CN117035710A - Talent assessment quantification method, system, equipment and medium - Google Patents
Talent assessment quantification method, system, equipment and medium Download PDFInfo
- Publication number
- CN117035710A CN117035710A CN202310882800.XA CN202310882800A CN117035710A CN 117035710 A CN117035710 A CN 117035710A CN 202310882800 A CN202310882800 A CN 202310882800A CN 117035710 A CN117035710 A CN 117035710A
- Authority
- CN
- China
- Prior art keywords
- evaluation
- data
- weight
- model
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 103
- 238000011002 quantification Methods 0.000 title claims abstract description 31
- 238000011156 evaluation Methods 0.000 claims abstract description 214
- 238000004364 calculation method Methods 0.000 claims abstract description 10
- 238000000926 separation method Methods 0.000 claims abstract description 5
- 238000012549 training Methods 0.000 claims description 68
- 230000006870 function Effects 0.000 claims description 32
- 230000008569 process Effects 0.000 claims description 30
- 238000010801 machine learning Methods 0.000 claims description 23
- 238000007781 pre-processing Methods 0.000 claims description 20
- 238000004422 calculation algorithm Methods 0.000 claims description 19
- 238000012360 testing method Methods 0.000 claims description 14
- 238000010606 normalization Methods 0.000 claims description 11
- 238000004140 cleaning Methods 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 7
- 238000011835 investigation Methods 0.000 claims description 7
- 238000012544 monitoring process Methods 0.000 claims description 7
- 238000003672 processing method Methods 0.000 claims description 6
- 238000003860 storage Methods 0.000 claims description 2
- 230000007115 recruitment Effects 0.000 abstract description 11
- 238000012545 processing Methods 0.000 description 9
- 238000007726 management method Methods 0.000 description 7
- 238000009826 distribution Methods 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 238000012417 linear regression Methods 0.000 description 5
- 238000009499 grossing Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000013459 approach Methods 0.000 description 3
- 238000007405 data analysis Methods 0.000 description 3
- 230000007423 decrease Effects 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 238000012216 screening Methods 0.000 description 3
- 230000009466 transformation Effects 0.000 description 3
- 230000002860 competitive effect Effects 0.000 description 2
- 238000003066 decision tree Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 238000000844 transformation Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000036961 partial effect Effects 0.000 description 1
- 238000013439 planning Methods 0.000 description 1
- 238000011158 quantitative evaluation Methods 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
- 238000002759 z-score normalization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/105—Human resources
- G06Q10/1053—Employment or hiring
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06398—Performance of employee with respect to a job function
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Software Systems (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Data Mining & Analysis (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Tourism & Hospitality (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Game Theory and Decision Science (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a talent assessment quantification method, a talent assessment quantification system, talent assessment equipment and a talent assessment medium, wherein the talent assessment quantification method comprises the following steps: information collection: collecting basic information according to information provided by job seekers, and carrying out weighted calculation on the basic information to obtain the score of an information part; and an evaluation step: inviting job seekers to participate in capability feature evaluation, learning capability evaluation and professional driving force evaluation, acquiring the completion condition and evaluation result, and comprehensively calculating the score of the evaluation part; the value separation obtaining step: and summarizing the scores of the information part and the scores of the evaluation part, calculating the value scores of the job seekers according to the adjustment of the dynamic weights, and presenting the value scores to the user. The invention can accurately evaluate and measure the true value and potential of talents, avoid the problem of subjective misjudgment and improve the efficiency and accuracy of talent management and recruitment.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a talent assessment quantification method, a talent assessment quantification system, talent assessment equipment and talent assessment media.
Background
In the current rapidly developing internet industry, recruitment of excellent talents becomes a key to the promotion of enterprise competitiveness. In the past, talent value assessment has been a very difficult problem because talent value has tended to be very subjective and complex. In this case, it is difficult to have an objective and standardized set of methods to evaluate talents. The reason for this has not been done before, mainly because of technical and data limitations. Past data management and analysis techniques have been relatively simple and straightforward, difficult to support for such complex data analysis and processing, and have placed high demands on the quality and integrity of talent data, which is difficult for many businesses and institutions.
But with the continued advancement of technology and data, the ability to process and analyze large-scale talent data has now been provided, which provides the possibility for the implementation and application of this scheme. In the current competitive employment market, a comprehensive and scientific assessment system is needed in order to more accurately assess the potential value of job seekers.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a talent assessment quantification method, a talent assessment quantification system, talent assessment equipment and talent assessment media.
According to the talent assessment quantification method, system, equipment and medium provided by the invention, the scheme is as follows:
in a first aspect, there is provided a talent assessment quantification method, the method comprising:
information collection: collecting basic information according to information provided by job seekers, and carrying out weighted calculation on the basic information to obtain the score of an information part;
and an evaluation step: inviting job seekers to participate in capability feature evaluation, learning capability evaluation and professional driving force evaluation, acquiring the completion condition and evaluation result, and comprehensively calculating the score of the evaluation part;
the value separation obtaining step: summarizing the scores of the information part and the scores of the evaluation part, calculating the value scores of the job seekers according to the adjustment of the dynamic weights, and presenting the value scores to the user;
The adjustment process of the dynamic weight is as follows:
initial weight allocation: when the system starts to evaluate, an initial weight is allocated to each evaluation index, and the initial weight is determined according to the importance of the index and the reliability of the data;
monitoring the data of which the evaluation is completed: when the recruiter completes one or more evaluations, the system monitors the collected data, so as to dynamically adjust the weight of each evaluation according to the actual situation;
weight adjustment based on machine learning: predicting the weight corresponding to each evaluation index through a training model by adopting a machine learning-based method; the training data comprises the data which is completely evaluated and the weights corresponding to the data, the model learns the weights corresponding to the indexes according to the data, and the weights are dynamically adjusted according to actual conditions;
dynamically adjusting weights: according to the weight and actual condition of model prediction, the system dynamically adjusts the weight of each evaluation index, and the rationality and accuracy of the value components are ensured;
generating final value points: after all the evaluation indexes are collected and weight adjusted, the system calculates the score of each evaluation index according to the weight after dynamic adjustment, and weights and sums the scores to obtain the final value score.
Preferably, in the information collecting step, collecting basic information includes: estimated compensation, school and working experience of job seekers;
wherein, the salary coefficient of the estimated salary is the level of the position dividing value of the estimated salary index in the position;
establishing a standard database of the split values of each job position, carrying out quarterly updating, wherein the investigation object of salary data is the data comparison in the limit range of cities and working years, the limited data is the index of 5 split values, dividing the whole function into a plurality of sections by adopting a piecewise linear interpolation method, and fitting a primary function in each section to ensure that the whole function continuously and monotonically increases in each section, and solving the split value of the estimated salary on the function, namely the salary coefficient;
and secondly, the academic coefficient and the working experience coefficient in the academic and working experiences are respectively set according to the duty ratio importance of the own required academic and working experience in the positions.
Preferably, in the value score acquisition step, in the case where both the information part and the evaluation part are completed, the weights each account for 50%, but if the evaluation part is not effectively collected, the weights of the information part are exclusively 100%; if the evaluation part only completes part of collection, the dynamic weight is adjusted according to specific numerical values.
Preferably, the weight adjustment based on machine learning includes:
data preprocessing: preprocessing the collected evaluation data for training a model, wherein the preprocessing process comprises data cleaning, data normalization and feature selection;
training a model: according to the data which have been evaluated and the weights corresponding to the data, the weights corresponding to the indexes are learned; in the training process, the model learns the weight corresponding to each index, dynamically adjusts the weight according to actual conditions, the training model needs to divide a data set into a training set and a test set, trains the model through the training set, and then evaluates the prediction capability of the model by using the test set;
model evaluation and tuning: the model evaluation is to evaluate the prediction capability and accuracy of the model, and the evaluation indexes comprise mean square error, average absolute error and decision coefficient; the evaluation result can guide the tuning of the model, including adjusting parameters of the model and selecting different characteristics or algorithms;
dynamic weight allocation: and predicting the data which is evaluated according to the trained model, so as to obtain the weight corresponding to each index.
Preferably, in the value score obtaining step, calculating a value score of the job seeker as an initial value score, and presenting the value score to a user to be processed after the value score is obtained, so as to eliminate deviation caused by single evaluation or information collection, wherein the processing method comprises a simple moving average method SMA and an exponential moving average method EMA;
The method comprises the steps of firstly storing original scores after initial value scores are generated, then storing processed final value score data, and calling the final value score data if an application scene is a relevant situation including a presentation report; if the application scene is the related situation including operation, the original score data of the initial value is called.
In a second aspect, there is provided a talent assessment quantification system, the system comprising:
an information collection module: collecting basic information according to information provided by job seekers, and carrying out weighted calculation on the basic information to obtain the score of an information part;
and an evaluation module: inviting job seekers to participate in capability feature evaluation, learning capability evaluation and professional driving force evaluation, acquiring the completion condition and evaluation result, and comprehensively calculating the score of the evaluation part;
the value score acquisition module: summarizing the scores of the information part and the scores of the evaluation part, calculating the value scores of the job seekers according to the adjustment of the dynamic weights, and presenting the value scores to the user;
the adjustment process of the dynamic weight is as follows:
initial weight allocation: when the system starts to evaluate, an initial weight is allocated to each evaluation index, and the initial weight is determined according to the importance of the index and the reliability of the data;
Monitoring the data of which the evaluation is completed: when the recruiter completes one or more evaluations, the system monitors the collected data, so as to dynamically adjust the weight of each evaluation according to the actual situation;
weight adjustment based on machine learning: predicting the weight corresponding to each evaluation index through a training model by adopting a machine learning-based method; the training data comprises the data which is completely evaluated and the weights corresponding to the data, the model learns the weights corresponding to the indexes according to the data, and the weights are dynamically adjusted according to actual conditions;
dynamically adjusting weights: according to the weight and actual condition of model prediction, the system dynamically adjusts the weight of each evaluation index, and the rationality and accuracy of the value components are ensured;
generating final value points: after all the evaluation indexes are collected and weight adjusted, the system calculates the score of each evaluation index according to the weight after dynamic adjustment, and weights and sums the scores to obtain the final value score.
Preferably, in the information collecting module, collecting basic information includes: estimated compensation, school and working experience of job seekers;
wherein, the salary coefficient of the estimated salary is the level of the position dividing value of the estimated salary index in the position;
Establishing a standard database of the split values of each job position, carrying out quarterly updating, wherein the investigation object of salary data is the data comparison in the limit range of cities and working years, the limited data is the index of 5 split values, dividing the whole function into a plurality of sections by adopting a piecewise linear interpolation method, and fitting a primary function in each section to ensure that the whole function continuously and monotonically increases in each section, and solving the split value of the estimated salary on the function, namely the salary coefficient;
secondly, the academic coefficient and the working experience coefficient in the academic and working experiences are respectively set according to the duty ratio importance of the own required academic and working experience in the positions;
in the value score acquisition module, under the condition that the information part and the evaluation part are completed, the weight accounts for 50 percent respectively, but if the evaluation part does not collect effectively, the weight of the information part is 100 percent exclusively; if the evaluation part only completes part collection, the dynamic weight is adjusted according to the specific numerical value;
in the value score acquisition module, calculating the value score of the job seeker as an initial value score, and presenting the value score to a user to be processed after the value score is obtained, so that deviation caused by single evaluation or information collection is eliminated, wherein the processing method comprises a simple moving average method SMA and an exponential moving average method EMA;
The method comprises the steps of firstly storing original scores after initial value scores are generated, then storing processed final value score data, and calling the final value score data if an application scene is a relevant situation including a presentation report; if the application scene is the related situation including operation, the original score data of the initial value is called.
Preferably, the weight adjustment based on machine learning includes:
data preprocessing: preprocessing the collected evaluation data for training a model, wherein the preprocessing process comprises data cleaning, data normalization and feature selection;
training a model: according to the data which have been evaluated and the weights corresponding to the data, the weights corresponding to the indexes are learned; in the training process, the model learns the weight corresponding to each index, dynamically adjusts the weight according to actual conditions, the training model needs to divide a data set into a training set and a test set, trains the model through the training set, and then evaluates the prediction capability of the model by using the test set;
model evaluation and tuning: the model evaluation is to evaluate the prediction capability and accuracy of the model, and the evaluation indexes comprise mean square error, average absolute error and decision coefficient; the evaluation result can guide the tuning of the model, including adjusting parameters of the model and selecting different characteristics or algorithms;
Dynamic weight allocation: and predicting the data which is evaluated according to the trained model, so as to obtain the weight corresponding to each index.
In a third aspect, there is provided a computer readable storage medium storing a computer program which, when executed by a processor, implements the steps of the talent assessment quantification method.
In a fourth aspect, an electronic device is provided, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the computer program implementing the steps of the talent assessment quantification method when executed by the processor.
Compared with the prior art, the invention has the following beneficial effects:
1. and (5) adaptively adjusting weights: according to the evaluation data completed by the recruiter, the system can intelligently adjust the weight of each evaluation so as to ensure the rationality and accuracy of the value components. This means that the system can generate meaningful value points even if the recruiter only completes a portion of the assessment;
2. the accuracy of the value score is improved: the dynamic weight distribution strategy enables the system to fully utilize available data, so that the accuracy of the value score is improved, and enterprises can effectively identify and select talents;
3. Enhancing the robustness of the system: the dynamic weight distribution strategy enables the system to have stronger robustness, can cope with the situation that different recruiters finish evaluation, and reduces the dependence on the data integrity. This makes the value score evaluation system more reliable and practical in practical applications.
Other advantages of the present invention will be set forth in the description of specific technical features and solutions, by which those skilled in the art should understand the advantages that the technical features and solutions bring.
Drawings
Other features, objects and advantages of the present invention will become more apparent upon reading of the detailed description of non-limiting embodiments, given with reference to the accompanying drawings in which:
FIG. 1 is a schematic flow chart of the method;
fig. 2 is a system frame diagram.
Detailed Description
The present invention will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the present invention, but are not intended to limit the invention in any way. It should be noted that variations and modifications could be made by those skilled in the art without departing from the inventive concept. These are all within the scope of the present invention.
The embodiment of the invention provides a talent assessment quantification method, which is shown by referring to fig. 1, and specifically comprises the following steps:
information collection: and collecting basic information according to the information provided by the job seeker, and carrying out weighted calculation on the basic information to obtain the score of the information part.
In the information collecting step, collecting basic information includes: estimated compensation, school and working experience of job seekers;
wherein, the salary coefficient of the estimated salary is the level of the position dividing value of the estimated salary index in the position;
establishing a standard database of the split values of each job position, carrying out quarterly updating, wherein the investigation object of salary data is the data comparison in the limit range of cities and working years, the limited data is the index of 5 split values, dividing the whole function into a plurality of sections by adopting a piecewise linear interpolation method, and fitting a primary function in each section to ensure that the whole function continuously and monotonically increases in each section, and solving the split value of the estimated salary on the function, namely the salary coefficient;
and secondly, the academic coefficient and the working experience coefficient in the academic and working experiences are respectively set according to the duty ratio importance of the own required academic and working experience in the positions.
And an evaluation step: inviting job seekers to participate in capability feature evaluation, learning capability evaluation and professional driving force evaluation, acquiring the completion condition and evaluation result, and comprehensively calculating the score of the evaluation part;
the value separation obtaining step: and summarizing the scores of the information part and the scores of the evaluation part, calculating the value scores of the job seekers according to the adjustment of the dynamic weights, and presenting the value scores to the user.
In the value division acquisition step, in the case where both the information part and the evaluation part are completed, the weights each account for 50%, but if the evaluation part does not perform effective collection, the weight of the information part is exclusive 100%; if the evaluation part only completes part of collection, the dynamic weight is adjusted according to specific numerical values.
The dynamic weight adjustment process is as follows:
initial weight allocation: when the system starts to evaluate, an initial weight is allocated to each evaluation index, and the initial weight is determined according to the importance of the index and the reliability of the data;
monitoring the data of which the evaluation is completed: when the recruiter completes one or more evaluations, the system monitors the collected data, so as to dynamically adjust the weight of each evaluation according to the actual situation;
weight adjustment based on machine learning: predicting the weight corresponding to each evaluation index through a training model by adopting a machine learning-based method; the training data comprises the data which is completely evaluated and the weights corresponding to the data, the model learns the weights corresponding to the indexes according to the data, and the weights are dynamically adjusted according to actual conditions;
Dynamically adjusting weights: according to the weight and actual condition of model prediction, the system dynamically adjusts the weight of each evaluation index, and the rationality and accuracy of the value components are ensured;
generating final value points: after all the evaluation indexes are collected and weight adjusted, the system calculates the score of each evaluation index according to the weight after dynamic adjustment, and weights and sums the scores to obtain the final value score.
Weight adjustment based on machine learning includes:
data preprocessing: preprocessing the collected evaluation data for training a model, wherein the preprocessing process comprises data cleaning, data normalization and feature selection;
training a model: according to the data which have been evaluated and the weights corresponding to the data, the weights corresponding to the indexes are learned; in the training process, the model learns the weight corresponding to each index, dynamically adjusts the weight according to actual conditions, the training model needs to divide a data set into a training set and a test set, trains the model through the training set, and then evaluates the prediction capability of the model by using the test set;
model evaluation and tuning: the model evaluation is to evaluate the prediction capability and accuracy of the model, and the evaluation indexes comprise mean square error, average absolute error and decision coefficient; the evaluation result can guide the tuning of the model, including adjusting parameters of the model and selecting different characteristics or algorithms;
Dynamic weight allocation: and predicting the data which is evaluated according to the trained model, so as to obtain the weight corresponding to each index.
In the value score acquisition step, calculating the value score of the job seeker as an initial value score, and presenting the value score to a user to be processed after the value score is acquired, so that deviation caused by single evaluation or information collection is eliminated, wherein the processing method comprises a simple moving average method SMA and an exponential moving average method EMA;
the method comprises the steps of firstly storing original scores after initial value scores are generated, then storing processed final value score data, and calling the final value score data if an application scene is a relevant situation including a presentation report; if the application scene is the related situation including operation, the original score data of the initial value is called.
The invention also provides a talent assessment quantification system which can be realized by executing the flow steps of the talent assessment quantification method, namely, the talent assessment quantification method can be understood as a preferred implementation mode of the talent assessment quantification system by a person skilled in the art. Referring to fig. 2, the system specifically includes the following:
An information collection module: and collecting basic information according to the information provided by the job seeker, and carrying out weighted calculation on the basic information to obtain the score of the information part.
In the information collection module, collecting basic information includes: estimated compensation, school and working experience of job seekers;
wherein, the salary coefficient of the estimated salary is the level of the position dividing value of the estimated salary index in the position;
establishing a standard database of the split values of each job position, carrying out quarterly updating, wherein the investigation object of salary data is the data comparison in the limit range of cities and working years, the limited data is the index of 5 split values, dividing the whole function into a plurality of sections by adopting a piecewise linear interpolation method, and fitting a primary function in each section to ensure that the whole function continuously and monotonically increases in each section, and solving the split value of the estimated salary on the function, namely the salary coefficient;
and secondly, the academic coefficient and the working experience coefficient in the academic and working experiences are respectively set according to the duty ratio importance of the own required academic and working experience in the positions.
And an evaluation module: inviting job seekers to participate in capability feature evaluation, learning capability evaluation and professional driving force evaluation, acquiring the completion condition and evaluation result, and comprehensively calculating the score of the evaluation part;
The value score acquisition module: and summarizing the scores of the information part and the scores of the evaluation part, calculating the value scores of the job seekers according to the adjustment of the dynamic weights, and presenting the value scores to the user.
In the value score acquisition module, under the condition that the information part and the evaluation part are completed, the weight accounts for 50% respectively, but if the evaluation part does not collect effectively, the weight of the information part is exclusive 100%; if the evaluation part only completes part of collection, the dynamic weight is adjusted according to specific numerical values.
The dynamic weight adjustment process is as follows:
initial weight allocation: when the system starts to evaluate, an initial weight is allocated to each evaluation index, and the initial weight is determined according to the importance of the index and the reliability of the data;
monitoring the data of which the evaluation is completed: when the recruiter completes one or more evaluations, the system monitors the collected data, so as to dynamically adjust the weight of each evaluation according to the actual situation;
weight adjustment based on machine learning: predicting the weight corresponding to each evaluation index through a training model by adopting a machine learning-based method; the training data comprises the data which is completely evaluated and the weights corresponding to the data, the model learns the weights corresponding to the indexes according to the data, and the weights are dynamically adjusted according to actual conditions;
Dynamically adjusting weights: according to the weight and actual condition of model prediction, the system dynamically adjusts the weight of each evaluation index, and the rationality and accuracy of the value components are ensured;
generating final value points: after all the evaluation indexes are collected and weight adjusted, the system calculates the score of each evaluation index according to the weight after dynamic adjustment, and weights and sums the scores to obtain the final value score.
Weight adjustment based on machine learning includes:
data preprocessing: preprocessing the collected evaluation data for training a model, wherein the preprocessing process comprises data cleaning, data normalization and feature selection;
training a model: according to the data which have been evaluated and the weights corresponding to the data, the weights corresponding to the indexes are learned; in the training process, the model learns the weight corresponding to each index, dynamically adjusts the weight according to actual conditions, the training model needs to divide a data set into a training set and a test set, trains the model through the training set, and then evaluates the prediction capability of the model by using the test set;
model evaluation and tuning: the model evaluation is to evaluate the prediction capability and accuracy of the model, and the evaluation indexes comprise mean square error, average absolute error and decision coefficient; the evaluation result can guide the tuning of the model, including adjusting parameters of the model and selecting different characteristics or algorithms;
Dynamic weight allocation: and predicting the data which is evaluated according to the trained model, so as to obtain the weight corresponding to each index.
In the value score acquisition module, calculating the value score of the job seeker as an initial value score, and presenting the value score to a user to be processed after the value score is obtained, so that deviation caused by single evaluation or information collection is eliminated, wherein the processing method comprises a simple moving average method SMA and an exponential moving average method EMA;
the method comprises the steps of firstly storing original scores after initial value scores are generated, then storing processed final value score data, and calling the final value score data if an application scene is a relevant situation including a presentation report; if the application scene is the related situation including operation, the original score data of the initial value is called.
Next, the present invention will be described in more detail.
In the current competitive employment market, a comprehensive and scientific assessment system is needed in order to more accurately assess the potential value of job seekers. The invention aims to develop a set of value score evaluation system according to the value score rule provided by the user, so as to help enterprises and recruiters to comprehensively evaluate job seekers, and provide effective basis for talent selection and human resource management. The talent assessment and quantification method provided by the invention comprises the following steps of, in combination with a specific example:
Information collection: basic information such as the academic, working experience and estimated salary of the job seeker is collected through information provided by the resume and the job seeker, and weighted calculation is carried out according to a value score rule to obtain the score of the information part.
Acquiring respective coefficients of the academic, the working experience and the estimated salary in the information collecting step:
salary coefficient: consider the level of the quantile value of the estimated salary index (calculated by another algorithm) at the position of the job (quantile value: refer to the data in the corresponding percentage position in the array by ordering the data from low to high.
(1) The standard of the dividing value of each position is a database established.
(2) Dynamic updates are made every quarter.
(3) Salary data investigation objects are data comparison within the limit range of cities and working years.
(4) The finite data is an index of 5 bit values, a piecewise linear interpolation method is adopted to divide the whole function into a plurality of intervals, a primary function is fitted in each interval, the whole function continuously and monotonically increases in each interval, and the bit values of the estimated salary on the function are solved, namely the salary coefficients.
The academic factors: (Adjustable)
(1) Doctor (1);
(2) Master (0.9);
(3) Family (0.85);
(4) Large specialty (0.75);
(5) High school (0.5);
(6) A department of Chinese/technical school (0.3);
(7) Junior middle school and below (0.1);
working experience coefficient: (Adjustable)
(1) 1 year or less (0.1)
(2) 2 years (0.2)
(3) 3 years (0.3)
(4) 4 years (0.4)
(5) 5 years (0.5)
(6) 6 years (0.6)
(7) 7 years (0.7)
(8) 8 years (0.8)
(9) 9 years (0.9)
(10) Over 10 years (1)
And an evaluation step: the job seeker is invited to participate in the ability characteristic evaluation (evaluation a), learning ability evaluation (evaluation B) and professional driving force evaluation (evaluation C), and the score of the evaluation portion is calculated according to the completion condition and the evaluation result and the value score rule.
The value separation obtaining step: and summarizing the scores of the information part and the evaluation part, calculating the total value score of the job seeker, and carrying out deep analysis by combining data such as historical values and the like, thereby providing the basis of talent selection and manpower resource management for enterprises and recruiters.
The weights are controlled by dynamic weight assignment, for example, in the case where both the information part and the evaluation part are filled in, the weights each account for 50%, but if the evaluation part is not collecting effectively, the weights of the information part can be exclusive by 100%. What the specific value is, depending on its accuracy, is dynamically adjusted, e.g. the big data found that the information part should not be too heavy, which may then gradually decrease from 50% specific gravity and vice versa.
After obtaining the total score of the valuation score, the total score is directly applied through processing and presented to the user for viewing as embodied in a report. The middle is indeed the process of processing. The main objective is to consider eliminating the deviation caused by single evaluation or information collection, and the specific means are not limited to a simple moving average method (Simple Moving Average, SMA is a common smoothing method, and data fluctuation and noise are reduced by calculating the average value of data in a given time window), an exponential moving average method (Exponential Moving Average, EMA), and the like to eliminate the deviation. Taking EMA as an example:
initializing the presentation value score to be equal to the hidden value score. For the second and subsequent hidden value points, an exponential moving average method is used to calculate the performance value points.
The specific calculation formula is as follows: representation value score (t) =α hidden value score (t) + (1- α) representation value score (t-1). Wherein t represents the current moment, alpha is a smoothing coefficient (the value range is between 0 and 1), and smaller alpha values can enable the expression value score to be smoother, and the influence of single deviation on the result is reduced. The exponential moving average method has the advantages that: for sudden increases or decreases, EMA may better eliminate these fluctuations, making the performance score more stable. EMA gives higher weight to the most recent data, meaning that it better reflects recent value component trends, while still maintaining some degree of smoothing.
The actual system selects a proper smoothing method according to the actual requirements and the data characteristics. And various schemes are combined to eliminate the deviation of single evaluation and improve the stability of the expression value. But this can be understood as a black box, a noise canceller, and the internal algorithm is dynamically optimized.
The flow after score generation is as follows:
1) After the score is generated, the original score is stored;
2) Processing data by the noise eliminator and then storing the processed data;
3) Directly calling the data of the 2) by using partial application scenes such as reports;
4) And (3) calling the data of 1) in part of application scenes such as operation, and finally generating some data to be displayed after the operation.
These presentation scenarios include, but are not limited to, value scores in value reports seen by the business inventory man-hours, valuation scores in resume images of candidates seen by the business recruitment, and the like.
The specific dynamic weight adjustment process is as follows:
(1) Initial weight allocation: the system assigns an initial weight to each assessment indicator at the beginning of the assessment, typically based on the importance of the indicator and the reliability of the data.
(2) Monitoring the data of which the evaluation is completed: when the recruiter completes one or more of the evaluations, the system will monitor the data collected to dynamically adjust the weight of each evaluation based on the actual situation.
(3) Weight adjustment based on machine learning: the system adopts a machine learning-based method, and predicts the weight corresponding to each evaluation index through a training model. The training data includes the data that has completed the evaluation and its corresponding weights. The model learns the weight corresponding to each index according to the data, and dynamically adjusts the weight according to the actual situation.
(4) Dynamically adjusting weights: according to the weight and actual condition of model prediction, the system can dynamically adjust the weight of each evaluation index so as to ensure the rationality and accuracy of the value components. For example, if the data for a certain metric is unreliable, the system may decrease the weight of that metric, thereby reducing its impact on the final score of value. Conversely, if the data of a certain index is very reliable, the system will increase the weight of that index, thereby increasing its impact on the final score of value.
(5) Generating final value points: after all the evaluation indexes are collected and weight adjusted, the system calculates the score of each evaluation index according to the weight after dynamic adjustment, and weights and sums the scores to obtain the final value score.
The procedure for (3) weight adjustment based on machine learning is developed as follows:
data preprocessing: the collected assessment data first needs to be preprocessed for training the model. The preprocessing process comprises the steps of data cleaning, data normalization, feature selection and the like. The data is cleaned to remove invalid data and abnormal values so as to ensure the quality of the data; normalization is to map the data of different indicators onto the same scale in order to compare their importance; feature selection is to select the most helpful feature for weight prediction.
Training a model: the training model aims at learning the weight corresponding to each index according to the data which has been evaluated and the weight corresponding to the data. This can be achieved by training a regression model, such as linear regression, ridge regression, support vector regression, etc. In the training process, the model learns the weights corresponding to the indexes, and dynamically adjusts the weights according to actual conditions. Training a model requires separating the data set into a training set and a test set, training the model through the training set, and then evaluating the predictive power of the model with the test set.
Model evaluation and tuning: model evaluation is to evaluate the predictive power and accuracy of the model. Common evaluation indexes include mean square error, average absolute error, decision coefficient, and the like. The evaluation result can guide the tuning of the model, including adjusting parameters of the model, selecting different features or algorithms, and the like.
Dynamic weight allocation: according to the trained model, the data which is evaluated can be predicted, so that the weight corresponding to each index is obtained. According to the actual situation, the weight can be dynamically adjusted to ensure the accuracy and rationality of the value score. The weights may be dynamically adjusted according to different criteria, such as according to reliability, importance, relevance, etc. of the data.
Examples of dynamic weighting:
assume that one recruiter needs to complete three assessment indicators: technical, communication and management capabilities, initial weights are 0.3, 0.3 and 0.4, respectively. When the recruiter completes the assessment of the technical ability and the communication ability, the system collects and records the relevant data, for example, the technical ability score is 80, the communication ability score is 70, and the predictive weights of the two indexes are calculated to be 0.35 and 0.25 respectively. The system can adjust the weight according to the predicted weight and the actual situation, and if the communication capacity score is lower but the predicted weight is higher, the system can reduce the weight of the index so as to reflect the importance of the index in the comprehensive evaluation of the recruiter.
Example of algorithm:
let us assume that we use a multiple linear regression model to predict the weights of the individual assessment indicators. For a certain evaluation index i, we define a multiple linear regression model:
y i ＝w 0 +w 1 x 1 +w 2 x 2 +...+w n * n
wherein y is i Weight, w, of index i 0 Is a constant term, w 1 To w n Weights, x, respectively representing other indices 1 To x n Respectively representing the scores of the other indicators that have been completed. We can train regression using the completed assessment data and its corresponding weightsAnd predicting the weight corresponding to each evaluation index according to the model.
Specifically, the step of training the regression model is as follows:
preparing training data: and taking the completed evaluation data and the corresponding weight thereof as training data. For example, for technical and communication capabilities, the following training data may be prepared:
training a regression model: the multiple linear regression model is trained using the training data. In particular, the parameters of the model may be solved using a least squares method such that the error between the predicted and actual values is minimized. The trained model can be used for predicting the weight corresponding to each evaluation index.
Prediction weights: when the recruiter completes a certain evaluation index, the system predicts the weight corresponding to each evaluation index by using the regression model obtained through training. Specifically, the system calculates the scores of other indexes according to the completed evaluation data, and then predicts the weights corresponding to the indexes by using a regression model.
Adjusting the weight according to the predicted weight and the actual situation: the system can adjust the weight according to the predicted weight and the actual situation. Specifically, if the predicted weight is consistent with the actual situation, the weight is kept unchanged; and if the predicted weight is inconsistent with the actual situation, corresponding adjustment is carried out.
Updating the adjusted weights: when the weight is adjusted, the system updates the adjusted weight and stores it in the database. These weights may be used in calculating the final value score.
Supplementing: in the training and predicting process of the regression model, data needs to be preprocessed and feature engineering so as to improve the accuracy and generalization capability of the model. Meanwhile, proper regression algorithm and super parameters are required to be selected, and parameters of the model are adjusted so as to achieve the optimal prediction effect.
Wherein, for the methods of pretreatment and feature engineering:
(1) Missing value processing: for missing values, the corresponding samples or features may be selected for deletion or filled in using interpolation.
(2) Data normalization: for data of different units and dimensions, a normalization process may be performed, such as Z-score normalization or min-max normalization.
(3) Feature selection: for excessive or redundant features, some of the most relevant or useful features may be selected for modeling to improve the accuracy and generalization ability of the model.
(4) Feature transformation: for some nonlinear relationships or outliers, feature transformations, such as polynomial expansions, logarithmic transformations, or discretizations, may be performed.
(5) Feature combination: feature combinations, such as feature interleaving or aggregation, may be performed for a plurality of related features.
For selecting proper regression algorithm and super parameters, and carrying out parameter adjustment on the model, the method comprises the following steps:
(1) A regression algorithm is selected: common regression algorithms are selected from linear regression, ridge regression, lasso regression, elastic network regression, decision tree regression, support vector machine regression and the like. When the regression algorithm is selected, factors such as complexity, generalization capability, stability, calculation efficiency and the like of the model are considered.
(2) Selecting super parameters: for certain regression algorithms, such as ridge regression and lasso regression, it is necessary to select a hyper-parameter, such as regularization strength. When the super parameters are selected, the performance of the model is evaluated by adopting methods such as cross validation and the like, and the proper super parameters are selected.
(3) Parameter adjustment and optimization: for decision tree regression and support vector machine regression, parameter adjustment optimization needs to be performed on the model, such as tree depth, node division mode, kernel function and the like. When parameter adjustment is optimized, the optimal super-parameter combination is found by adopting methods such as grid search, random search and the like.
In the current rapidly developing internet industry, recruitment of excellent talents becomes a key to the promotion of enterprise competitiveness. In this context, the value score evaluation system is applied to talent selection of an internet company.
Application cases: internet company a is recruiting a product manager. To better screen candidates, company a decides to employ our score evaluation system to evaluate the overall quality of the job seeker.
The implementation steps are as follows:
and (3) information collection: the recruiter submits the duration, and the system automatically acquires basic information such as the academic experience, the working experience and the like of the recruiter as an information part of the value score evaluation.
Self-evaluation: the recruiter needs to complete evaluation a (ability trait evaluation), evaluation B (learning ability evaluation), and evaluation C (professional driving force evaluation).
Calculating the value score: the system automatically calculates the value score according to the basic information of the recruiter and the self-evaluation result.
Talent screening and pulling out: company a can sort according to the corresponding recruiters of the value points, and select candidates with higher value points to enter the next round of interviews. To simplify the screening process, company a may even perform a manual screening with direct reference to the value score. Through the process, the company A can evaluate the comprehensive quality of the job seeker more objectively and comprehensively, and the talent selection accuracy is improved.
According to the embodiment, the value score evaluation system is applied to a specific recruitment scene, so that enterprises can be helped to more efficiently screen and select excellent talents, and the competitiveness of the enterprises is improved.
In the value score evaluation system, the following technical difficulties exist:
data integration and processing: the score evaluation system needs to integrate and process data from different sources, including basic information, self-evaluation results, etc. The technical difficulty is how to efficiently collect and process such data so that it can be compatible with and adapt to the needs of the evaluation algorithm. To solve this problem, we have devised a flexible data processing framework that automatically recognizes and integrates data from different sources and converts it into a format suitable for the evaluation algorithm.
Dynamic weight allocation: because the recruiter may only complete a portion of the evaluations, the system needs to dynamically adjust the weights of the evaluations according to the actual situation. The technical challenge here is to implement an intelligent weight distribution strategy to ensure the rationality and accuracy of the value scores. The weight distribution method based on machine learning is adopted, and the weight can be automatically adjusted according to the data which has been evaluated, so that the accuracy of the value score is improved.
Cross-module co-operation: the value score evaluation system involves a number of modules (e.g., information collection modules, self-assessment modules, etc.) that need to work together to generate the final value score. The technical difficulty is to realize efficient and stable cooperative work among the modules. To solve this problem, we have introduced a modular system design approach that enables each module to be developed and optimized independently while maintaining good synergy.
System expansibility: along with the continuous change of application scenes and demands, the value score evaluation system needs to have good expansibility so as to adapt to new data sources and evaluation indexes. The technical challenge is how to design an easily scalable system architecture. To achieve this goal, we have adopted a plug-in design concept that allows the system to easily add new modules and functions to meet changing demands.
By fusing the solutions in the technical scheme, the technical difficulties are successfully overcome, and a value score evaluation system which is efficient, accurate and easy to expand is realized.
In summary, the effects of the present invention are specifically expressed as follows:
recruitment efficiency is improved: the recruitment efficiency is improved by automatically collecting and processing the data of the recruiter, such as information of the recruiter, self-evaluation results and the like, and generating a comprehensive value score for the recruiter according to the data, so that enterprises can be helped to rapidly identify and screen talents with higher potential among a plurality of recruiters.
Optimizing talent selection: the system adopts a multidimensional evaluation index and a dynamic weight distribution strategy, so that the value score can more comprehensively and accurately reflect the capability and potential of the recruiter. This helps the enterprise to make more scientific and reasonable talent selection decisions.
According to the talent assessment quantification method, system, equipment and medium provided by the embodiment of the invention, a set of objective evaluation criteria is established by collecting, processing and analyzing a large amount of talent data based on data analysis and machine learning technology. These data are used to build a talent assessment model that can assess talents, such as skills, experience, competency, etc., according to different dimensions.
The scheme has the advantages that talent assessment can be more objective and accurate, the value of talents can be quantized, and recruitment and talent management are more efficient. This can also avoid evaluation errors and unfairness due to subjective factors.
The key meaning of the scheme is that the talents can be measured and evaluated more objectively and accurately through quantitative evaluation of talent data. Past recruitment and talent management processes are often based on subjective impressions and experience judgment, and this approach may lead to many bias and misjudgment such that truly valuable talents are ignored or underestimated, while talents that do not have practical value are overestimated or oversubscribed. By means of the scheme, the real value and potential of talents can be estimated and measured more accurately based on analysis of data and algorithms, the problem of subjective misjudgment is avoided, and the efficiency and accuracy of talent management and recruitment are improved.
The application field of this solution is very broad. Besides recruitment and talent management, the system can be applied to talent cultivation, professional development and other aspects. For example, enterprises may provide more personalized and accurate career development planning and training schemes for staff based on analysis of their data, thereby better exploiting the potential and value of the staff. In addition, government and social organizations may also utilize this approach to analyze and predict talent flows and distributions to better formulate talent policies and development plans. In a word, the scheme has very broad application prospect and market value.
Those skilled in the art will appreciate that the invention provides a system and its individual devices, modules, units, etc. that can be implemented entirely by logic programming of method steps, in addition to being implemented as pure computer readable program code, in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers, etc. Therefore, the system and various devices, modules and units thereof provided by the invention can be regarded as a hardware component, and the devices, modules and units for realizing various functions included in the system can also be regarded as structures in the hardware component; means, modules, and units for implementing the various functions may also be considered as either software modules for implementing the methods or structures within hardware components.
The foregoing describes specific embodiments of the present application. It is to be understood that the application is not limited to the particular embodiments described above, and that various changes or modifications may be made by those skilled in the art within the scope of the appended claims without affecting the spirit of the application. The embodiments of the application and the features of the embodiments may be combined with each other arbitrarily without conflict.
Claims (10)
1. A talent assessment quantification method, comprising:
information collection: collecting basic information according to information provided by job seekers, and carrying out weighted calculation on the basic information to obtain the score of an information part;
and an evaluation step: inviting job seekers to participate in capability feature evaluation, learning capability evaluation and professional driving force evaluation, acquiring the completion condition and evaluation result, and comprehensively calculating the score of the evaluation part;
the value separation obtaining step: summarizing the scores of the information part and the scores of the evaluation part, calculating the value scores of the job seekers according to the adjustment of the dynamic weights, and presenting the value scores to the user;
the adjustment process of the dynamic weight is as follows:
initial weight allocation: when the system starts to evaluate, an initial weight is allocated to each evaluation index, and the initial weight is determined according to the importance of the index and the reliability of the data;
Monitoring the data of which the evaluation is completed: when the recruiter completes one or more evaluations, the system monitors the collected data, so as to dynamically adjust the weight of each evaluation according to the actual situation;
weight adjustment based on machine learning: predicting the weight corresponding to each evaluation index through a training model by adopting a machine learning-based method; the training data comprises the data which is completely evaluated and the weights corresponding to the data, the model learns the weights corresponding to the indexes according to the data, and the weights are dynamically adjusted according to actual conditions;
dynamically adjusting weights: according to the weight and actual condition of model prediction, the system dynamically adjusts the weight of each evaluation index, and the rationality and accuracy of the value components are ensured;
generating final value points: after all the evaluation indexes are collected and weight adjusted, the system calculates the score of each evaluation index according to the weight after dynamic adjustment, and weights and sums the scores to obtain the final value score.
2. The talent assessment quantification method according to claim 1, wherein in the information collection step, collecting basic information includes: estimated compensation, school and working experience of job seekers;
wherein, the salary coefficient of the estimated salary is the level of the position dividing value of the estimated salary index in the position;
Establishing a standard database of the split values of each job position, carrying out quarterly updating, wherein the investigation object of salary data is the data comparison in the limit range of cities and working years, the limited data is the index of 5 split values, dividing the whole function into a plurality of sections by adopting a piecewise linear interpolation method, and fitting a primary function in each section to ensure that the whole function continuously and monotonically increases in each section, and solving the split value of the estimated salary on the function, namely the salary coefficient;
and secondly, the academic coefficient and the working experience coefficient in the academic and working experiences are respectively set according to the duty ratio importance of the own required academic and working experience in the positions.
3. The talent assessment quantification method according to claim 1, wherein in the value division acquisition step, in the case where both the information section and the assessment section are completed, weights each account for 50%, but if the assessment section does not perform effective collection, the weight of the information section is exclusive 100%; if the evaluation part only completes part of collection, the dynamic weight is adjusted according to specific numerical values.
4. The talent assessment quantification method of claim 1, wherein the machine learning based weight adjustment comprises:
Data preprocessing: preprocessing the collected evaluation data for training a model, wherein the preprocessing process comprises data cleaning, data normalization and feature selection;
training a model: according to the data which have been evaluated and the weights corresponding to the data, the weights corresponding to the indexes are learned; in the training process, the model learns the weight corresponding to each index, dynamically adjusts the weight according to actual conditions, the training model needs to divide a data set into a training set and a test set, trains the model through the training set, and then evaluates the prediction capability of the model by using the test set;
model evaluation and tuning: the model evaluation is to evaluate the prediction capability and accuracy of the model, and the evaluation indexes comprise mean square error, average absolute error and decision coefficient; the evaluation result can guide the tuning of the model, including adjusting parameters of the model and selecting different characteristics or algorithms;
dynamic weight allocation: and predicting the data which is evaluated according to the trained model, so as to obtain the weight corresponding to each index.
5. The talent assessment quantification method according to claim 1, wherein in the value score obtaining step, a value score of a job seeker is calculated as an initial value score, and the value score is presented to a user to be processed after the value score is obtained, so that deviation caused by single assessment or information collection is eliminated, and the processing method comprises a simple moving average method SMA and an exponential moving average method EMA;
The method comprises the steps of firstly storing original scores after initial value scores are generated, then storing processed final value score data, and calling the final value score data if an application scene is a relevant situation including a presentation report; if the application scene is the related situation including operation, the original score data of the initial value is called.
6. A talent assessment quantification system, comprising:
an information collection module: collecting basic information according to information provided by job seekers, and carrying out weighted calculation on the basic information to obtain the score of an information part;
and an evaluation module: inviting job seekers to participate in capability feature evaluation, learning capability evaluation and professional driving force evaluation, acquiring the completion condition and evaluation result, and comprehensively calculating the score of the evaluation part;
the value score acquisition module: summarizing the scores of the information part and the scores of the evaluation part, calculating the value scores of the job seekers according to the adjustment of the dynamic weights, and presenting the value scores to the user;
the adjustment process of the dynamic weight is as follows:
initial weight allocation: when the system starts to evaluate, an initial weight is allocated to each evaluation index, and the initial weight is determined according to the importance of the index and the reliability of the data;
Monitoring the data of which the evaluation is completed: when the recruiter completes one or more evaluations, the system monitors the collected data, so as to dynamically adjust the weight of each evaluation according to the actual situation;
weight adjustment based on machine learning: predicting the weight corresponding to each evaluation index through a training model by adopting a machine learning-based method; the training data comprises the data which is completely evaluated and the weights corresponding to the data, the model learns the weights corresponding to the indexes according to the data, and the weights are dynamically adjusted according to actual conditions;
dynamically adjusting weights: according to the weight and actual condition of model prediction, the system dynamically adjusts the weight of each evaluation index, and the rationality and accuracy of the value components are ensured;
generating final value points: after all the evaluation indexes are collected and weight adjusted, the system calculates the score of each evaluation index according to the weight after dynamic adjustment, and weights and sums the scores to obtain the final value score.
7. The talent assessment quantification system of claim 6, wherein the information gathering module gathers basic information comprising: estimated compensation, school and working experience of job seekers;
wherein, the salary coefficient of the estimated salary is the level of the position dividing value of the estimated salary index in the position;
Establishing a standard database of the split values of each job position, carrying out quarterly updating, wherein the investigation object of salary data is the data comparison in the limit range of cities and working years, the limited data is the index of 5 split values, dividing the whole function into a plurality of sections by adopting a piecewise linear interpolation method, and fitting a primary function in each section to ensure that the whole function continuously and monotonically increases in each section, and solving the split value of the estimated salary on the function, namely the salary coefficient;
secondly, the academic coefficient and the working experience coefficient in the academic and working experiences are respectively set according to the duty ratio importance of the own required academic and working experience in the positions;
in the value score acquisition module, under the condition that the information part and the evaluation part are completed, the weight accounts for 50 percent respectively, but if the evaluation part does not collect effectively, the weight of the information part is 100 percent exclusively; if the evaluation part only completes part collection, the dynamic weight is adjusted according to the specific numerical value;
in the value score acquisition module, calculating the value score of the job seeker as an initial value score, and presenting the value score to a user to be processed after the value score is obtained, so that deviation caused by single evaluation or information collection is eliminated, wherein the processing method comprises a simple moving average method SMA and an exponential moving average method EMA;
The method comprises the steps of firstly storing original scores after initial value scores are generated, then storing processed final value score data, and calling the final value score data if an application scene is a relevant situation including a presentation report; if the application scene is the related situation including operation, the original score data of the initial value is called.
8. The talent assessment quantification system of claim 6, wherein the machine learning based weight adjustment comprises:
data preprocessing: preprocessing the collected evaluation data for training a model, wherein the preprocessing process comprises data cleaning, data normalization and feature selection;
training a model: according to the data which have been evaluated and the weights corresponding to the data, the weights corresponding to the indexes are learned; in the training process, the model learns the weight corresponding to each index, dynamically adjusts the weight according to actual conditions, the training model needs to divide a data set into a training set and a test set, trains the model through the training set, and then evaluates the prediction capability of the model by using the test set;
model evaluation and tuning: the model evaluation is to evaluate the prediction capability and accuracy of the model, and the evaluation indexes comprise mean square error, average absolute error and decision coefficient; the evaluation result can guide the tuning of the model, including adjusting parameters of the model and selecting different characteristics or algorithms;
Dynamic weight allocation: and predicting the data which is evaluated according to the trained model, so as to obtain the weight corresponding to each index.
9. A computer-readable storage medium storing a computer program, characterized in that the computer program, when executed by a processor, implements the steps of the talent assessment quantification method of any of claims 1 to 5.
10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the computer program when executed by the processor implements the steps of the talent assessment quantification method of any of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310882800.XA CN117035710A (en) | 2023-07-18 | 2023-07-18 | Talent assessment quantification method, system, equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310882800.XA CN117035710A (en) | 2023-07-18 | 2023-07-18 | Talent assessment quantification method, system, equipment and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117035710A true CN117035710A (en) | 2023-11-10 |
Family
ID=88623461
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310882800.XA Pending CN117035710A (en) | 2023-07-18 | 2023-07-18 | Talent assessment quantification method, system, equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117035710A (en) |
-
2023
- 2023-07-18 CN CN202310882800.XA patent/CN117035710A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110110981B (en) | Credit rating default probability measure and risk early warning method | |
| US20180053092A1 (en) | Method and System for Innovation Management and Optimization Under Uncertainty | |
| US8200526B2 (en) | Method and system for collecting stakeholder relationship data | |
| Liu et al. | A competency evaluation method of human resources managers based on multi-granularity linguistic variables and VIKOR method | |
| CN111950708B (en) | Neural network structure and method for finding daily life habits of college students | |
| AU2019100968A4 (en) | A Credit Reporting Evaluation System Based on Mixed Machine Learning | |
| CN110782375B (en) | Online learning overall process dynamic analysis method and system based on data | |
| WO2017071369A1 (en) | Method and device for predicting user unsubscription | |
| DE112020002684T5 (en) | A multi-process system for optimal predictive model selection | |
| CN116611546B (en) | Knowledge-graph-based landslide prediction method and system for target research area | |
| CN111476274A (en) | Big data prediction analysis method, system, device and storage medium | |
| CN115249081A (en) | Object type prediction method and device, computer equipment and storage medium | |
| Shnorr | Integral assessment of retail digitalization | |
| Kosylo et al. | Artificial intelligence on job-hopping forecasting: AI on job-hopping | |
| Yet et al. | Estimating criteria weight distributions in multiple criteria decision making: a Bayesian approach | |
| CN110738565A (en) | Real estate finance artificial intelligence composite wind control model based on data set | |
| Fouladgar et al. | Project portfolio selection using VIKOR technique under Fuzzy environment | |
| CN117035710A (en) | Talent assessment quantification method, system, equipment and medium | |
| CN114841497A (en) | Large-cluster satellite emergency scheme decision method and system considering non-cooperative behaviors | |
| Li et al. | University Students' behavior characteristics analysis and prediction method based on combined data mining model | |
| CN113159634A (en) | Financial product management method and device and electronic equipment | |
| Hrinchenko et al. | Qualimetric approaches to assessing sustainable development indicators | |
| Patayon et al. | Time Series Analysis on Enrolment Data: A case in a State University in Zamboanga del Norte, Philippines | |
| Shen et al. | Enhance the evaluation quality of project performance based on fuzzy aggregation weight effect | |
| WO2022227213A1 (en) | Industry recommendation method and apparatus, computer device and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |