A method and device for processing human resource information

Abstract

The present invention provides a method and apparatus for processing human resource information. The method includes: obtaining a plurality of resume data; preprocessing the plurality of resume data to obtain preprocessed resume data; extracting features from the preprocessed resume data to obtain feature vectors; determining a linear combination result according to the feature vectors and the model weights of a job matching model; obtaining a first matching result according to the linear combination result and a preset conversion function; performing a first screening on the plurality of resume data according to the first matching result and a preset screening value to obtain a first screening result; and performing a second screening on the plurality of resume data according to the first screening result, preset screening conditions and a resume screening model to obtain a target screening result. By performing two screenings on the resumes, the present invention can improve the matching degree between the resumes and the target job, which is beneficial to improving the screening efficiency and accuracy of the resumes.

Description

Technical Field

[0001] The present invention relates to the technical field of human resources, and also relates to a method and device for processing human resource information. Background Art

[0002] In human resource management, resume screening is a very important link. With the development of Internet technology, the number of resumes submitted by job seekers through online platforms has increased sharply, which makes the human resources department face the pressure of screening a large number of resumes. Traditional resume screening methods rely too much on keyword matching, resulting in some candidates who are highly matched with the position being missed because of improper use of keywords, or some candidates who do not meet the requirements being wrongly included in the consideration range because of "keyword stuffing". Therefore, how to solve the low accuracy of resume screening has become an urgent technical problem to be solved. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method and device for processing human resource information.

[0004] To solve the above technical problem, the technical solution of the present invention is as follows:

[0005] In a first aspect of the present invention, a method for processing human resource information is provided, including:

[0006] Obtain a plurality of resume data;

[0007] Preprocess the plurality of resume data to obtain preprocessed resume data;

[0008] Extract features from the preprocessed resume data to obtain feature vectors;

[0009] Determine a linear combination result according to the feature vectors and the model weights of the position matching model; the position matching model is obtained by training a preset network model according to the collected historical resume data and the corresponding position matching degree;

[0010] Obtain a first matching result according to the linear combination result and a preset conversion function;

[0011] Perform a first screening on the plurality of resume data according to the first matching result and a preset screening value to obtain a first screening result;

[0012] Perform a second screening on the plurality of resume data according to the first screening result, a preset screening condition and a resume screening model to obtain a target screening result.

[0013] Optionally, preprocessing the plurality of resume data to obtain preprocessed resume data includes:

[0014] Clean the multiple resume data to obtain cleaned data;

[0015] Perform word segmentation on the cleaned data to obtain preprocessed resume data.

[0016] Optionally, perform feature extraction on the preprocessed resume data to obtain feature vectors, including:

[0017] Extract keywords from the preprocessed resume data to obtain keywords;

[0018] Perform format conversion on the keywords to obtain word vectors;

[0019] Determine feature vectors based on the keywords and the word vectors.

[0020] Optionally, obtain a first matching result based on the linear combination result and a preset conversion function, including:

[0021] Obtain a preset conversion function; the preset conversion function is ;

[0022] Obtain a first matching result based on the linear combination result and the preset conversion function; the first matching result includes the probability value of the matching degree between the resume and the target position;

[0023] Among them, is the probability value of the matching degree between the resume and the target position, z is the linear combination result, and e is the base of the natural logarithm.

[0024] Optionally, the training process of the position matching model includes:

[0025] Obtain sample data; the sample data includes historical resume data and the corresponding position matching degree;

[0026] Train a preset network model with initialized model parameters according to the sample data to obtain a first preset network model;

[0027] Determine the target model parameters of the first preset network model according to a preset loss function or a preset number of iterations;

[0028] Determine a position matching model according to the target model parameters.

[0029] Optionally, perform a first screening on the multiple resume data according to the first matching result and a preset screening value to obtain a first screening result, including:

[0030] Sort the multiple resume data according to the first matching result to obtain a sorting result;

[0031] Perform a first screening on the multiple resume data according to the sorting result and a preset screening value to obtain a first screening result.

[0032] Optionally, perform a second screening on the multiple resume data according to the first screening result, a preset screening condition, and a resume screening model to obtain a target screening result, including:

[0033] Obtain a preset screening condition; the preset screening condition includes an educational background condition and a work experience condition;

[0034] Extract key features from the first screening result according to the preset screening condition to obtain key feature data;

[0035] Perform a second screening on the multiple resume data according to the key feature data and the resume screening model to obtain a target screening result.

[0036] In a second aspect of the present invention, a processing device for human resource information is provided, including:

[0037] An acquisition module, configured to acquire multiple resume data;

[0038] A processing module, configured to perform preprocessing on the multiple resume data to obtain preprocessed resume data; perform feature extraction on the preprocessed resume data to obtain feature vectors; determine a linear combination result according to the feature vectors and the model weights of a position matching model; the position matching model is obtained by training a preset network model according to the collected historical resume data and the corresponding position matching degree; obtain a first matching result according to the linear combination result and a preset conversion function; perform a first screening on the multiple resume data according to the first matching result and a preset screening value to obtain a first screening result; perform a second screening on the multiple resume data according to the first screening result, a preset screening condition, and a resume screening model to obtain a target screening result.

[0039] In a third aspect of the present invention, a computing device is provided, including: a processor and a memory storing a computer program, and when the computer program is run by the processor, the method described in the first aspect is executed.

[0040] In a fourth aspect of the present invention, a computer-readable storage medium is provided, storing instructions, and when the instructions are run on a computer, the computer is made to execute the method described in the first aspect.

[0041] The above solution of the present invention has at least the following beneficial effects:

[0042] In the above solution of the present invention, by preprocessing the obtained multiple resume data, then extracting features to obtain feature vectors, determining the linear combination result according to the feature vectors and the model weights of the job matching model, obtaining the first matching result according to the linear combination result and the preset conversion function, and performing the first screening on the multiple resume data according to the first matching result and the preset screening value to obtain the first screening result, and then performing the second screening on the multiple resume data according to the first screening result, the preset screening conditions and the resume screening model to obtain the target screening result. Through the two screenings, the matching degree between the obtained resumes and the target job can be improved, which is beneficial to improving the screening efficiency and accuracy of the resumes. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 is a schematic flowchart of the method for processing human resource information in an embodiment of the present invention;

[0044] Figure 2 is a schematic structural diagram of the device for processing human resource information in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0045] Hereinafter, exemplary embodiments of the present invention will be described in more detail with reference to the drawings. Although the exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present invention can be more thoroughly understood and the scope of the present invention can be fully conveyed to those skilled in the art.

[0046] As Figure 1 shown, an embodiment of the present invention provides a method for processing human resource information, including the following steps:

[0047] Step 101, obtaining multiple resume data;

[0048] Step 102, preprocessing the multiple resume data to obtain preprocessed resume data;

[0049] Step 103, extracting features from the preprocessed resume data to obtain feature vectors;

[0050] Step 104, determining a linear combination result according to the feature vectors and the model weights of the job matching model; the job matching model is obtained by training a preset network model according to the collected historical resume data and the corresponding job matching degree;

[0051] Step 105, obtaining a first matching result according to the linear combination result and a preset conversion function;

[0052] Step 106: Perform a first screening on the multiple resume data according to the first matching result and a preset screening value to obtain a first screening result;

[0053] Step 107: Perform a second screening on the multiple resume data according to the first screening result, a preset screening condition, and a resume screening model to obtain a target screening result.

[0054] In the method for processing human resource information according to the embodiment of the present invention, by preprocessing the obtained multiple resume data, then performing feature extraction to obtain feature vectors, determining a linear combination result according to the feature vectors and the model weights of the job matching model, obtaining a first matching result according to the linear combination result and a preset conversion function, performing a first screening on the multiple resume data according to the first matching result and a preset screening value to obtain a first screening result, and then performing a second screening on the multiple resume data according to the first screening result, a preset screening condition, and a resume screening model to obtain a target screening result. Through the two screenings, the matching degree between the obtained resumes and the target job can be improved, which is beneficial to improving the screening efficiency and accuracy of the resumes.

[0055] In an alternative embodiment of the present invention, the obtaining of the multiple resume data in step 101 may be obtaining the multiple resume data from channels such as recruitment platforms and internal databases. The resume data may include information such as personal basic information, work experience, education level, and work skills.

[0056] In an alternative embodiment of the present invention, step 102 includes:

[0057] Step 1021: Clean the multiple resume data to obtain cleaned data;

[0058] Specifically, data cleaning is to remove irrelevant characters, HTML (HyperText Markup Language) tags, special symbols, etc. from the resume data to ensure the purity of the text data and improve the subsequent processing efficiency.

[0059] Step 1022: Perform word segmentation processing on the cleaned data to obtain preprocessed resume data.

[0060] Specifically, use a pre-stored dictionary to cut the text in the cleaned data into independent lexical units to obtain a first word segmentation result; delete the words in the preset stop word list from the word segmentation result, such as "de", "shi", "zai", etc., to obtain a second word segmentation result; recombine the second word segmentation result into text to obtain preprocessed resume data for subsequent processing. The purpose of performing word segmentation processing is to provide high-quality data input for subsequent processing, which is beneficial to improving the processing efficiency and the accuracy of the result.

[0061] In an alternative embodiment of the present invention, step 103 includes:

[0062] Step 1031: Extract keywords from the preprocessed resume data to obtain keywords.

[0063] Specifically, by counting the word frequency of each word in a single preprocessed resume data; then counting the occurrences of each word in all preprocessed resume data and calculating the inverse document frequency ; multiplying the TF and IDF of each word to obtain the TF-IDF value; finally, sorting the vocabulary according to the TF-IDF value and extracting the top-ranked vocabulary as keywords.

[0064] Step 1032: Convert the format of the keywords to obtain word vectors.

[0065] Specifically, the keywords can be input into a trained word embedding model for format conversion. The processing process of the word embedding model for keywords includes: in the model, each keyword q corresponds to a word vector , where a is the embedding dimension (e.g., a can be 300), and the word vector is obtained by extracting through the weight matrix of the word embedding model, and the calculation formula is v q =Q[q] , where c is the size of the vocabulary, and the word vector of keyword q is the q-th row of the weight matrix Q.

[0066] Step 1033: Determine the feature vector according to the keywords and the word vectors.

[0067] Specifically, the feature vector can be calculated through , where F is the feature vector, n is the number of keywords, is the word vector of the i-th keyword. The feature vector is used as the input for the subsequent model.

[0068] In an optional embodiment of the present invention, step 104 includes:

[0069] Step 10411: Determine the linear combination result according to the feature vector, the model weights of the job matching model, and .

[0070] Specifically, the feature vector and the model weights of the job matching model can be input into to obtain the linear combination result z, where F is the feature vector, w is the target weight vector in the model weights, and b is the target bias term in the model weights.

[0071] In an optional embodiment of the present invention, the training process of the job matching model in step 104 includes:

[0072] Step 10421: Obtain sample data; the sample data includes historical resume data and the corresponding job matching degree.

[0073] Specifically, the sample data can be the resume data and the corresponding job matching degree within a preset time period in history. The specific value of the preset time period can be set according to needs, such as within the past year or within the past three years, etc. Obtaining the sample data can be used for subsequent training of the model by the user.

[0074] Step 10422: Train a preset network model with initialized model parameters according to the sample data to obtain a first preset network model.

[0075] Step 10423: Determine the target model parameters of the first preset network model according to a preset loss function or a preset number of iterations.

[0076] Specifically, the preset network model initializes the weight vector w and the bias term b, initialized to zero or small random numbers. Then, the sample data is input into the initialized preset network model for training to obtain a first preset network model. If the preset loss function of the first preset network model converges to a smaller value (such as a preset convergence value) or reaches the preset number of iterations, then the model parameters of the first preset network model are determined as the target model parameters.

[0077] Among them, the preset loss function is L w,b =- 1 m ∑ i=1 m [ y i log y i ^ +(1- y i ) log 1- y i ^ ] , where m is the number of samples in the sample data, w is the weight vector, b is the bias term, is the predicted probability, is the true matching degree.

[0078] Step 10424: Determine a job matching model according to the target model parameters.

[0079] Specifically, input the target model parameters into the preset network model to determine the job matching model.

[0080] In an optional embodiment of the present invention, step 105 includes:

[0081] Step 1051: Obtain a preset conversion function; the preset conversion function is ;

[0082] Step 1052: Obtain a first matching result according to the linear combination result and the preset conversion function; the first matching result includes the probability value of the matching degree between the resume and the target job.

[0083] Among them, is the probability value of the matching degree between the resume and the target position, z is the result of the linear combination, and e is the base of the natural logarithm.

[0084] Specifically, through the formula the first matching result is obtained, where is the probability value of the matching degree between the resume and the target position output by the preset conversion function, representing the probability of the resume matching the position, which is between 0 and 1, z is the result of the linear combination, and e is the base of the natural logarithm (approximately equal to 2.71828). Here, the first matching result includes each resume data and its corresponding probability value, which is convenient for subsequent sorting and screening.

[0085] In an optional embodiment of the present invention, step 106 includes:

[0086] Step 1061, sorting the multiple resume data according to the first matching result to obtain a sorting result;

[0087] Specifically, the resume data can be sorted from high to low according to the probability value in the first matching result to obtain a sorting result, and the sorting result includes the sorting serial number, the corresponding probability value, and the resume data.

[0088] Step 1062, performing a first screening on the multiple resume data according to the sorting result and a preset screening value to obtain a first screening result.

[0089] Specifically, the preset screening value can be the number of resumes selected. For example, in a specific embodiment, the preset screening value is 5, then the first 5 resume data in the sorting result are extracted, and the first screening result includes the first 5 resume data, the sorting serial number, the corresponding probability value, etc.

[0090] In an optional embodiment of the present invention, step 107 includes:

[0091] Step 1071, obtaining a preset screening condition; the preset screening condition includes an educational background condition and a work experience condition;

[0092] Specifically, the educational background condition, work experience condition, educational level condition, professional skill condition, etc. required for the target position can be obtained for further screening of the resume data later to improve the accuracy of the final screening. In a specific embodiment, the preset screening condition includes: the educational background condition is a bachelor's degree, the work experience condition is at least one year, and the professional skill condition is having a CET-6 certificate. The preset screening condition can be set according to the specific position requirements.

[0093] Step 1072, extracting key features from the first screening result according to the preset screening condition to obtain key feature data;

[0094] Specifically, if the preset screening conditions include educational background conditions, work experience conditions, education level conditions, professional skill conditions, etc., then the educational background conditions, work experience conditions, education level conditions, and professional skill conditions in the resumes in the first screening result are respectively encoded to obtain numerical data. For example, if the educational background condition in a resume is undergraduate, the corresponding numerical data is 1, and if the educational background condition is master, the corresponding numerical data is 2, and so on. The key feature data includes the numerical data corresponding to the preset screening conditions. In a specific embodiment, the key feature data includes 1 (educational background condition is undergraduate), 0.1 (work experience condition is 1 year), 2 (professional skill condition is having a CET-6 certificate).

[0095] Step 1073, perform a second screening on the multiple resume data according to the key feature data and the resume screening model to obtain a target screening result.

[0096] Specifically, input the key feature data into the resume screening model, and use the formula , , to obtain the probability value that the resume data meets the preset screening conditions. If the probability value that the resume data meets the preset screening conditions is greater than the preset screening probability value, it can be determined as the target resume. Among them, is the decision function, is the weight vector of the resume screening model, T is the transpose operation, is the bias term of the resume screening model, is the mapping of the key feature data in the high-dimensional space, is the probability value that the resume data meets the preset screening conditions, and both A and B are fitting parameters (in a specific embodiment, A is -0.1 and B is 0.5).

[0097] Here, the target screening result may include target resume data, the probability value in the first screening result, the probability value that the resume data meets the preset screening conditions, etc. Among them, the training process of the resume screening model includes:

[0098] Collect historical resume data; each piece of resume data is marked with whether it meets the preset screening conditions (such as 1 means meeting, 0 means not meeting).

[0099] Perform data cleaning and data format conversion on the historical resume data to obtain a sample set; the purpose of data cleaning is to improve the accuracy of the training set and the effectiveness of subsequent model training. Among them, data format conversion is to convert non-numerical data in the historical resume data into numerical data.

[0100] Divide the sample set into a training set and a test set; 80% of the sample set can be divided into the training set and 20% into the test set.

[0101] Initialize the parameters of the preset network model, which may include: determining the kernel function as , setting the regularization parameter C to 1.0, where is the key feature data of the i-th historical resume data, is the key feature data of the j-th historical resume data, T represents the transpose operation, is the kernel function, which is used to map and in the low-dimensional feature space to the high-dimensional space.

[0102] Use the training set to train the initialized preset network model to obtain the trained model; then use the test set and the preset verification metrics to verify the performance of the trained model. If the verification metrics reach the preset metric values, then use the trained model as the resume screening model. Among them, the preset verification metrics may be: accuracy and / or recall , where is the accuracy, is the recall, TP is the number of samples that are actually positive and are predicted as positive by the model, TN is the number of samples that are actually negative and are predicted as negative by the model, FP is the number of samples that are actually negative but are predicted as positive by the model (misjudgment), and FN is the number of samples that are actually positive but are predicted as negative by the model (omission).

[0103] In an optional embodiment of the present invention, the method further includes:

[0104] Step 108, presenting the target screening result in a visual manner.

[0105] Specifically, the target screening result can be converted according to a preset presentation type (such as a chart type), which is convenient for human resource managers or recruiters to intuitively see the resume screening result and facilitate their recruitment decisions based on the scoring and ranking.

[0106] A specific embodiment of the human resource information processing method according to the embodiment of the present invention includes:

[0107] Step 111, obtain multiple resume data;

[0108] Obtain multiple resume data from channels such as recruitment platforms and internal databases for subsequent screening.

[0109] Step 112, preprocess;

[0110] Perform data cleaning and word segmentation on the multiple resume data to obtain preprocessed resume data, so as to improve data accuracy and subsequent processing efficiency.

[0111] Step 113, feature extraction;

[0112] By performing keyword extraction on the preprocessed resume data, calculating word vectors, and determining feature vectors, it is convenient for subsequent model input and processing.

[0113] Step 114, calculate the job matching probability;

[0114] Use the trained job matching model to calculate the job matching probability values for each resume data, providing a basis for subsequent resume screening, which is conducive to saving labor and improving screening efficiency.

[0115] Step 115, first screening;

[0116] First, sort the resume data from high to low according to the probability values in the first matching result to obtain a sorting result, and then screen out the top N resume data in the sorting result according to a preset screening value as the first screening result, where N is less than or equal to the preset screening value.

[0117] Step 116, second screening.

[0118] During the second screening, it is necessary to further screen the resume data in the first screening result according to the specific conditions required by the target position to improve the matching degree between the resume and the target position. The finally screened resume data is the target screening result, and the target screening result is presented in a visual manner in a preset way for human resource managers or recruiters to view.

[0119] The method for processing human resource information according to the embodiment of the present invention saves the time for recruiters to screen out personnel matching the target position from a large number of resume data through multiple screenings of the obtained multiple resume data, which is conducive to saving labor costs and improving resume screening efficiency.

[0120] As Figure 2 shown, an embodiment of the present invention proposes a human resource information processing device 200, including:

[0121] An acquisition module 201 for acquiring multiple resume data;

[0122] The processing module 202 is configured to preprocess the multiple resume data to obtain preprocessed resume data; extract features from the preprocessed resume data to obtain feature vectors; determine a linear combination result according to the feature vectors and the model weights of the job matching model; the job matching model is obtained by training a preset network model according to the collected historical resume data and the corresponding job matching degree; obtain a first matching result according to the linear combination result and a preset conversion function; perform a first screening on the multiple resume data according to the first matching result and a preset screening value to obtain a first screening result; and perform a second screening on the multiple resume data according to the first screening result, a preset screening condition, and a resume screening model to obtain a target screening result.

[0123] Optionally, preprocessing the multiple resume data to obtain preprocessed resume data includes:

[0124] Performing data cleaning on the multiple resume data to obtain cleaned data;

[0125] Performing word segmentation on the cleaned data to obtain preprocessed resume data.

[0126] Optionally, extracting features from the preprocessed resume data to obtain feature vectors includes:

[0127] Extracting keywords from the preprocessed resume data to obtain keywords;

[0128] Converting the format of the keywords to obtain word vectors;

[0129] Determining feature vectors according to the keywords and the word vectors.

[0130] Optionally, obtaining a first matching result according to the linear combination result and a preset conversion function includes:

[0131] Obtaining a preset conversion function; the preset conversion function is ;

[0132] Obtaining a first matching result according to the linear combination result and the preset conversion function; the first matching result includes the probability value of the matching degree between the resume and the target position;

[0133] Wherein, is the probability value of the matching degree between the resume and the target position, z is the linear combination result, and e is the base of the natural logarithm.

[0134] Optionally, the training process of the job matching model includes:

[0135] Obtaining sample data; the sample data includes historical resume data and the corresponding job matching degree;

[0136] Train a preset network model with initialized model parameters according to the sample data to obtain a first preset network model;

[0137] Determine the target model parameters of the first preset network model according to a preset loss function or a preset number of iterations;

[0138] Determine a job matching model according to the target model parameters.

[0139] Optionally, perform a first screening on the multiple resume data according to the first matching result and a preset screening value to obtain a first screening result, including:

[0140] Sort the multiple resume data according to the first matching result to obtain a sorting result;

[0141] Perform a first screening on the multiple resume data according to the sorting result and the preset screening value to obtain a first screening result.

[0142] Optionally, perform a second screening on the multiple resume data according to the first screening result, preset screening conditions, and a resume screening model to obtain a target screening result, including:

[0143] Obtain preset screening conditions; the preset screening conditions include educational background conditions and work experience conditions;

[0144] Extract key features from the first screening result according to the preset screening conditions to obtain key feature data;

[0145] Perform a second screening on the multiple resume data according to the key feature data and the resume screening model to obtain a target screening result.

[0146] The human resource information processing device according to the embodiment of the present invention preprocesses the obtained multiple resume data, then extracts feature vectors, determines a linear combination result according to the feature vectors and the model weights of the job matching model, obtains a first matching result according to the linear combination result and a preset conversion function, performs a first screening on the multiple resume data according to the first matching result and a preset screening value to obtain a first screening result, and then performs a second screening on the multiple resume data according to the first screening result, preset screening conditions, and a resume screening model to obtain a target screening result. Through the two screenings, the matching degree of the obtained resumes and the target positions can be improved, which is beneficial to improving the screening efficiency and accuracy of the resumes.

[0147] It should be noted that this device corresponds to the above method, and all implementation manners in the above method embodiments are applicable to the embodiments of this device and can achieve the same technical effects. Details are not described herein again.

[0148] An embodiment of the present invention further provides a computing device, including: a processor and a memory storing a computer program. When the computer program is run by the processor, the method described in any one of the above embodiments is executed. All implementation manners in the above method embodiments are applicable to the embodiments of this device and can achieve the same technical effects. Details are not described herein again.

[0149] An embodiment of the present invention further provides a computer-readable storage medium, on which instructions are stored. When the instructions are run on a computer, the computer is caused to execute the method described in any one of the above embodiments. All implementation manners in the above method embodiments are applicable to the embodiments of this device and can achieve the same technical effects. Details are not described herein again.

[0150] It should be noted that in the device and method of the present invention, obviously, each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent solutions of the present invention. And, the steps of performing the above series of processes can naturally be executed in chronological order according to the described order, but it is not necessary to be executed in chronological order. Certain steps can be executed in parallel, crosswise or independently of each other.

[0151] It should be noted that in the above embodiments, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article or device including that element. In addition, it should be pointed out that the scope of the method and device in the above embodiments of the implementation manner is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in a reverse order according to the functions involved. For example, the described method may be executed in an order different from that described, and various steps may be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0152] The above is the preferred implementation manner of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A method for processing human resource information, characterized in that Including: Obtain multiple resume data; The resume data includes personal basic information, work experience, education level, and work skill information; Preprocess the multiple resume data to obtain preprocessed resume data; Extract features from the preprocessed resume data to obtain feature vectors; Determine the linear combination result according to the feature vectors and the model weights of the job matching model; the job matching model is obtained by training a preset network model according to the collected historical resume data and the corresponding job matching degree; Obtain the first matching result according to the linear combination result and a preset conversion function; Perform the first screening on the multiple resume data according to the first matching result and a preset screening value to obtain the first screening result; Perform the second screening on the multiple resume data according to the first screening result, preset screening conditions, and a resume screening model to obtain the target screening result; Among them, obtaining the first matching result according to the linear combination result and a preset conversion function includes: Obtain a preset conversion function; the preset conversion function is ; Obtain the first matching result according to the linear combination result and the preset conversion function; the first matching result includes the probability value of the matching degree between the resume and the target job; Among them, is the probability value of the matching degree between the resume and the target position, z is the result of the linear combination, and e is the base of the natural logarithm; Among them, preprocessing the multiple resume data to obtain preprocessed resume data includes: Perform data cleaning on the multiple resume data to obtain cleaned data; Perform word segmentation on the cleaned data to obtain preprocessed resume data.

2. The method for processing human resource information according to claim 1, wherein Extracting features from the preprocessed resume data to obtain feature vectors includes: Extract keywords from the preprocessed resume data to obtain keywords; Convert the format of the keywords to obtain word vectors; Determine the feature vectors according to the keywords and the word vectors.

3. The processing method of human resource information according to claim 1, wherein The training process of the job matching model includes: Obtain sample data; the sample data includes historical resume data and the corresponding job matching degree; Train a preset network model with initialized model parameters according to the sample data to obtain a first preset network model; Determine the target model parameters of the first preset network model according to a preset loss function or a preset number of iterations; Determine the job matching model according to the target model parameters.

4. The method for processing human resource information according to claim 1, wherein Performing the first screening on the multiple resume data according to the first matching result and a preset screening value to obtain the first screening result includes: Sort the multiple resume data according to the first matching result to obtain a sorting result; Perform the first screening on the multiple resume data according to the sorting result and the preset screening value to obtain the first screening result.

5. The processing method of human resource information according to claim 1, characterized in that Performing the second screening on the multiple resume data according to the first screening result, preset screening conditions, and a resume screening model to obtain the target screening result includes: Obtain preset screening conditions; the preset screening conditions include educational background conditions and work experience conditions; Extract key features from the first screening result according to the preset screening conditions to obtain key feature data; Perform the second screening on the multiple resume data according to the key feature data and the resume screening model to obtain the target screening result.

6. A processing device for human resource information, characterized in that, Including: An acquisition module for acquiring multiple resume data; the resume data includes personal basic information, work experience, educational background, and work skill information; A processing module for preprocessing the multiple resume data to obtain preprocessed resume data; Performing feature extraction on the preprocessed resume data to obtain feature vectors; determining a linear combination result according to the feature vectors and the model weights of a job matching model; the job matching model is obtained by training a preset network model according to the collected historical resume data and the corresponding job matching degree; Obtaining a first matching result according to the linear combination result and a preset conversion function; Performing a first screening on the multiple resume data according to the first matching result and a preset screening value to obtain a first screening result; Performing a second screening on the multiple resume data according to the first screening result, a preset screening condition, and a resume screening model to obtain a target screening result; Among them, obtaining a first matching result according to the linear combination result and a preset conversion function includes: Obtain a preset conversion function; the preset conversion function is ; Obtaining a first matching result according to the linear combination result and the preset conversion function; the first matching result includes the probability value of the matching degree between the resume and the target job; Among them, is the probability value of the matching degree between the resume and the target position, z is the result of the linear combination, and e is the base of the natural logarithm; Among them, preprocessing the multiple resume data to obtain preprocessed resume data includes: Performing data cleaning on the multiple resume data to obtain cleaned data; Performing word segmentation on the cleaned data to obtain preprocessed resume data.

7. A computing device, characterized in that, Including: A processor and a memory storing a computer program, and when the computer program is run by the processor, it executes the method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, Storing instructions, and when the instructions are run on a computer, the computer is made to execute the method according to any one of claims 1 to 5.

Images