's skill growth order data to the personal information database, and update the talent ability profile in real time; The dynamic data of talents at all levels can be visualized on the PC webpage through the IP port, including the progress of skill ordering and ability improvement trajectory; Access to the enterprise talent double-selection database is open to qualified talents. Talents that do not meet the requirements will automatically enter the elimination process and will not be included in the talent database display.

8. A standardized system for visual intelligent assessment of talents, characterized by: include, Data acquisition module, used to collect talent-related data from different platforms and perform pre-processing; The feature extraction module is used to pre-process the collected talent-related data and then extract features through the feature extraction model to obtain semantic features and numerical features; The evaluation module is used to build an evaluation model, input semantic features and numerical features into the evaluation model for training and testing, and obtain comprehensive ability scores and job matching; The visualization module is used to set classification criteria, classify talents according to comprehensive ability scores and job matching, and visualize the classification results.

9. An electronic device comprising: memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the standardized talent visualization intelligent assessment method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions, when executed by a processor, implement the steps of the standardized talent visualization intelligent assessment method according to any one of claims 1 to 7.

Images