Employee intention post prediction method and device
By building and optimizing the employee intended job prediction model, the problem of low accuracy in employee intended job prediction under the traditional job matching model is solved, more efficient employee job matching is achieved, and the risk of talent loss is reduced.
Patent Information
- Application Number
- CN202511018376.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-23
- Publication Date
- 2025-10-10
AI Technical Summary
The traditional job matching model cannot meet the career growth needs of employees, resulting in mismatch between people and jobs, affecting business operations and talent loss. In addition, traditional data analysis methods cannot comprehensively analyze the characteristic attributes of employee information, resulting in low accuracy in predicting intended jobs.
By obtaining employee information samples, a prediction model for employee intended positions is constructed. The optimization algorithm is used to adjust the model parameters and fill in missing data. Iterative training is performed to generate a trained prediction model, and finally the target employee's intended position is predicted.
It improves the accuracy and adaptability of employee intended job prediction, enhances the model's ability to process information of different employees, and avoids the impact of missing data on prediction results.
Smart Images
Figure CN120764787A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of big data technology, and in particular to a method and device for predicting employees' intended positions. Background Art
[0002] With the dynamic evolution of corporate organizational structures and job requirements, as well as the diversification of employee career development aspirations, traditional job matching models are becoming increasingly inadequate to meet employees' career growth needs. This mismatch can significantly impact the normal operation and development of a company. On the one hand, employees are unable to leverage their expertise in appropriate roles, leading to low productivity. On the other hand, increased talent turnover increases recruitment and training costs, and can even lead to the leakage of trade secrets. Therefore, improving the efficiency of job matching is crucial. Traditional job matching relies primarily on manual screening, keyword matching on resumes, and questionnaires. These methods are not only inefficient but also fail to fully tap into employees' potential and hidden job requirements. To improve job matching efficiency and prevent significant talent loss, companies must continuously optimize their human resource management methods. However, due to the complexity of employee information attributes and the pain point of missing data, traditional data analysis methods cannot fully analyze these attributes, resulting in low accuracy in predicting employees' preferred positions.
[0003] In summary, how to improve the accuracy of employee intended job prediction is a technical problem that technical personnel in this field urgently need to solve. Summary of the Invention
[0004] In view of this, the present application provides a method and device for predicting employee intended positions, aiming to improve the accuracy of employee intended position prediction.
[0005] In a first aspect, the present application provides a method for predicting employees' intended positions, comprising:
[0006] Obtaining sample employee information;
[0007] Inputting the employee information sample into a pre-built employee intended position prediction model;
[0008] Adjusting the model parameters of the employee intended position prediction model through an optimization algorithm, and filling in the missing data in the employee information sample to obtain an updated employee information sample;
[0009] Iteratively training the employee intended position prediction model using the updated employee information sample to obtain a trained employee intended position prediction model;
[0010] The trained employee intended position prediction model is used to predict the intended position of the target employee.
[0011] Optionally, before inputting the employee information sample into a pre-built employee intended position prediction model, the method further includes:
[0012] Preprocessing the employee information sample;
[0013] The preprocessing of the employee information sample includes:
[0014] Performing data cleaning on the employee information sample to obtain a cleaned employee information sample;
[0015] performing numerical processing on the string data in the cleaned employee information sample, performing binarization processing on the discrete data in the cleaned employee information sample, and performing normalization processing on the continuous data in the cleaned employee information sample to obtain a processed employee information sample;
[0016] extracting target features from the processed employee information sample to construct a first feature vector;
[0017] The job transfer tendency is extracted from the processed employee information sample to construct a first target vector.
[0018] Optionally, extracting target features from the processed employee information sample to construct a first feature vector includes:
[0019] Pre-filling missing values in the processed employee information sample;
[0020] Target features are extracted from the pre-filled employee information sample to construct the first feature vector.
[0021] Optionally, before inputting the employee information sample into a pre-built employee intended position prediction model, the method further includes:
[0022] Splitting the employee information sample into a training set and a validation set;
[0023] Inputting the employee information sample into a pre-built employee intended position prediction model includes:
[0024] The training set is input into a pre-built employee intended position prediction model.
[0025] Optionally, after iteratively training the employee intended position prediction model using the updated employee information sample to obtain the trained employee intended position prediction model, the method further includes:
[0026] The trained employee intended job prediction model is verified using the verification set.
[0027] Optionally, using the trained employee intended position prediction model to predict the intended position of a target employee includes:
[0028] Obtain employee information of target employees;
[0029] The employee information of the target employee is input into the trained employee intended position prediction model to obtain a prediction result of the target employee's intended position.
[0030] Optionally, before inputting the employee information of the target employee into the trained employee intended position prediction model to obtain the target employee's intended position prediction result, the method further includes:
[0031] Preprocessing the employee information of the target employee;
[0032] The pre-processing of the employee information of the target employee includes:
[0033] Performing data cleansing on the employee information of the target employee to obtain cleansed employee information;
[0034] digitizing the string data in the cleaned employee information, binarizing the discrete data in the cleaned employee information, and normalizing the continuous data in the cleaned employee information to obtain processed employee information;
[0035] extracting target features from the processed employee information to construct a second feature vector;
[0036] The job transfer tendency is extracted from the processed employee information to construct a second target vector.
[0037] Optionally, extracting target features from the processed employee information to construct a feature vector includes:
[0038] Pre-filling missing values in the processed employee information;
[0039] Target features are extracted from the pre-filled employee information to construct the second feature vector.
[0040] In a second aspect, the present application provides an employee intended job prediction device, comprising:
[0041] Acquisition module, used to obtain employee information samples;
[0042] An input module, configured to input the employee information sample into a pre-built employee intended position prediction model;
[0043] A filling module is used to adjust the model parameters of the employee intended position prediction model through an optimization algorithm and fill in the missing data in the employee information sample to obtain an updated employee information sample;
[0044] A training module, configured to iteratively train the employee intended position prediction model using the updated employee information samples to obtain a trained employee intended position prediction model;
[0045] The prediction module is used to use the trained employee intended position prediction model to predict the intended position of the target employee.
[0046] Optionally, the device further comprises:
[0047] A first preprocessing module, configured to preprocess the employee information sample;
[0048] The first preprocessing module is specifically used for:
[0049] Performing data cleaning on the employee information sample to obtain a cleaned employee information sample;
[0050] performing numerical processing on the string data in the cleaned employee information sample, performing binarization processing on the discrete data in the cleaned employee information sample, and performing normalization processing on the continuous data in the cleaned employee information sample to obtain a processed employee information sample;
[0051] extracting target features from the processed employee information sample to construct a first feature vector;
[0052] The job transfer tendency is extracted from the processed employee information sample to construct a first target vector.
[0053] Optionally, extracting target features from the processed employee information sample to construct a first feature vector includes:
[0054] Pre-filling missing values in the processed employee information sample;
[0055] Target features are extracted from the pre-filled employee information sample to construct the first feature vector.
[0056] Optionally, the device further comprises:
[0057] A splitting module, used to split the employee information sample into a training set and a validation set;
[0058] The input module is specifically used for:
[0059] The training set is input into a pre-built employee intended position prediction model.
[0060] Optionally, the device further comprises:
[0061] A verification module is used to verify the trained employee intended job prediction model using the verification set.
[0062] Optionally, the prediction module includes:
[0063] An acquisition unit, used to acquire employee information of a target employee;
[0064] The prediction unit is used to input the employee information of the target employee into the trained employee intended position prediction model to obtain the intended position prediction result of the target employee.
[0065] Optionally, the device further comprises:
[0066] A second preprocessing module, configured to preprocess the employee information of the target employee;
[0067] The second preprocessing module is specifically used for:
[0068] Performing data cleansing on the employee information of the target employee to obtain cleansed employee information;
[0069] digitizing the string data in the cleaned employee information, binarizing the discrete data in the cleaned employee information, and normalizing the continuous data in the cleaned employee information to obtain processed employee information;
[0070] extracting target features from the processed employee information to construct a second feature vector;
[0071] The job transfer tendency is extracted from the processed employee information to construct a second target vector.
[0072] Optionally, extracting target features from the processed employee information to construct a feature vector includes:
[0073] Pre-filling missing values in the processed employee information;
[0074] Target features are extracted from the pre-filled employee information to construct the second feature vector.
[0075] This application provides a method for predicting employee intended positions. When executing the method, employee information samples are first obtained. The employee information samples are then input into a pre-built employee intended position prediction model. Then, the model parameters of the employee intended position prediction model are adjusted using an optimization algorithm, and missing data in the employee information sample is filled in to obtain an updated employee information sample. The updated employee information sample is then used to iteratively train the employee intended position prediction model, obtaining a trained employee intended position prediction model. Finally, the trained employee intended position prediction model is used to predict the intended position of a target employee. In this manner, employee information samples are first obtained to provide basic data for subsequent analysis; the samples are then input into the pre-built model, establishing an initial framework for prediction. Adjusting model parameters through the optimization algorithm improves the model's accuracy and adaptability, making it more relevant to actual conditions; filling in missing data ensures data integrity and quality, preventing the impact of missing data on prediction results. Iteratively training the model using the updated samples continuously optimizes model performance and enhances its ability to process and predict different employee information. Finally, the trained model is used to predict the target employee's intended position, thereby improving the accuracy of employee intended position predictions. BRIEF DESCRIPTION OF THE DRAWINGS
[0076] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.
[0077] Figure 1 A flowchart of a method for predicting employee intended positions provided in an embodiment of the present application;
[0078] Figure 2 A schematic diagram of the structure of an employee intended position prediction device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0079] The following, in conjunction with the accompanying drawings, provides a clear and complete description of the technical solutions in the embodiments of this application. This application provides a method and device for predicting employee intended positions, which are used in the field of big data technology. The above is merely an example and does not limit the application areas of the method and device provided in this application.
[0080] With the dynamic evolution of corporate organizational structures and job requirements, as well as the diversification of employee career development aspirations, traditional job matching models are becoming increasingly inadequate to meet employees' career growth needs. This mismatch can significantly impact the normal operation and development of a company. On the one hand, employees are unable to leverage their expertise in appropriate roles, leading to low productivity. On the other hand, increased talent turnover increases recruitment and training costs, and can even lead to the leakage of trade secrets. Therefore, improving the efficiency of job matching is crucial. Traditional job matching relies primarily on manual screening, keyword matching on resumes, and questionnaires. These methods are not only inefficient but also fail to fully tap into employees' potential and hidden job requirements. To improve job matching efficiency and prevent significant talent loss, companies must continuously optimize their human resource management methods. However, due to the complexity of employee information attributes and the pain point of missing data, traditional data analysis methods cannot fully analyze these attributes, resulting in low accuracy in predicting employees' preferred positions.
[0081] After research, the inventors proposed the technical solution of this application. This solution first obtains employee information samples and then inputs them into a pre-built employee intended position prediction model. Then, an optimization algorithm is used to adjust the model parameters of the employee intended position prediction model and fill in missing data in the employee information sample, resulting in an updated employee information sample. The updated employee information sample is then used to iteratively train the employee intended position prediction model, resulting in a trained employee intended position prediction model. Finally, the trained employee intended position prediction model is used to predict the intended position of a target employee. In this manner, employee information samples are first obtained to provide basic data for subsequent analysis; inputting the samples into the pre-built model establishes an initial framework for prediction. Adjusting model parameters through the optimization algorithm improves the model's accuracy and adaptability, making it more relevant to actual conditions; filling in missing data ensures data integrity and quality, preventing the impact of missing data on prediction results. Iteratively training the model using the updated samples continuously optimizes model performance and enhances its ability to process and predict diverse employee information. Finally, the trained model is used to predict the target employee's intended position, thereby improving the accuracy of employee intended position predictions.
[0082] In order to enable those skilled in the art to better understand the present application, the present application is further described in detail below in conjunction with the accompanying drawings and specific embodiments. Obviously, the embodiments described are only a part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without making creative work are within the scope of protection of the present application. It should be noted that, for ease of description, only the parts related to the relevant invention are shown in the accompanying drawings. In the absence of conflict, the embodiments in the present application and the features in the embodiments can be combined with each other.
[0083] First, some terminology needs to be explained:
[0084] Missing values: The values of one or more attributes in the existing dataset are empty.
[0085] Data imputation: short for imputation. Data imputation can be understood as the process of estimating missing values in a dataset by using values calculated using appropriate methods. In other words, data imputation can form a complete dataset by filling in the missing values in an incomplete dataset without response errors.
[0086] Binarization: Convert numerical feature data into binary data, that is, feature data with a value greater than the set threshold is set to 1, otherwise it is 0.
[0087] Normalization: Normalization can eliminate the influence of dimension (that is, the size of the value is related to the unit of measurement) and limit the distribution of the data to a given minimum and maximum range after being processed by a specific algorithm.
[0088] Feature vector: A sample is composed of different attributes, and different attributes are represented by different attribute values. Multiple attribute values combined together can be represented by a vector, which is called a feature vector.
[0089] Artificial neural network: abbreviated as neural network, is a network composed of interconnected nonlinear processing units, which is used to perform distributed parallel processing of input values to obtain a set of output values.
[0090] Autoassociative neural network: It is a fully connected neural network that copies input to output, that is, the attributes of the input sample are consistent with the attributes of the output sample.
[0091] Multi-task learning: Use shared learning to solve a main problem and multiple other related secondary problems simultaneously.
[0092] Generalization error: It is used to measure the ability of a machine learning model to generalize unknown data, that is, the error in predicting unknown data based on the rules learned from sample data.
[0093] See also Figure 1 , Figure 1 A flowchart of a method for predicting employee intended positions provided in an embodiment of the present application includes:
[0094] S101: Obtain employee information sample.
[0095] Collect employee information samples, including both current and former employees. The sample data includes qualitative and quantitative data, and its attributes include basic information (employee department, age, education, marital status, salary, position and rank, job changes, etc.), intended positions (professional, skilled, management, etc.), and job transfer tendencies (the transfer tendency rate of former employees is the highest by default). It should be noted here that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions.
[0096] The collected employee information samples were preprocessed as follows: data cleaning was performed to filter out invalid information, such as employee home addresses. String data in the samples was converted to numeric values, discrete data was binarized, and continuous data was normalized to the range [0, 1]. Feature vectors and target vectors were then extracted from the employee information sample set. Usable features were extracted from the processed sample set and a feature vector was constructed, with the target vector representing job transfer propensity. The collection of employee information samples presented challenges in collecting some information items about current employees. Since neural network models cannot handle missing values when generating feature vectors, simple imputation methods such as the K-nearest neighbor algorithm were used to pre-fill missing values.
[0097] S102: Inputting employee information samples into a pre-built employee intended position prediction model.
[0098] First, we need to build a prediction model for employees' intended positions. Based on employee information, we use a random data partitioning method to split the employee information sample set into a training set and a test set in a ratio of 8:2 to build an employee intended position prediction model, thereby obtaining the transfer tendency and intended positions of employees.
[0099] The employee intended job prediction model is an improved multi-task learning architecture based on an autoassociative neural network that performs both prediction and filling tasks. This architecture predicts employee transfer intentions and fills employee intended positions. The calculation rules for the output layer nodes of the autoassociative neural network are as follows:
[0100] ;
[0101] Among them, d represents the number of attributes of the employee information sample, Represents the employee information input sample of the model, Represents the output vector of the model.
[0102] The transmission path of employee information feature vectors from the model input layer to the output layer is reconstructed to eliminate the direct mapping of input attributes to output attributes, thereby enhancing the mutual correlation between employee sample attributes. The calculation rules for the output layer nodes in the model filling and prediction tasks are as follows:
[0103] ;
[0104] ;
[0105] Among them, q represents the number of hidden layer nodes; represents the threshold of the hth hidden layer node; represents the threshold of the j-th output layer node; Represents the connection weight between the kth node in the input layer and the hth node in the hidden layer; Represents the connection weight between the hth node in the hidden layer and the jth node in the output layer; represents a nonlinear activation function, and c represents the number of job transfer tendency classifications of employee information samples.
[0106] This model leverages multi-task learning to balance the prediction of employee transfer propensity with the filling of missing data (e.g., intended positions) within employee samples. Filling in missing data is considered a secondary task, while predicting employee transfer propensity is the primary task. This model not only improves the utilization of known employee information but also reduces the impact of errors in estimating employees' intended positions on model parameter optimization, thereby improving the accuracy of employee transfer propensity prediction.
[0107] The entire employee information sample set is input into the employee intended position prediction model, and the cost function is calculated based on the model output.
[0108] S103: Adjust the model parameters of the employee intended position prediction model through the optimization algorithm, and fill in the missing data in the employee information sample to obtain an updated employee information sample.
[0109] The model parameters are adjusted through the optimization algorithm, and the missing data in the employee information sample (intended positions, etc.) are estimated. The missing data in the employee information sample set is updated, and the updated employee information sample set is used as the input of the model to complete the next iteration.
[0110] S104: Iteratively train the employee intended job prediction model using the updated employee information samples to obtain a trained employee intended job prediction model.
[0111] When the iteration stopping condition is reached, the model training is terminated, the missing employee information data is filled in, and the employee transfer tendency rate is output.
[0112] Finally, the validation group is used to verify the trained model.
[0113] S105: Using the trained employee intended job prediction model, predict the intended job of the target employee.
[0114] Obtain the target employee's employee information and input the target employee's employee information into the trained employee intended position prediction model to obtain a prediction result for the target employee's intended position. In some embodiments, the target employee's employee information can be preprocessed in the same manner as the preprocessing of the employee information sample, which will not be further described here.
[0115] In an embodiment of the present application, an employee information sample is first obtained, and then the employee information sample is input into a pre-built employee intended position prediction model. Then, the model parameters of the employee intended position prediction model are adjusted through an optimization algorithm, and the missing data in the employee information sample is filled to obtain an updated employee information sample. Then, the employee intended position prediction model is iteratively trained using the updated employee information sample to obtain a trained employee intended position prediction model. Finally, the trained employee intended position prediction model is used to predict the intended position of the target employee. In this way, the employee information sample is first obtained to provide basic data for subsequent analysis; the sample is input into the pre-built model to build an initial framework for prediction. Adjusting the model parameters through the optimization algorithm can improve the accuracy and adaptability of the model, making it more in line with the actual situation; filling in the missing data ensures the integrity and quality of the data, avoiding the impact of missing data on the prediction results. Using the updated sample to iteratively train the model can continuously optimize the model performance and enhance its ability to process and predict different employee information. Finally, the trained model is used to predict the target employee's intended position, which can improve the accuracy of the employee intended position prediction.
[0116] The above are some specific implementations of the employee intended job prediction method provided in the embodiment of the present application. Based on this, the present application also provides a corresponding device. The device provided in the embodiment of the present application will be introduced from the perspective of functional modularization.
[0117] See also Figure 2 , Figure 2 This is a schematic diagram of the structure of a device for predicting employee intended positions provided in an embodiment of the present application. The device 200 for predicting employee intended positions includes:
[0118] An acquisition module 210 is used to acquire employee information samples;
[0119] An input module 220 is used to input the employee information sample into a pre-built employee intended position prediction model;
[0120] A filling module 230 is used to adjust the model parameters of the employee intended position prediction model through an optimization algorithm and fill in the missing data in the employee information sample to obtain an updated employee information sample;
[0121] A training module 240 is configured to iteratively train the employee intended position prediction model using the updated employee information samples to obtain a trained employee intended position prediction model;
[0122] The prediction module 250 is used to predict the target employee's intended position using the trained employee intended position prediction model.
[0123] Optionally, the apparatus 200 further includes:
[0124] A first preprocessing module, configured to preprocess the employee information sample;
[0125] The first preprocessing module is specifically used for:
[0126] Performing data cleaning on the employee information sample to obtain a cleaned employee information sample;
[0127] performing numerical processing on the string data in the cleaned employee information sample, performing binarization processing on the discrete data in the cleaned employee information sample, and performing normalization processing on the continuous data in the cleaned employee information sample to obtain a processed employee information sample;
[0128] extracting target features from the processed employee information sample to construct a first feature vector;
[0129] The job transfer tendency is extracted from the processed employee information sample to construct a first target vector.
[0130] Optionally, extracting target features from the processed employee information sample to construct a first feature vector includes:
[0131] Pre-filling missing values in the processed employee information sample;
[0132] Target features are extracted from the pre-filled employee information sample to construct the first feature vector.
[0133] Optionally, the apparatus 200 further includes:
[0134] A splitting module, used to split the employee information sample into a training set and a validation set;
[0135] The input module is specifically used for:
[0136] The training set is input into a pre-built employee intended position prediction model.
[0137] Optionally, the apparatus 200 further includes:
[0138] A verification module is used to verify the trained employee intended job prediction model using the verification set.
[0139] Optionally, the prediction module 250 includes:
[0140] An acquisition unit, used to acquire employee information of a target employee;
[0141] The prediction unit is used to input the employee information of the target employee into the trained employee intended position prediction model to obtain the intended position prediction result of the target employee.
[0142] Optionally, the apparatus 200 further includes:
[0143] A second preprocessing module, configured to preprocess the employee information of the target employee;
[0144] The second preprocessing module is specifically used for:
[0145] Performing data cleansing on the employee information of the target employee to obtain cleansed employee information;
[0146] digitizing the string data in the cleaned employee information, binarizing the discrete data in the cleaned employee information, and normalizing the continuous data in the cleaned employee information to obtain processed employee information;
[0147] extracting target features from the processed employee information to construct a second feature vector;
[0148] The job transfer tendency is extracted from the processed employee information to construct a second target vector.
[0149] Optionally, extracting target features from the processed employee information to construct a feature vector includes:
[0150] Pre-filling missing values in the processed employee information;
[0151] Target features are extracted from the pre-filled employee information to construct the second feature vector.
[0152] The embodiments of the present application also provide corresponding devices and computer storage media for implementing the solutions provided by the embodiments of the present application.
[0153] The device includes a memory and a processor, the memory is used to store instructions or codes, and the processor is used to execute the instructions or codes so that the device executes the method described in any embodiment of the present application.
[0154] The computer storage medium stores code, and when the code is executed, the device executing the code implements the method described in any embodiment of the present application.
[0155] Through the description of the above embodiments, it can be seen that those skilled in the art can clearly understand that all or part of the steps in the above-mentioned embodiment methods can be implemented by means of software plus a general hardware platform. Based on this understanding, the technical solution of the present application can be embodied in the form of a software product. The computer software product can be stored in a storage medium, such as a read-only memory (ROM) / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network communication device such as a router) to execute the methods described in various embodiments or certain parts of the embodiments of the present application.
[0156] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.
[0157] It should also be noted that the various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the equipment and device embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments. The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components indicated as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without expending creative work.
[0158] The above is merely one specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.
Claims
1. A method for predicting employees' intended positions, characterized by: include: Obtaining sample employee information; Inputting the employee information sample into a pre-built employee intended position prediction model; Adjusting the model parameters of the employee intended position prediction model through an optimization algorithm, and filling in the missing data in the employee information sample to obtain an updated employee information sample; Iteratively training the employee intended position prediction model using the updated employee information sample to obtain a trained employee intended position prediction model; The trained employee intended position prediction model is used to predict the intended position of the target employee.
2. The method according to claim 1, characterized in that Before inputting the employee information sample into the pre-built employee intended position prediction model, the method further includes: Preprocessing the employee information sample; The preprocessing of the employee information sample includes: Performing data cleaning on the employee information sample to obtain a cleaned employee information sample; performing numerical processing on the string data in the cleaned employee information sample, performing binarization processing on the discrete data in the cleaned employee information sample, and performing normalization processing on the continuous data in the cleaned employee information sample to obtain a processed employee information sample; extracting target features from the processed employee information sample to construct a first feature vector; The job transfer tendency is extracted from the processed employee information sample to construct a first target vector.
3. The method according to claim 2, characterized in that Extracting target features from the processed employee information sample to construct a first feature vector includes: Pre-filling missing values in the processed employee information sample; Target features are extracted from the pre-filled employee information sample to construct the first feature vector.
4. The method according to claim 1, wherein Before inputting the employee information sample into the pre-built employee intended position prediction model, the method further includes: Splitting the employee information sample into a training set and a validation set; Inputting the employee information sample into a pre-built employee intended position prediction model includes: The training set is input into a pre-built employee intended position prediction model.
5. The method according to claim 4, characterized in that After iteratively training the employee intended position prediction model using the updated employee information sample to obtain the trained employee intended position prediction model, the method further includes: The trained employee intended job prediction model is verified using the verification set.
6. The method according to claim 1, characterized in that The method of using the trained employee intended position prediction model to predict the intended position of the target employee includes: Obtain employee information of target employees; The employee information of the target employee is input into the trained employee intended position prediction model to obtain a prediction result of the target employee's intended position.
7. The method according to claim 6, characterized in that Before inputting the employee information of the target employee into the trained employee intended position prediction model to obtain the intended position prediction result of the target employee, the method further includes: Preprocessing the employee information of the target employee; The pre-processing of the employee information of the target employee includes: Performing data cleansing on the employee information of the target employee to obtain cleansed employee information; digitizing the string data in the cleaned employee information, binarizing the discrete data in the cleaned employee information, and normalizing the continuous data in the cleaned employee information to obtain processed employee information; extracting target features from the processed employee information to construct a second feature vector; The job transfer tendency is extracted from the processed employee information to construct a second target vector.
8. The method according to claim 7, characterized in that Extracting target features from the processed employee information to construct a feature vector includes: Pre-filling missing values in the processed employee information; Target features are extracted from the pre-filled employee information to construct the second feature vector.
9. A device for predicting employees' intended positions, characterized in that: include: Acquisition module, used to obtain employee information samples; An input module, configured to input the employee information sample into a pre-built employee intended position prediction model; A filling module is used to adjust the model parameters of the employee intended position prediction model through an optimization algorithm and fill in the missing data in the employee information sample to obtain an updated employee information sample; A training module, configured to iteratively train the employee intended position prediction model using the updated employee information samples to obtain a trained employee intended position prediction model; The prediction module is used to use the trained employee intended position prediction model to predict the intended position of the target employee.
10. The device according to claim 9, characterized in that The device further comprises: A first preprocessing module, configured to preprocess the employee information sample; The first preprocessing module is specifically used for: Performing data cleaning on the employee information sample to obtain a cleaned employee information sample; performing numerical processing on the string data in the cleaned employee information sample, performing binarization processing on the discrete data in the cleaned employee information sample, and performing normalization processing on the continuous data in the cleaned employee information sample to obtain a processed employee information sample; extracting target features from the processed employee information sample to construct a first feature vector; The job transfer tendency is extracted from the processed employee information sample to construct a first target vector.