Talent evaluation analysis method, device, equipment, medium and product
Through the method of training multiple machine learning models and verifying feature weights, the problem of subjective judgment impact in talent evaluation analysis is solved, and the objectivity and accuracy of talent evaluation is achieved.
Patent Information
- Application Number
- CN202411829930.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-12
- Publication Date
- 2025-04-15
AI Technical Summary
Talent evaluation analysis methods are easily affected by subjective judgments, resulting in a lack of objectivity in the scoring results.
By obtaining multi-source talent data, training multiple machine learning models based on these data, calculating the variance value between the feature weights of each model in the model set and the custom feature weights, verifying the feature weights, obtaining the average level feature weights, and talent evaluation is performed based on this.
The accurate acquisition of the average level feature weight is achieved, the influence of subjective judgment is avoided, the objectivity and accuracy of talent evaluation is ensured, and the comprehensive evaluation of talents is achieved.
Smart Images
Figure CN119831405A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of talent evaluation and analysis, and in particular to a talent evaluation and analysis method, device, equipment, medium and product. Background Art
[0002] With the development of science and technology, various technologies have been rapidly developed. Among them, there are more and more people with professional knowledge or special skills in all walks of life. Therefore, talent evaluation has become an important means to select outstanding talents.
[0003] The talent evaluation analysis method adopts the hierarchical analysis method. When determining the importance rating, all factors are compared in pairs, and a relative scale is used to reduce the difficulty of comparing factors of different natures and improve accuracy. However, the hierarchical analysis method may be affected by subjective judgment during the comparison process, resulting in a lack of objectivity in the scoring results. Summary of the invention
[0004] In view of this, the present invention provides a talent evaluation and analysis method, device, equipment, medium and product to solve the problem that the talent evaluation and analysis method is easily affected by subjective judgment, resulting in a lack of objectivity in the scoring results.
[0005] In a first aspect, the present invention provides a talent evaluation and analysis method, the method comprising:
[0006] Obtain multi-source talent data, train multiple machine learning models based on the multi-source talent data, and obtain a model set;
[0007] Get the custom feature weights, and calculate the variance between the feature weights of each model in the model set and the custom feature weights;
[0008] The variance value is compared with the preset threshold condition, and the feature weight is verified based on the comparison result to obtain the average level feature weight;
[0009] Talent evaluation is performed based on the average level feature weights to obtain talent evaluation scores.
[0010] A talent evaluation and analysis method provided in this embodiment trains multiple machine learning models with multi-source talent data to obtain a model set, calculates the variance value between the feature weight of each model in the model set and the custom feature weight, compares the variance value with a preset threshold condition, verifies the feature weight based on the comparison result to obtain the average level feature weight, and then evaluates the talent based on the average level feature weight to obtain a talent evaluation score, wherein the feature weight verification is performed using the comparison result of the variance value between the feature weight of each model and the custom feature weight and the preset threshold condition, thereby avoiding the influence of subjective judgment and achieving accurate acquisition of the average level feature weight, and performing talent evaluation through the average level feature weight, thereby ensuring the objectivity and accuracy of the talent evaluation and achieving a comprehensive assessment of the talent.
[0011] In an optional implementation, multiple machine learning models are trained based on multi-source talent data to obtain a model set, including:
[0012] Preprocessing the multi-source talent data, and dividing the preprocessed multi-source talent data into a feature vector space and a target vector space;
[0013] Divide the feature vector space and target vector space into training set and test set respectively;
[0014] The training set is used to train multiple machine learning models to obtain a model set.
[0015] A talent evaluation and analysis method provided in this embodiment divides the feature vector space and target vector space obtained by dividing the preprocessed talent multi-source data into a training set and a test set, and then uses the training set to train multiple machine learning models, thereby improving the prediction performance of the models in the model set and providing a data basis for the subsequent calculation of the variance value between the feature weights of each model in the model set and the custom feature weights.
[0016] In an optional implementation, the variance value is compared with a preset threshold condition, and feature weight verification is performed based on the comparison result to obtain the average level feature weight, including:
[0017] The variance value is compared with the preset threshold condition. If the variance value meets the preset threshold condition, the custom feature weight or the model weight feature corresponding to the minimum variance value is used as the average horizontal feature weight.
[0018] A talent evaluation and analysis method provided in this embodiment realizes accurate acquisition of the average level feature weight by selecting the model weight feature corresponding to the custom feature weight or the minimum variance value as the average level feature weight when the variance value meets the preset threshold condition.
[0019] In an optional implementation, the variance value is compared with a preset threshold condition, and feature weight verification is performed based on the comparison result to obtain the average level feature weight, further comprising:
[0020] If the variance value does not meet the preset threshold condition, the model validation index is used to validate and screen each model in the model set based on the test set to obtain the optimal model, and the feature weight corresponding to the optimal model is used as the average level feature weight.
[0021] A talent evaluation and analysis method provided in this embodiment uses a model verification indicator to verify and screen each model in the model set based on a test set when the variance value does not meet the preset threshold condition, and uses the feature weight corresponding to the optimal model as the horizontal level feature weight, thereby realizing the screening of the optimal features, making the horizontal level feature weight consistent with the actual data, and improving the accuracy and objectivity of the talent evaluation.
[0022] In an optional implementation, it also includes:
[0023] Comprehensive evaluation and analysis is performed based on the talent evaluation scores to obtain talent analysis results.
[0024] The talent evaluation and analysis method provided in this embodiment achieves the evaluation of the overall performance of talents and the exploration of the potential of talents in specific fields by performing comprehensive evaluation and analysis on talent evaluation scores.
[0025] In an optional implementation, a comprehensive evaluation analysis is performed based on the talent evaluation scores to obtain talent analysis results, including:
[0026] Build evaluation forms based on talent evaluation scores;
[0027] Arrange the talent evaluation scores in the evaluation form in descending order, and based on the descending order results, select a preset number of talent evaluation scores to construct an excellent talent form;
[0028] Obtain the preset preferred number, screen talents based on the preset preferred number, and obtain a list of experts;
[0029] Compare and analyze the list of specialists and the list of outstanding talents to obtain the preferred list;
[0030] Determine talent analysis results based on evaluation forms, talent forms, expert lists, and preference forms.
[0031] The talent evaluation and analysis method provided in this embodiment performs comprehensive evaluation and analysis through talent evaluation scores, and constructs evaluation forms, talent forms, expert lists and preference forms according to the comprehensive analysis results, thereby achieving a comprehensive evaluation of talents and the mining of talent potential.
[0032] In a second aspect, the present invention provides a talent evaluation and analysis device, the device comprising:
[0033] The training module is used to obtain multi-source talent data, train multiple machine learning models based on the multi-source talent data, and obtain a model set;
[0034] A calculation module is used to obtain custom feature weights and calculate the variance value between each model feature weight in the model set and the custom feature weight;
[0035] A comparison module, used to compare the variance value with a preset threshold condition, perform feature weight verification based on the comparison result, and obtain an average level feature weight;
[0036] The talent evaluation module is used to evaluate talents based on average level feature weights to obtain talent evaluation scores.
[0037] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the talent evaluation and analysis method of the first aspect or any corresponding embodiment thereof by executing the computer instructions.
[0038] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the talent evaluation and analysis method of the first aspect or any corresponding embodiment thereof.
[0039] In a fifth aspect, the present invention provides a computer program product, including computer instructions, which are used to enable a computer to execute the talent evaluation and analysis method of the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0041] Figure 1 is a flow chart of a talent evaluation and analysis method according to an embodiment of the present invention;
[0042] Figure 2 is a flow chart of another talent evaluation and analysis method according to an embodiment of the present invention;
[0043] Figure 3is a flow chart of another talent evaluation and analysis method according to an embodiment of the present invention;
[0044] Figure 4 is a structural block diagram of a talent evaluation and analysis device according to an embodiment of the present invention;
[0045] Figure 5 It is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION
[0046] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.
[0047] The embodiment of the present invention provides a talent evaluation and analysis method. It should be noted that the talent evaluation and analysis method provided by the embodiment of the present invention can be executed by a talent evaluation and analysis device, which can be implemented as part or all of an electronic device through software, hardware, or a combination of software and hardware. The electronic device can be a server or a terminal. The server in the embodiment of the present application can be a single server or a server cluster composed of multiple servers. The terminal in the embodiment of the present application can be a smart phone, a personal computer, a tablet computer, an intelligent robot, or other intelligent hardware devices. In the following method embodiments, the execution subject is an electronic device as an example for explanation.
[0048] According to an embodiment of the present invention, an embodiment of a talent evaluation and analysis method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.
[0049] In this embodiment, a talent evaluation and analysis method is provided, which can be used in the above-mentioned electronic device. Figure 1 is a flow chart of a talent evaluation and analysis method according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:
[0050] Step S101, obtaining multi-source talent data, training multiple machine learning models based on the multi-source talent data, and obtaining a model set.
[0051] Specifically, the multi-source talent data includes the source data in the scientific research talent element library and the engineering talent element library. The characteristics of the multi-source talent data include but are not limited to the number of patent publications, the number of works, the number of papers, journal levels, impact factors, project sources, project levels, project funds, project roles, and the amount of results transformation.
[0052] Furthermore, the classifier algorithms corresponding to the multiple machine learning models include but are not limited to Bayesian algorithm, linear regression algorithm, decision tree algorithm, random forest algorithm, etc.
[0053] Step S102, obtaining the custom feature weights, and respectively calculating the variance values between the feature weights of each model in the model set and the custom feature weights.
[0054] Step S103, comparing the variance value with a preset threshold condition, performing feature weight verification based on the comparison result, and obtaining a horizontal feature weight.
[0055] Step S104, performing talent evaluation based on the average level feature weights to obtain a talent evaluation score.
[0056] Specifically, talent analysis is conducted based on the average level feature weights, in-depth analysis is conducted on the specific features of each talent, and feedback and output are provided on the talent's features that need to be strengthened in the assessment and evaluation, as well as their features that currently have higher scores.
[0057] Furthermore, based on the average level feature weights, the features with lower and higher scores for each talent in the prediction process are calculated, and the calculation results are fed back to the corresponding talents.
[0058] A talent evaluation and analysis method provided in this embodiment trains multiple machine learning models with multi-source talent data to obtain a model set, calculates the variance value between the feature weight of each model in the model set and the custom feature weight, compares the variance value with a preset threshold condition, verifies the feature weight based on the comparison result to obtain the average level feature weight, and then evaluates the talent based on the average level feature weight to obtain a talent evaluation score, wherein the feature weight verification is performed using the comparison result of the variance value between the feature weight of each model and the custom feature weight and the preset threshold condition, thereby avoiding the influence of subjective judgment and achieving accurate acquisition of the average level feature weight, and performing talent evaluation through the average level feature weight, thereby ensuring the objectivity and accuracy of the talent evaluation and achieving a comprehensive assessment of the talent.
[0059] In this embodiment, a talent evaluation and analysis method is provided, which can be used in the above-mentioned electronic device. Figure 2 is a flow chart of a talent evaluation and analysis method according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps:
[0060] Step S201, obtain multi-source talent data, train multiple machine learning models based on the multi-source talent data, and obtain a model set.
[0061] Specifically, the above step S201 includes:
[0062] Step S2011, preprocessing the talent multi-source data, and dividing the preprocessed talent multi-source data into a feature vector space and a target vector space.
[0063] Specifically, the preprocessing of multi-source talent data includes but is not limited to outlier processing and feature processing, wherein outlier processing is used to identify and process observations in multi-source talent data that deviate significantly from other data points, and its specific steps include: first calculating the mean and standard deviation of the data, and then identifying values whose mean and standard deviation exceed a certain interval (i.e., outliers), and replacing the outliers with statistics such as the mean, median or mode; feature processing is used to perform steps such as data cleaning and data standardization on multi-source talent data, specifically, including but not limited to data scaling of data whose values deviate from the normal range, scaling them to the interval [0,1], and generating new features by combining and transforming existing features, etc.
[0064] Furthermore, the feature vector space refers to a collection of feature dimensions of the element library, where each group of vectors is regarded as an independent feature, and each point in the vector group represents the specific value corresponding to the feature. The feature vector space can be expressed as:
[0065]
[0066] Among them, F represents the feature vector space, which contains v features, n data, and f i represents the feature dimension of the i-th feature, f ij represents an eigenvalue, where 1 <j<n,i<v。
[0067] Furthermore, the target vector space refers to a set of historical scores. In the target vector space, each vector value represents the score value of a certain feature. The target vector space can be expressed as:
[0068]
[0069] Among them, S represents the target vector space, each s i Represents a target score value.
[0070] Step S2012, dividing the feature vector space and the target vector space into a training set and a test set respectively.
[0071] Specifically, the feature vector space and the target vector space are divided into a training set and a test set respectively according to preset proportions.
[0072] Step S2013, use the training set to train multiple machine learning models to obtain a model set.
[0073] Specifically, the training sets are input into multiple machine learning models for training, and then a model set is constructed based on a set of trained machine learning models. The model set can be expressed as:
[0074] M={m1 m2…m g} (4)
[0075] Where M represents the model set, m1, m2, ..., m g Represents the trained machine learning model. There are g trained machine learning models in the model set.
[0076] Step S202, obtaining the custom feature weights, and respectively calculating the variance values between the feature weights of each model in the model set and the custom feature weights.
[0077] Specifically, the feature weight of each model in the model set can be expressed as:
[0078] W={w1 w2…w g} (5)
[0079]
[0080] Where W represents the set of model feature weights. The set of model feature weights W has the same number of elements as the model set M. i represents the feature weight of the i-th model, w iv represents the weight of the vth feature in the ith model.
[0081] Furthermore, the variance value between the custom feature weight and the feature weights of each model in the model set is calculated. The calculation formula of the variance value is as follows:
[0082]
[0083] Among them, E i represents the variance value, w ij Represents the model feature weight, w' j represents the custom feature weight, where the set of custom feature weights W′ can be expressed as:
[0084]
[0085] Furthermore, a variance list EList is constructed according to the variance values between the custom feature weights and the feature weights of each model in the model set.
[0086] Step S203, comparing the variance value with a preset threshold condition, performing feature weight verification based on the comparison result, and obtaining a horizontal feature weight.
[0087] Specifically, the above step S203 includes:
[0088] Step S2031, compare the variance value with the preset threshold condition. If the variance value meets the preset threshold condition, the custom feature weight or the model weight feature corresponding to the minimum variance value is used as the average level feature weight.
[0089] Specifically, the smaller the variance value is, the closer it is to the preset threshold condition. It can be expressed as follows to determine whether there is an element in the variance list that meets the preset threshold condition:
[0090] EList=if E i <threshold (9)
[0091] Here, threshold represents a preset threshold condition.
[0092] Furthermore, if there are elements in the variance list that meet the preset threshold condition, that is, EList>0, then the average level feature weight w * is the model weight corresponding to the minimum variance value in the variance list EList, or the average level feature weight w * Custom feature weights.
[0093] Step S2032: If the variance value does not meet the preset threshold condition, then based on the test set, the model validation index is used to validate and screen each model in the model set to obtain the optimal model, and the feature weight corresponding to the optimal model is used as the average level feature weight.
[0094] Specifically, if there is no element in the variance list that meets the preset threshold condition, that is, EList<1, the test set is used to verify each model in the model set, and the optimal model is selected according to the model verification index, and the horizontal level feature weight is set to the feature weight corresponding to the optimal model.
[0095] Furthermore, the data in the test set are input into each model in the model set one by one for prediction to obtain the target value predicted by the model. The target value predicted by the model can be expressed as:
[0096]
[0097] Among them, S iRepresents the target value predicted by the ith model.
[0098] Furthermore, the optimal model is selected according to the model validation index. The calculation formula of model validation is as follows:
[0099]
[0100] Among them, s ij represents the predicted target value, t i Represents the evaluation index value of the model.
[0101] Furthermore, the evaluation index value list T of the models in the model set can be expressed as:
[0102] T={t1 t2…t g} (12)
[0103] Furthermore, the optimal model M * is the minimum t i The corresponding model has the average level feature weight w * The optimal model M * The corresponding feature weights.
[0104] Step S204: Perform talent evaluation based on the average level feature weights to obtain talent evaluation scores. Figure 1 Step S104 of the illustrated embodiment will not be described in detail here.
[0105] A talent evaluation and analysis method provided in this embodiment divides the feature vector space and target vector space obtained by dividing the pre-processed talent multi-source data into a training set and a test set, and then uses the training set to train multiple machine learning models, thereby improving the prediction performance of the models in the model set and providing a data basis for the subsequent calculation of the variance value between the feature weights of each model in the model set and the custom feature weights; secondly, when the variance value meets the preset threshold condition, by selecting the model weight feature or the custom feature weight corresponding to the minimum variance value as the horizontal level feature weight, the accurate acquisition of the horizontal level feature weight is achieved; when the variance value does not meet the preset threshold condition, based on the test set, the model verification indicator is used to verify and screen each model in the model set, and the feature weight corresponding to the optimal model is used as the horizontal level feature weight, thereby achieving the screening of the optimal feature, so that the horizontal level feature weight is consistent with the actual data, and the accuracy and objectivity of the talent evaluation are improved.
[0106] In this embodiment, a talent evaluation and analysis method is provided, which can be used in the above-mentioned electronic device. Figure 3 is a flow chart of a talent evaluation and analysis method according to an embodiment of the present invention. Figure 3 As shown, the process includes the following steps:
[0107] Step S301, obtain multi-source talent data, train multiple machine learning models based on the multi-source talent data, and obtain a model set. Figure 2 Step S201 of the illustrated embodiment will not be described in detail here.
[0108] Step S302: Obtain the custom feature weights, and calculate the variance between the feature weights of each model in the model set and the custom feature weights. Figure 2 Step S202 of the illustrated embodiment will not be described in detail here.
[0109] Step S303, compare the variance value with the preset threshold condition, perform feature weight verification based on the comparison result, and obtain the average level feature weight. Figure 2 Step S203 of the illustrated embodiment will not be described in detail here.
[0110] Step S304: Perform talent evaluation based on the average level feature weights to obtain talent evaluation scores. Figure 2 Step S204 of the illustrated embodiment will not be described in detail here.
[0111] Step S305: Perform comprehensive evaluation and analysis based on the talent evaluation scores to obtain talent analysis results.
[0112] Specifically, the above step S305 includes:
[0113] Step S3051, constructing an evaluation form based on the talent evaluation scores.
[0114] Specifically, we collect the data set of talents to be evaluated and use the average level feature weight w * Predict the talent evaluation scores in the data set, and then construct an evaluation form based on the talent evaluation scores. The evaluation form P can be expressed as:
[0115]
[0116] Among them, p ij represents the score of the i-th talent corresponding to the j-th feature, c i represents the evaluation score of the i-th talent, i <h,P h Represents the evaluation form of the hth talent.
[0117] Furthermore, the higher characteristic score values among the characteristic score values corresponding to each talent reflect the current advantages of the talent in the scoring system, while the characteristics with lower characteristic score values indicate the weak links in the scoring system. The characteristics with higher characteristic score values and the characteristics with lower characteristic score values to be strengthened will be fed back to the relevant talents for their reference.
[0118] Step S3052: sort the talent evaluation scores in the evaluation form in descending order, and based on the descending order result, select a preset number of talent evaluation scores to construct an excellent talent form.
[0119] Specifically, the talent evaluation score c i Sort from high to low, and select the top-ranked talent evaluation scores to build a list of outstanding talents.
[0120] Step S3053, obtaining a preset preference number, performing talent screening based on the preset preference number, and obtaining a list of experts.
[0121] Specifically, talents are screened one by one according to the preset preference quantity, talents with higher characteristic score values corresponding to different characteristics are summarized, and the names of the talents are integrated into the list of talents.
[0122] Step S3054, compare and analyze the list of experts and the list of outstanding talents to obtain a preferred list.
[0123] Specifically, the list of specialists is compared and analyzed with the list of outstanding talents, and a list of talents that are not included in the list of outstanding talents is screened out, and this list of talents is used as a preference list.
[0124] Step S3055, determining the talent analysis results based on the evaluation form, talent form, expert list and preference form.
[0125] The talent evaluation and analysis method provided in this embodiment performs comprehensive evaluation and analysis through talent evaluation scores, and constructs evaluation forms, talent forms, expert lists and preference forms according to the comprehensive analysis results, thereby achieving a comprehensive evaluation of talents and the mining of talent potential.
[0126] The specific steps of a talent evaluation and analysis method are described below through a specific embodiment.
[0127] Embodiment 1:
[0128] The specific steps of a talent evaluation and analysis method include:
[0129] S1, obtain multi-source talent data, feature weights set by users (i.e., custom feature weights) and preset threshold conditions, and continue to S2;
[0130] S2, preprocessing the multi-source talent data, and dividing the multi-source talent data into a feature vector space and a target vector space, and continuing to S3;
[0131] S3, determine whether the algorithm model component module input standard is met, if not, return to S3, if yes, continue to S4;
[0132] S4, dividing the feature vector space and the target vector space into a test set and a training set respectively, and continuing to S5;
[0133] S5, train multiple machine learning models using the training set, and then build a group of models, i.e., a model set, and continue to S6;
[0134] S6, calculating the variance value between the weight set by the user and the weight of each model feature in the model set, and constructing a corresponding variance list, and continuing to S7;
[0135] S7, determine whether there is an element in the variance list that meets the preset threshold condition. If so, set the average level feature weight to the user-defined feature weight or the model weight feature corresponding to the minimum variance value, and continue to S9; if not, continue to S8;
[0136] S8, using the test set to verify each model in the model set, and based on the relevant indicators of model verification, screen out the optimal model, set the average level feature weight as the feature weight corresponding to the optimal model, and continue to S9;
[0137] S9. Calculate the features with lower and higher scores for each talent in the prediction process according to the average level feature weights, and feed back the calculation results to the corresponding talent, and continue to S10;
[0138] S10. Calculate the evaluation scores of all talents according to the average level of feature weights, generate an evaluation form, and sort them according to the scores, and then generate a form of excellent talents with corresponding top rankings, and continue to S12;
[0139] S11. Screen talents one by one according to the preferred quantity, and further generate corresponding expert lists according to the frequency of appearance of expert talents, and continue to S12;
[0140] S12. Compare and analyze the list of specialists with the list of outstanding talents, screen out the list of talents that are not included in the list of outstanding talents, that is, the preference list, and output the preference list, evaluation form and outstanding talent form.
[0141] Embodiment 2:
[0142] A talent evaluation and analysis system is constructed for the talent evaluation and analysis method in the above-mentioned embodiment 1, including an input processing module, an algorithm model component module, a feature weight determination module, an optimal feature screening module, a talent assessment and analysis module, and a comprehensive evaluation and analysis module;
[0143] The input processing module is used to receive multi-source talent data, user-defined feature weights and preset threshold conditions, and perform pre-processing of the multi-source talent data on this basis. In this process, the multi-source talent data is divided into feature vector space and target vector space to ensure that the data set can meet the analysis requirements of the feature weight determination module;
[0144] The algorithm model component module is used to divide the feature vector space and the target vector space into a test set and a training set respectively, and then use the training set to train multiple machine learning algorithms, and finally generate a model set;
[0145] The feature weight determination module is used to calculate the variance between the user-defined weight and the feature weights of each model in the model set, and to determine whether there is a feature weight that meets the preset threshold requirements. If there is a feature weight that meets the conditions, its average level feature weight is set as the user-defined feature weight, or the model weight feature corresponding to the minimum variance value; if there is no feature weight that meets the conditions, the optimal feature screening module is performed;
[0146] The optimal feature screening module is used to use the test set data to predict each model in the model set one by one, screen out the optimal model based on the model verification indicators, and determine the features and feature weights corresponding to the optimal model. The optimal feature weight corresponding to the optimal model is set to the average level of feature weights;
[0147] The talent assessment analysis module is used to analyze talents based on the average level of feature weights, conduct in-depth analysis of the specific features of each talent, and provide feedback and output on the features that need to be strengthened in the assessment and evaluation of the talent and the features with higher current scores;
[0148] The comprehensive evaluation and analysis module scores talents according to the average level of feature weights, and generates a list of excellent talents for the top several talents to be extracted. Talents are screened one by one according to the preset number of preferences to determine the list of specialists. The list of specialists is compared and analyzed with the list of excellent talents, and the talents not included in the list constitute the preference list, and the preference list, evaluation form and excellent talent form are output.
[0149] In the above embodiment, a detailed and comprehensive weight verification process is proposed, which can use machine learning algorithms to determine the optimal weights, thereby providing a comprehensive evaluation for each talent, clearly pointing out the strengths and weaknesses of each talent in various aspects, which not only helps to understand the overall performance of talents, but also can deeply explore the potential in specific areas. Moreover, by using this system, not only can evaluation scores be obtained, but also experts or specialists with specific expertise can be found according to preferences.
[0150] In this embodiment, a talent evaluation and analysis device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and will not be repeated hereafter. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.
[0151] This embodiment provides a talent evaluation and analysis device, such as Figure 4 As shown, including:
[0152] A training module 401 is used to obtain multi-source talent data, and train multiple machine learning models based on the multi-source talent data to obtain a model set;
[0153] A calculation module 402 is used to obtain a custom feature weight and calculate the variance value between each model feature weight in the model set and the custom feature weight;
[0154] Comparison module 403, used to compare the variance value with a preset threshold condition, perform feature weight verification based on the comparison result, and obtain an average level feature weight;
[0155] The talent evaluation module 404 is used to perform talent evaluation based on the average level feature weights to obtain a talent evaluation score.
[0156] In some optional implementations, the training module 401 includes:
[0157] A preprocessing unit, used for preprocessing the multi-source talent data, and dividing the preprocessed multi-source talent data into a feature vector space and a target vector space;
[0158] A partitioning unit, used for partitioning the feature vector space and the target vector space into a training set and a test set respectively;
[0159] The training unit is used to train multiple machine learning models using the training set to obtain a model set.
[0160] In some optional implementations, the comparison module 403 includes:
[0161] The comparison unit is used to compare the variance value with a preset threshold condition. If the variance value meets the preset threshold condition, the custom feature weight or the model weight feature corresponding to the minimum variance value is used as the average level feature weight.
[0162] In some optional implementations, the comparison module 403 further includes:
[0163] The verification and screening unit is used to verify and screen each model in the model set based on the test set using the model verification indicator if the variance value does not meet the preset threshold condition, obtain the optimal model, and use the feature weight corresponding to the optimal model as the average level feature weight.
[0164] In some optional embodiments, the method further comprises:
[0165] The comprehensive evaluation and analysis module is used to conduct comprehensive evaluation and analysis based on the talent evaluation scores to obtain talent analysis results.
[0166] In some optional implementations, the comprehensive evaluation and analysis module includes:
[0167] A construction unit for constructing an evaluation form based on talent evaluation scores;
[0168] An arrangement unit, used to arrange the talent evaluation scores in the evaluation form in descending order, and based on the descending arrangement result, select a preset number of talent evaluation scores to construct an excellent talent form;
[0169] A talent screening unit is used to obtain a preset preference number, screen talents based on the preset preference number, and obtain a list of experts;
[0170] The analysis unit is used to compare and analyze the list of specialists and the list of outstanding talents to obtain a preferred list;
[0171] The determination unit is used to determine the talent analysis results based on the evaluation form, talent form, expert list and preference form.
[0172] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.
[0173] A talent evaluation and analysis device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.
[0174] The embodiment of the present invention also provides a computer device having the above Figure 4 A talent evaluation and analysis device is shown.
[0175] See also Figure 5 , Figure 5 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Figure 5As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 5 A processor 10 is taken as an example.
[0176] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.
[0177] The memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.
[0178] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
[0179] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.
[0180] The computer device also includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 may be connected via a bus or other means. Figure 5 The example of connecting through bus is taken in the following.
[0181] The input device 30 can receive input digital or character information, and generate key signal input related to the user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The above-mentioned display device includes but is not limited to a liquid crystal display, a light emitting diode, a display and a plasma display. In some optional embodiments, the display device can be a touch screen.
[0182] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.
[0183] A part of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the existence of the computer program instruction in a computer-readable medium includes, but is not limited to, a source file, an executable file, an installation package file, etc., and accordingly, the way in which the computer program instruction is executed by the computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium accessible to the computer.
[0184] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.
Claims
1. A talent evaluation and analysis method, characterized in that: The method comprises: Acquire multi-source talent data, and train multiple machine learning models based on the multi-source talent data to obtain a model set; Obtaining a custom feature weight, and respectively calculating a variance value between each model feature weight in the model set and the custom feature weight; Compare the variance value with a preset threshold condition, perform feature weight verification based on the comparison result, and obtain an average level feature weight; Talent evaluation is performed based on the average level feature weights to obtain a talent evaluation score.
2. The method according to claim 1, characterized in that The training of the multiple machine learning models based on the multi-source talent data to obtain a model set includes: Preprocessing the talent multi-source data, and dividing the preprocessed talent multi-source data into a feature vector space and a target vector space; Dividing the feature vector space and the target vector space into a training set and a test set respectively; The multiple machine learning models are trained using the training set to obtain the model set.
3. The method according to claim 2, characterized in that The comparing the variance value with a preset threshold condition, and performing feature weight verification based on the comparison result to obtain the average level feature weight, includes: The variance value is compared with a preset threshold condition. If the variance value meets the preset threshold condition, the model weight feature corresponding to the custom feature weight or the minimum variance value is used as the horizontal level feature weight.
4. The method according to claim 3, characterized in that The step of comparing the variance value with a preset threshold condition, and performing feature weight verification based on the comparison result to obtain the average level feature weight, further includes: If the variance value does not meet the preset threshold condition, then based on the test set, each model in the model set is verified and screened using a model verification indicator to obtain an optimal model, and the feature weight corresponding to the optimal model is used as the average level feature weight.
5. The method according to claim 1, characterized in that Also includes: A comprehensive evaluation and analysis is performed based on the talent evaluation scores to obtain talent analysis results.
6. The method according to claim 5, characterized in that The comprehensive evaluation and analysis based on the talent evaluation scores to obtain talent analysis results includes: Constructing an evaluation form based on the talent evaluation scores; Arrange the talent evaluation scores in the evaluation form in descending order, and based on the descending order result, select a preset number of talent evaluation scores to construct an excellent talent form; Obtaining a preset number of preferences, performing talent screening based on the preset number of preferences, and obtaining a list of talents; Compare and analyze the list of experts and the list of outstanding talents to obtain a preferred list; The talent analysis result is determined based on the evaluation form, the talent form, the expert list and the preference form.
7. A talent evaluation and analysis device, characterized in that: The device comprises: A training module, used to obtain multi-source talent data, and train multiple machine learning models based on the multi-source talent data to obtain a model set; A calculation module, used to obtain a custom feature weight, and respectively calculate a variance value between each model feature weight in the model set and the custom feature weight; A comparison module, used to compare the variance value with a preset threshold condition, perform feature weight verification based on the comparison result, and obtain an average level feature weight; The talent evaluation module is used to perform talent evaluation based on the average level feature weights to obtain a talent evaluation score.
8. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the talent evaluation and analysis method according to any one of claims 1 to 6 by executing the computer instructions.
9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the talent evaluation and analysis method according to any one of claims 1 to 6.
10. A computer program product, characterized in that The method comprises computer instructions for causing a computer to execute the talent evaluation and analysis method according to any one of claims 1 to 6.