CN117114514A

CN117114514A - Talent information analysis management method, system and device based on big data

Info

Publication number: CN117114514A
Application number: CN202311377485.1A
Authority: CN
Inventors: 黄思齐; 曹杨; 谢真强; 龙光涛; 卢南方; 张南楠; 甘盛霖
Original assignee: Guizhou Vocational Technology College Of Electronics & Information; CETC Big Data Research Institute Co Ltd
Current assignee: Guizhou Vocational Technology College Of Electronics & Information; CETC Big Data Research Institute Co Ltd
Priority date: 2023-10-24
Filing date: 2023-10-24
Publication date: 2023-11-24
Anticipated expiration: 2043-10-24
Also published as: CN117114514B

Abstract

The application discloses a talent information analysis and management method, system and device based on big data, which are used for reducing labor cost and accurately providing talent recommendation for enterprises. The method comprises the following steps: acquiring talent resume information of talents to be analyzed; carrying out identity confirmation on talent resume information based on the identity information of talents to be analyzed; after the identity is confirmed, inputting talent resume information into an information extraction model to extract keywords, and determining talent data information; directionally storing talent data information to a talent information management database; when the talent demand of the target enterprise is determined, talent demand intention information of the target enterprise is determined, and a plurality of talent data information is acquired from a talent information management database; calculating the matching degree of talent data information and talent demand intention information by using a data matching algorithm; and sorting the matching degrees from high to low, determining the matching level, and determining the target talents according to the matching level.

Description

Talent information analysis management method, system and device based on big data

Technical Field

The application relates to the field of data processing, in particular to a talent information analysis and management method, system and device based on big data.

Background

In the current society, the competition of the market is the competition of talents, and each stage of the enterprise operation strategy must have proper talents as support.

At present, the enterprise searches talents matched with positions mainly depend on a search function of a service platform or performs off-line market recruitment to collect talent resume, talent information of the talents needs to be screened manually, a large amount of labor cost is spent, and the final screening result is a rough result and cannot be recommended to the enterprise well.

Disclosure of Invention

The application provides a talent information analysis management method, system and device based on big data, which are used for automatically determining target talents corresponding to target enterprise talent demands, and talent information of job seekers does not need to be manually screened, so that labor cost is reduced, talent recommendation is accurately provided for enterprises, and user experience is improved.

The first aspect of the application provides a talent information analysis and management method based on big data, which comprises the following steps:

acquiring talent resume information of talents to be analyzed;

carrying out identity confirmation on the talent resume information based on the identity information of the talents to be analyzed;

After the identity confirmation is passed, inputting the talent resume information into an information extraction model to extract keywords, and determining talent data information, wherein the talent data information comprises identity information, learning history information, work history information, skill information and job hunting intention information, and the keyword extraction process comprises the following steps:

extracting specific keywords based on an NLP technology, performing field matching on the extracted keywords and a pre-constructed vocabulary to obtain corresponding fields and field values, storing the fields and the field values through a data structure to obtain talent data information, and pre-training an information extraction model by inputting a plurality of talent resume information samples, wherein the specific keywords comprise keywords related to identity information, academic information, work history information, skill information and job hunting intention information;

directionally storing the talent data information to a talent information management database;

when determining that the target enterprise has talent demand, determining talent demand intention information of the target enterprise, and acquiring a plurality of talent data information from a talent information management database;

determining a first characteristic of the talent data information and a second characteristic of the talent demand intention information, calculating a coincidence value of the first characteristic in accordance with the second characteristic, determining a value of importance of the talent demand intention information on the second characteristic, and calculating the matching degree of the talent data information and the talent demand intention information by using a data matching algorithm based on the first characteristic, the second characteristic, the coincidence value and the importance value, wherein the first characteristic comprises a first identity, a first academy, a first work history, a first skill and a first job intention, and the second characteristic comprises a second identity, a second academy, a second work history, a second skill and a second job intention;

And sorting the matching degrees from high to low, determining a matching level, and determining target talents according to the matching level.

Optionally, the identifying the talent resume information based on the identity information of the talents to be analyzed includes:

determining a first face image of the talents to be analyzed based on the identity information of the talents to be analyzed;

inputting the first face image and the talent resume information into a face discrimination model, determining a second face image contained in the talent resume information, extracting first face contour features of the first face image and second face contour features of the second face image, extracting first face five-sense organ point features of the first face image and second face five-sense organ point features of the second face image, and pre-training the face discrimination model by inputting a plurality of face image samples and talent resume information samples;

combining the first face contour feature and the first face five-sense organ point feature into a group of first feature vectors based on a full connection layer of the face discrimination model, and combining the second face contour feature and the second face five-sense organ point feature into a group of second feature vectors;

Inputting the first feature vector and the second feature vector into a softmax classifier of the face discrimination model, outputting the identity discrimination result of the talents to be analyzed, and carrying out identity confirmation on the talent resume information according to the identity discrimination result.

Optionally, the determining the first characteristic of the talent data information and the second characteristic of the talent demand intention information, calculating a coincidence value of the first characteristic with the second characteristic, determining a value of importance of the talent demand intention information on the second characteristic, and calculating the matching degree of the talent data information and the talent demand intention information by using a data matching algorithm based on the first characteristic, the second characteristic, the coincidence value and the value of importance includes:

respectively determining a first identity, a first academic, a first working history, a first skill and a first job-seeking intention of talents corresponding to the talent data information, and setting five first characteristics of i= {1,2,3,4,5} respectively representing the first identity, the first academic, the first working history, the first skill and the first job-seeking intention;

determining a second identity, a second academic, a second working history, a second skill and a second job-seeking intention of the talent demand corresponding to the talent demand intention information, and setting five second characteristics of j= {1,2,3,4,5} respectively representing the second identity, the second academic, the second working history, the second skill and the second job-seeking intention;

Calculating a coincidence value of the first characteristic i with the second characteristic j, and determining a value of importance of the talent demand intention information on the second characteristic j;

calculating the matching degree of the talent data information and the talent demand intention information based on a data matching algorithm, wherein the data matching algorithm is as follows:

wherein P represents the matching degree, u represents the talent data information, v represents the talent demand intention information of the target enterprise,weight value representing the corresponding weighting of said second property j +.>A coincidence value of a first characteristic i representing said talent data information u to said second characteristic j,/or->A value representing the importance of the talent demand intent information v to the second characteristic j;

wherein the weight value is calculated according to the weight calculation formulaThe weight calculation formula is as follows:

wherein N represents the talent information pipeThe first intention is the total number of talents of the second intention in the management database,in talents whose first job-seeking intention is second job-seeking intention, whether a first characteristic i of an a-th talent satisfies a j-th characteristic in the second characteristics, if so, the a-th talent is treated with>And assigning a preset value, and if the preset value is not met, assigning 0.

Optionally, the calculating the coincidence value of the first characteristic i, which is coincident with the second characteristic j, includes:

constructing a characteristic association vocabulary, wherein the characteristic association vocabulary comprises the first characteristic i, the second characteristic j and a coincidence value;

and detecting the first characteristic i and the second characteristic j by using the characteristic association vocabulary, if the first characteristic i accords with the second characteristic j, assigning the coincidence value to be 1, and if the first characteristic i does not accord with the second characteristic j, assigning the coincidence value to be 0.

Optionally, after the talent data information is directionally stored in the talent information management database, the talent information analysis management method further includes:

the talent data information in the talent information management database is subjected to label positioning, and the labels of the talents are positioned in an incumbent state and/or a free state according to the current working state of the talents corresponding to the talents;

the step of obtaining a plurality of talent data information from the talent information management database comprises the following steps:

and determining talent data information corresponding to the talents with the current labels positioned in the free state from the talent information management database, and acquiring the talent data information.

Optionally, the determining the target talents according to the matching level includes:

And screening talents with the matching level higher than a preset threshold, determining the talents as target talents, and directionally pushing talent data information of the target talents to recruitment ends of the target enterprises.

Optionally, the directionally storing the talent data information in a talent information management database includes:

and carrying out structuring treatment on the talent data information, and directionally storing the talent data information into a talent information management database in a json format.

setting a talent information storage limit value to be stored currently in a talent information management database;

judging whether five characteristics of identity information, learning history information, work history information, skill information and job hunting intention information included in the talent data information reach the talent information storage limit value;

if yes, directionally storing the talent data information to a first gradient talent database of the talent information management database;

if not, the talent data information is directionally stored to a standby talent database of the talent information management database, and the talent data information is marked.

The second aspect of the present application provides a talent information analysis management system based on big data, comprising:

the acquisition unit is used for acquiring talent resume information of talents to be analyzed;

the confirmation unit is used for carrying out identity confirmation on the talent resume information based on the identity information of the talents to be analyzed;

the first determining unit is configured to input the talent resume information into the information extraction model to perform keyword extraction after the identity confirmation is passed, and determine talent data information, where the talent data information includes identity information, learning information, work history information, skill information and job hunting intention information, and the keyword extraction process includes:

The storage unit is used for directionally storing the talent data information to a talent information management database;

the second determining unit is used for determining talent demand intention information of the target enterprise when determining that the target enterprise has talent demand, and acquiring a plurality of talent data information from the talent information management database;

the matching unit is used for determining a first characteristic of the talent data information and a second characteristic of the talent demand intention information, calculating a coincidence value of the first characteristic in accordance with the second characteristic, determining a value of importance of the talent demand intention information on the second characteristic, and calculating the matching degree of the talent data information and the talent demand intention information by using a data matching algorithm based on the first characteristic, the second characteristic, the coincidence value and the importance value, wherein the first characteristic comprises a first identity, a first academy, a first working history, a first skill and a first job intention, and the second characteristic comprises a second identity, a second academy, a second working history, a second skill and a second job intention;

and the third determining unit is used for determining the matching level according to the sequence of the matching degrees from high to low, and determining the target talents according to the matching level.

Optionally, the confirmation unit is specifically configured to determine a first face image of the talents to be analyzed based on the identity information of the talents to be analyzed;

Optionally, the matching unit is specifically configured to determine a first identity, a first academic, a first working history, a first skill, and a first job-seeking intent of a talent corresponding to the talent data information, and set i= {1,2,3,4,5} to respectively represent five first characteristics of the first identity, the first academic, the first working history, the first skill, and the first job-seeking intent;

wherein P represents the matching degree, and u represents the matching degreeTalent data information, v representing talent demand intent information of the target enterprise, Weight value representing the corresponding weighting of said second property j +.>A coincidence value of a first characteristic i representing said talent data information u to said second characteristic j,/or->A value representing the importance of the talent demand intent information v to the second characteristic j;

wherein N represents the total number of talents for which the first job hunting intention is the second job hunting intention in the talent information management database,in talents whose first job-seeking intention is second job-seeking intention, whether a first characteristic i of an a-th talent satisfies a j-th characteristic in the second characteristics, if so, the a-th talent is treated with>And assigning a preset value, and if the preset value is not met, assigning 0.

Optionally, the matching unit is specifically configured to construct a feature-related vocabulary, where the feature-related vocabulary includes the first feature i, the second feature j, and a coincidence value;

Optionally, the talent information analysis management system further includes:

The label positioning unit is used for carrying out label positioning on the talent data information in the talent information management database, and positioning the label of the talent data information into an incumbent state and/or a free state according to the current working state of the talent corresponding to the talent data information;

the second determining unit is specifically configured to determine talent data information corresponding to talents whose current tags are located in a free state from the talent information management database, and obtain the talent data information.

Optionally, the third determining unit is specifically configured to screen talents with a matching level higher than a preset threshold, determine the talents as target talents, and directionally push talent data information of the target talents to a recruitment end of the target enterprise.

Optionally, the storage unit is specifically configured to perform a structuring process on the talent data information, and directionally store the talent data information in a json format in a talent information management database.

Optionally, the storage unit is specifically configured to set a talent information storage limit value to be currently stored in a talent information management database;

The third aspect of the present application provides a talent information analysis and management device based on big data, the device comprising:

a processor, a memory, an input-output unit, and a bus;

the processor is connected with the memory, the input/output unit and the bus;

the memory holds a program that the processor invokes to perform the talent information analysis management method of the first aspect and any one of the alternatives of the first aspect.

A fourth aspect of the present application provides a computer-readable storage medium having a program stored thereon, which when executed on a computer performs the talent information analysis management method of the first aspect and any one of the alternatives of the first aspect.

From the above technical scheme, the application has the following advantages:

the method comprises the steps of obtaining talent resume information of talents to be analyzed; carrying out identity confirmation on talent resume information based on the identity information of talents to be analyzed; after the identity is confirmed, inputting talent resume information into an information extraction model to extract keywords, and determining talent data information, wherein the talent data information comprises identity information, learning information, work history information, skill information and job hunting intention information; directionally storing talent data information to a talent information management database; when the talent demand of the target enterprise is determined, determining talent demand intention information of the target enterprise; the method comprises the steps of obtaining a plurality of talent data information from a talent information management database, calculating the matching degree of the plurality of talent data information and talent demand intention information respectively according to a data matching algorithm, sorting the matching degree from high to low, determining a matching level, and determining target talents according to the matching level, so that target talents corresponding to target enterprise talents are automatically determined, personnel information of job seekers does not need to be screened manually, labor cost is reduced, talent recommendation is accurately provided for enterprises, and user experience is improved.

It should be noted that, all actions of acquiring signals, information or data in the present application are performed under the condition of conforming to the corresponding data protection rule policy of the country of the location and obtaining the authorization given by the owner of the corresponding device.

Drawings

In order to more clearly illustrate the technical solutions of the present application, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic flow chart of an embodiment of a talent information analysis and management method based on big data provided by the application;

FIG. 2 is a schematic flow chart of another embodiment of a talent information analysis and management method based on big data provided by the application;

FIG. 3 is a schematic diagram of an embodiment of a big data based talent information analysis management system according to the present application;

FIG. 4 is a schematic structural diagram of another embodiment of a talent information analysis management system based on big data according to the present application;

fig. 5 is a schematic structural diagram of an embodiment of a talent information analysis management device based on big data according to the present application.

Detailed Description

The talent information analysis and management method based on big data provided by the application can be applied to a terminal, and can also be applied to a server, for example, the terminal can be a smart phone or a fixed terminal such as a computer, a tablet computer, a smart television, a smart watch, a portable computer terminal and the like. For convenience of explanation, the present application is exemplified by using the terminal as the execution subject.

Referring to fig. 1, fig. 1 is a schematic diagram illustrating an embodiment of a talent information analysis and management method based on big data according to the present application, the method includes:

s101, acquiring talent resume information of talents to be analyzed;

the terminal acquires talent resume information of a plurality of talents to be analyzed according to big data, wherein the talent resume information can exist in the form of talent resume images or talent resume documents. Specifically, the terminal may obtain talent resume information of to-be-analyzed talents delivered to the target enterprise in each recruitment software, or obtain talent resume information of to-be-analyzed talents browsed by the target enterprise in each recruitment software, or obtain talent resume information of to-be-analyzed talents collected and stored in the terminal by the target enterprise offline recruitment, or obtain talent resume information of various talents from other modes based on big data, which is not limited herein.

S102, confirming the identity of talent resume information based on the identity information of talents to be analyzed;

before the terminal performs talent information analysis management on the talents to be analyzed, the terminal needs to perform identity confirmation on the talents to be analyzed and verify talent resume information of the talents to be analyzed. Specifically, the identity information of the talents to be analyzed can be obtained, the talent resume information is subjected to identity verification through the identity information, and the identity information is information capable of uniquely proving the talents to be analyzed, including identity cards, information of a learning information network and the like. It should be noted that, the acquisition of the identity information of the talents to be analyzed is subject to the consent or authorization of the person. For example, the terminal may obtain an identification card of the talent to be analyzed, determine an identification card image, and perform identity verification according to the identification card image and a face image included in talent resume information. Or acquiring the learning information network learning inquiry code of the talents to be analyzed, and carrying out learning verification on the learning information included in the talent resume information according to the learning information network learning inquiry code so as to confirm that the information included in the talent resume information belongs to the talents to be analyzed. Or, the terminal may perform identity confirmation on the talent resume information in other manners, which is not limited herein.

S103, after the identity confirmation is passed, inputting talent resume information into an information extraction model to extract keywords, and determining talent data information, wherein the talent data information comprises identity information, learning history information, work history information, skill information and job hunting intention information, and the keyword extraction process comprises the following steps: extracting specific keywords based on an NLP technology, performing field matching on the extracted keywords and a pre-constructed vocabulary to obtain corresponding fields and field values, storing the fields and the field values through a data structure to obtain talent data information, and pre-training an information extraction model by inputting a plurality of talent resume information samples;

the terminal inputs the acquired talent resume information into a pre-trained information extraction model to extract keywords, and determines talent data information of the talents to be analyzed. The talent data information comprises identity information, learning information, work history information, skill information and job hunting intention information, wherein the identity information can comprise information such as photos, names, sexes, identification numbers, contact ways, mailbox account numbers and the like of talents to be analyzed, the learning information comprises information such as learning grades, reading institutions, professions and the like, the work history information comprises information such as work ages, companies who have employment, work posts, work history achievements and the like, the skill information comprises information such as special skills, proficiency and corresponding skill certificates, and the job hunting intention information comprises information such as intention enterprises, intention posts, intention payroll and intention profits, intention work places and the like.

Specifically, the keyword extraction process includes the steps of:

first, talent resume information is input into an information extraction model, keywords or phrases in the talent resume information are identified based on an NLP technology, and specific keywords are extracted. The specific keywords include keywords related to (or touching) information such as identity information, learning information, work history information, skill information, and job intention information. For example, the information including the photo, name, sex, identification number, contact information, mailbox account number, etc. of the talents to be analyzed in the above mentioned identity information may include information including the academic class, reading institution, specialty, etc. in the academic information, the work age, company who has employment, work post, work history achievement, etc. in the work history information, information including the features, proficiency, skill certificates, etc. in the skill information, and keywords corresponding to the information including the intention business, intention post, intention salary, welfare, intention work place, etc. in the intention information.

Then, field matching is carried out on the extracted keywords and fields contained in a vocabulary table pre-stored in a database, corresponding fields and field values are obtained, and the extracted keywords are filled into the vocabulary table. The fields in the vocabulary include the fields contained in the above mentioned information such as identity information, learning information, work history information, skill information and job intention information, and the values of the fields corresponding to the fields are extracted by using an NLP technology.

And finally, storing the fields and the field values in the vocabulary into a data structure, and determining and outputting talent data information according to the fields and the field values in the data structure.

The information extraction model comprises a structural network based on NLP technology and a data structure network based on a database, wherein the database can be an SQL database or other databases, and the information extraction model is obtained by inputting a plurality of talent resume information samples into an initial model for training.

S104, directionally storing talent data information to a talent information management database;

and the terminal directionally stores the talent data information output by the information extraction model into the talent information management database according to the storage requirement of the talent information management database. For example, the talent data information may be stored in a talent information management database in the form of a knowledge graph to improve the observability of the information, or the talent data information may be stored in a specific location of the talent information management database in a targeted manner, or may be stored in another manner, which is not limited herein. By directionally storing the talent data information of a plurality of talents in the talent information management database, the talents can be quickly and effectively determined and solicited according to the talent data information when the enterprise solicits talents.

S105, when the talent demand of the target enterprise is determined, talent demand intention information of the target enterprise is determined, and a plurality of talent data information is acquired from a talent information management database;

when talent demand exists in a target enterprise and talent solicitation is required, the terminal determines talent demand intention information of the target enterprise according to recruitment requirements of the target enterprise, wherein the talent demand intention information can comprise requirements corresponding to information such as identity information, learning information, work history information, skill information and job seeking intention information of a required talent, so that follow-up talent solicitation is carried out according to the requirements.

Then, the terminal acquires the talent data information corresponding to the plurality of previously stored talents from the talent information management database so as to perform talent matching according to the talent demand intention information and the talent data information.

S106, determining a first characteristic of talent data information and a second characteristic of talent demand intention information, calculating a coincidence value of the first characteristic in accordance with the second characteristic, determining a value of importance of the talent demand intention information on the second characteristic, and calculating the matching degree of the talent data information and the talent demand intention information by using a data matching algorithm based on the first characteristic, the second characteristic, the coincidence value and the value of importance, wherein the first characteristic comprises a first identity, a first academic, a first work history, a first skill and a first job intention, and the second characteristic comprises a second identity, a second academic, a second work history, a second skill and a second job intention;

The terminal determines five first characteristics of representing talents in talent data information, namely a first academic, a first working history, a first skill and a first job seeking intention, determines a second identity representing talents in talent demand intention information, a second academic, a second working history, a second skill and a second job seeking intention, calculates a conforming value conforming to the second characteristic in the first characteristic and determines a value of importance of talents demand intention information to the second characteristic. The coincidence value may be calculated according to a string similarity algorithm or a field matching determination may be made by inputting the first characteristic and the second characteristic into an associated vocabulary. The importance value represents the preference degree of the target enterprise to one of the second characteristics so as to distinguish the importance degree of the target enterprise to different characteristics.

And calculating the matching degree of the talent data information contained in the talent information management database and the talent demand intention information respectively according to a data matching algorithm. The data matching algorithm may use a similarity algorithm, such as a cosine similarity algorithm, a character string similarity algorithm, or a text similarity algorithm, or may use other data matching algorithms, which are not limited herein.

And S107, sorting the matching degrees from high to low, determining the matching level, and determining the target talents according to the matching level.

The terminal sequentially sorts the calculated talent data information of the plurality of talents and the matching degree of the enterprise from top to bottom, and determines the matching level corresponding to the plurality of talent data information, so that the target talents which are aimed to be solicited by the target enterprise are determined according to the matching level.

In the embodiment, the terminal acquires talent resume information of talents to be analyzed; carrying out identity confirmation on talent resume information based on the identity information of talents to be analyzed; after the identity is confirmed, inputting talent resume information into an information extraction model to extract keywords, and determining talent data information, wherein the talent data information comprises identity information, learning information, work history information, skill information and job hunting intention information; directionally storing talent data information to a talent information management database; when the talent demand of the target enterprise is determined, determining talent demand intention information of the target enterprise; the method comprises the steps of obtaining a plurality of talent data information from a talent information management database, calculating the matching degree of the plurality of talent data information and talent demand intention information respectively according to a data matching algorithm, sorting the matching degree from high to low, determining a matching level, and determining target talents according to the matching level, so that target talents corresponding to target enterprise talents are automatically determined, personnel information of job seekers does not need to be screened manually, labor cost is reduced, talent recommendation is accurately provided for enterprises, and user experience is improved.

In order to make the talent information analysis and management method based on big data provided by the application more obvious and understandable, the talent information analysis and management method based on big data provided by the application is described in detail below:

referring to fig. 2, fig. 2 is a schematic diagram of another embodiment of a talent information analysis and management method based on big data according to the present application, the method includes:

s201, acquiring talent resume information of talents to be analyzed;

step S201 in this embodiment is similar to step S101 in the embodiment shown in fig. 1, and detailed description thereof will be omitted here.

S202, confirming the identity of talent resume information based on the identity information of talents to be analyzed;

optionally, the terminal acquires identity image information of talents to be analyzed, and identity confirmation is carried out on talent resume information through a face discrimination mode. Specifically, a first face image of the talents to be analyzed is determined based on the identity information of the talents to be analyzed. The identity information can be an identity card, and the terminal acquires a first face image on the talent identity card to be analyzed.

And then, inputting the first face image and talent resume information into a pre-trained face discrimination model, and determining a second face image in the talent resume information according to the face discrimination model, wherein the talent resume information comprises the second face image. And then, extracting the first face contour feature of the first face image and the second face contour feature of the second face image through a face discrimination model, and extracting the first face five-sense organ point feature of the first face image and the second face five-sense organ point feature of the second face image. Further, the face discrimination model comprises a neural network model with a combined outline and five-sense-organ part thickness extraction, the convolution pool included by the neural network model is used for respectively carrying out rough extraction on the first face outline feature of the first face image and the second face outline feature of the second face image, and carrying out fine extraction on the first face five-sense-organ point feature of the first face image and the second face five-sense-organ point feature of the second face image.

Combining the first face contour feature and the first face five-sense organ point feature which are combined by the thickness extraction into a group of first feature vectors based on the full connection layer of the face discrimination model, and combining the second face contour feature and the second face five-sense organ point feature which are combined by the thickness extraction into a group of second feature vectors;

and finally, inputting the first feature vector and the second feature vector into a softmax classifier of a face discrimination model, outputting an identity discrimination result of talents to be analyzed, and carrying out identity confirmation on talent resume information according to the identity discrimination result. For example, the first feature vector and the second feature vector are input into a softmax classifier, if the identity discrimination result of the to-be-analyzed talents is that the first feature vector and the second feature vector are consistent, the identity of the talent resume information is confirmed, the talent resume information is output to be true, and if the first feature vector and the second feature vector are inconsistent, the talent resume information is output to be doubtful.

It should be noted that, the face discrimination model is obtained by inputting a plurality of face image samples and talent resume information samples for pre-training. Specifically, a face image sample set and a talent resume information sample set are constructed, wherein the face image sample set comprises a plurality of face image samples, the talent resume information sample set comprises a plurality of resume information samples, and the plurality of resume information samples comprise face images of talents of resume objects.

And constructing a neural network model based on face recognition and combining contour and thickness extraction of the five sense organs. The face image sample set and talent resume information sample set are input into a neural network model to conduct image feature extraction, face outline features and specific facial feature point features are determined according to the extracted image features, specifically, a large number of convolution pools are used for roughly extracting the face outline features of the image, the specific facial feature point features of the face image are finely extracted by using an acceptance-ResNet network, the face outline features and the specific facial feature point features are input into a full-connection layer of the network model to conduct feature combination, then the combined feature vectors are input into a softmax classifier to conduct discrimination training, a loss function and a gradient are calculated, model parameters of the neural network model and a sub-network model are optimized and updated by means of a back propagation algorithm according to the loss function value and the gradient, the model is iteratively trained until the loss function converges by means of the face image sample set and the talent resume information sample set, and the trained face discrimination model is obtained.

S203, after the identity confirmation is passed, inputting talent resume information into an information extraction model to extract keywords, and determining talent data information, wherein the talent data information comprises identity information, learning history information, work history information, skill information and job hunting intention information, and the keyword extraction process comprises the following steps: extracting specific keywords based on an NLP technology, performing field matching on the extracted keywords and a pre-constructed vocabulary to obtain corresponding fields and field values, storing the fields and the field values through a data structure to obtain talent data information, and pre-training an information extraction model by inputting a plurality of talent resume information samples;

Step S203 in this embodiment is similar to step S103 in the embodiment shown in fig. 1, and detailed description thereof will be omitted herein.

S204, directionally storing talent data information to a talent information management database;

optionally, the terminal directionally stores the talent data information output by the information extraction model into the talent information management database according to the storage requirement of the talent information management database. Specifically, the terminal can perform structuring processing on the talent data information, directionally store the talent data information in a json format into a talent information management database, and improve the visual effect of data storage.

Further, the terminal may set a talent information storage limit value to be currently stored in the talent information management database, the talent information storage limit value being used to distinguish the degree of talents' excellence. The terminal judges whether five characteristics of identity, academic, work history, skills and job hunting intention included in the talent data information reach the talent information storage limit value; if yes, determining that the talents reach the preset excellent degree, and storing talent data information into a first gradient talent database of a talent information management database, wherein the first gradient talent database is used for storing the talent data information of more excellent talents; if not, the talent data information is stored in a standby talent database of the talent information management database, and the talent data information is marked. For example, if the talent information storage limit is set as a local talent, and a graduate of a major college or a three-year working experience is set, when the personal data information of a certain talent reaches the talent information storage limit, the personal information data information of the talent is stored in a first talent base; if the talents are non-local people, only important college graduations and three-year working experiences in the personal data information of the talents accord with the talent information storage limit value, the talents are stored in a standby talent library, and the identity information of the non-local people is marked.

The talents are distinguished through the talent information storage limit value and are directionally stored in different talents libraries, so that different talents can be effectively distinguished, different talents are effectively managed, and the follow-up target enterprises can conveniently solicit the talents from the different talents libraries.

S205, positioning the talents data information in the talents information management database, and positioning the label of the talents data information to be in an incumbent state and/or a free state according to the current working state of the talents corresponding to the talents data information;

optionally, the terminal may update the talent data information in the talent information management library in real time, so as to improve timeliness of the talent data information. Specifically, the talent data information in the talent information management database is subjected to label positioning, the current talent employment state is mainly tracked, if the talents are employment, the talent data information corresponding to the talents is subjected to label positioning in the talent information management database, the labels are positioned in the incumbent state, and if the talents are not presently employment, the labels are positioned in the free state, so that a follow-up target enterprise can selectively solicit for the talents.

S206, when the talent demand of the target enterprise is determined, talent demand intention information of the target enterprise is determined, talent data information corresponding to the talents with the current labels positioned in a free state is determined from a talent information management database, and the talent data information is acquired;

Optionally, when the target enterprise has talent demand and needs talent solicitation, the terminal determines talent demand intention information of the target enterprise according to recruitment requirements of the target enterprise, where the talent demand intention information may include requirements corresponding to information such as identity information, learning information, work history information, skill information, job seeking intention information, and the like of a demanded talent, so that subsequent talent solicitation is performed according to the requirements.

Then, the talents are subjected to label positioning screening in a talent information management database, and talent data information corresponding to the talents with labels positioned in a free state is determined and acquired so as to solicit the talents.

S207, determining a first characteristic of the talent data information and a second characteristic of the talent demand intention information, calculating a coincidence value of the first characteristic in accordance with the second characteristic, determining a value of importance of the talent demand intention information on the second characteristic, and calculating the matching degree of the talent data information and the talent demand intention information by using a data matching algorithm based on the first characteristic, the second characteristic, the coincidence value and the value of importance, wherein the first characteristic comprises a first identity, a first academy, a first work history, a first skill and a first job intention, and the second characteristic comprises a second identity, a second academy, a second work history, a second skill and a second job intention;

The terminal respectively determines a first identity, a first academic, a first working history, a first skill and a first job-seeking intention of talents corresponding to talent data information, and sets five first characteristics of i= {1,2,3,4,5} respectively representing the first identity, the first academic, the first working history, the first skill and the first job-seeking intention; and determining a second identity, a second academic, a second working history, a second skill and a second job-seeking intention of the talent demand corresponding to the talent demand intention information, and setting j= {1,2,3,4,5} to respectively represent five second characteristics of the second identity, the second academic, the second working history, the second skill and the second job-seeking intention.

Then, a coincidence value is calculated in the first characteristic i, which coincides with the second characteristic j. Specifically, a characteristic association vocabulary is constructed, and the characteristic association vocabulary comprises a first characteristic i, a second characteristic j and a coincidence value; and detecting the first characteristic i and the second characteristic j by using the characteristic association vocabulary, if the first characteristic i accords with the second characteristic j, assigning a coincidence value to be 1, and if the first characteristic i does not accord with the second characteristic j, assigning a coincidence value to be 0.

Further, the characteristic association vocabulary comprises an association main vocabulary and an association sub-vocabulary, wherein the association main vocabulary is used for determining whether the first characteristic i accords with the second characteristic j, and the association sub-vocabulary is used for judging whether specific information contained in the first characteristic i accords with specific information required in the second characteristic j. For example, the associated primary vocabulary is shown in Table 1 and the associated sub-vocabulary is shown in Table 2. Then the coincidence value of the first characteristic i of the talent's personal data information and the second characteristic j of the talent's demand intention information can be determined according to tables 1 and 2.

Table 1:

table 2:

alternatively, the terminal may calculate the coincidence value of the first characteristic i, which is coincident with the second characteristic j, according to other calculation methods, which is not limited herein.

Then, the terminal determines the importance value of the talent demand intention information on the second characteristic j, wherein the importance value represents the preference degree, that is, when the target enterprise views or prefers one of the second characteristics, different weight assignments can be performed on the characteristic, so as to represent the importance value of the characteristic, as shown in the following table 3.

Table 3:

/>

specifically, the value of importance can be set according to the intention requirement of the target enterprise, or can be determined by calculating the preference degree of the current market for each characteristic of talents according to big data, and the preference degree is not limited herein.

Finally, calculating the matching degree of talent data information and talent demand intention information based on a data matching algorithm, wherein the data matching algorithm is as follows:

wherein P represents the matching degree, u represents talent data information, v represents talent demand intention information of a target enterprise,weight value representing the corresponding weighting of the second property j, < >>A coincidence value of a first characteristic i representing talent data information u to a second characteristic j,/-, is->A value representing the importance of talent demand intent information v to the second characteristic j;

wherein N represents the total number of talents whose first job-seeking intent is the second job-seeking intent in the talent information management database,of talents whose first job-seeking intention is second job-seeking intention, whether the first characteristic i of the a-th talent satisfies the j-th characteristic of the second characteristics, if so, pair->And assigning a preset value, wherein the preset value can be 1 or other values, the preset value is not limited herein, and if the preset value is not satisfied, the preset value is assigned 0.

S208, sorting the matching degrees from high to low, determining the matching level, screening talents with the matching level higher than a preset threshold, determining the talents as target talents, and directionally pushing talent data information of the target talents to recruitment ends of target enterprises.

Optionally, the terminal sequentially sorts the calculated matching degree of the talent data information of the plurality of talents and the enterprise from top to bottom, and determines matching levels corresponding to the plurality of talent data information, wherein the matching levels can be classified into a low matching level, a medium matching level and a high matching level. The terminal screens out the target talents corresponding to the talent data information with the high matching level, and directionally pushes the talent data information of the target talents to the recruitment end of the target enterprise, so that recruiters of the target enterprise can solicit target talents meeting talent demand intention.

In the embodiment, the terminal acquires talent resume information of talents to be analyzed; carrying out identity confirmation on talent resume information based on the identity information of talents to be analyzed; after the identity is confirmed, inputting talent resume information into an information extraction model to extract keywords, and determining talent data information, wherein the talent data information comprises identity information, learning information, work history information, skill information and job hunting intention information; and directionally storing talent data information to a talent information management database to effectively manage talents.

Meanwhile, when the talent demand of the target enterprise is determined, talent demand intention information of the target enterprise is determined; the method comprises the steps of obtaining a plurality of talent data information from a talent information management database, calculating the matching degree of the plurality of talent data information and talent demand intention information respectively according to a data matching algorithm, sorting the matching degree from high to low, determining a matching level, and determining target talents according to the matching level, so that target talents corresponding to target enterprise talents are automatically determined, personnel information of job seekers does not need to be screened manually, labor cost is reduced, talent recommendation is accurately provided for enterprises, and user experience is improved.

The above description is given to the talent information analysis and management method based on big data provided by the application, and the following description is given to the talent information analysis and management system based on big data provided by the application:

referring to fig. 3, fig. 3 is a schematic diagram illustrating an embodiment of a talent information analysis management system based on big data according to the present application, the system includes:

an acquiring unit 301, configured to acquire talent resume information of a talent to be analyzed;

a confirmation unit 302, configured to confirm the identity of the talent resume information based on the identity information of the talents to be analyzed;

the first determining unit 303 is configured to input the talent resume information into an information extraction model to perform keyword extraction after the identity confirmation is passed, and determine talent data information, where the talent data information includes identity information, learning information, work history information, skill information, and job hunting intention information, and the keyword extraction process includes:

A storage unit 304, configured to directionally store the talent data information to a talent information management database;

a second determining unit 305, configured to determine talent demand intention information of a target enterprise when it is determined that the target enterprise has talent demand, and acquire a plurality of talent data information from the talent information management database;

a matching unit 306, configured to determine a first characteristic of the talent data information and a second characteristic of the talent demand intention information, calculate a compliance value of the first characteristic to the second characteristic, and determine a value of importance of the talent demand intention information to the second characteristic, and calculate a matching degree of the talent data information and the talent demand intention information using a data matching algorithm based on the first characteristic, the second characteristic, the compliance value, and the importance value, where the first characteristic includes a first identity, a first academic, a first working history, a first skill, and a first intention, and the second characteristic includes a second identity, a second academic, a second working history, a second skill, and a second intention;

and a third determining unit 307, configured to determine a matching level by sorting the matching degrees from high to low, and determine a target talent according to the matching level.

In the system of this embodiment, the functions executed by each unit correspond to the steps in the foregoing embodiment of the method shown in fig. 1, and are not described herein in detail.

In the embodiment, the system acquires talent resume information of talents to be analyzed; carrying out identity confirmation on talent resume information based on the identity information of talents to be analyzed; after the identity is confirmed, inputting talent resume information into an information extraction model to extract keywords, and determining talent data information, wherein the talent data information comprises identity information, learning information, work history information, skill information and job hunting intention information; the talent data information is directionally stored in a talent information management database, and when the talent demand of the target enterprise is determined, talent demand intention information of the target enterprise is determined; the method comprises the steps of obtaining a plurality of talent data information from a talent information management database, calculating the matching degree of the plurality of talent data information and talent demand intention information respectively according to a data matching algorithm, sorting the matching degree from high to low, determining a matching level, and determining target talents according to the matching level, so that target talents corresponding to target enterprise talents are automatically determined, personnel information of job seekers does not need to be screened manually, labor cost is reduced, talent recommendation is accurately provided for enterprises, and user experience is improved.

Referring to fig. 4, fig. 4 is a schematic diagram of another embodiment of a big data-based talent information analysis management system according to the present application, where the system includes:

an acquiring unit 401, configured to acquire talent resume information of talents to be analyzed;

a confirmation unit 402, configured to confirm the identity of the talent resume information based on the identity information of the talents to be analyzed;

a first determining unit 403, configured to input the talent resume information into an information extraction model to perform keyword extraction after the identity confirmation is passed, and determine talent data information, where the talent data information includes identity information, learning information, work history information, skill information, and job hunting intention information, and the keyword extraction process includes:

A storage unit 404, configured to directionally store the talent data information to a talent information management database;

a second determining unit 406, configured to determine talent demand intention information of a target enterprise when it is determined that the target enterprise has talent demand, and acquire a plurality of talent data information from the talent information management database;

a matching unit 407, configured to determine a first characteristic of the talent data information and a second characteristic of the talent demand intention information, calculate a coincidence value of the first characteristic with the second characteristic, and determine a value of importance of the talent demand intention information to the second characteristic, and calculate a matching degree of the talent data information and the talent demand intention information using a data matching algorithm based on the first characteristic, the second characteristic, the coincidence value, and the importance value, where the first characteristic includes a first identity, a first academic, a first working history, a first skill, and a first intention, and the second characteristic includes a second identity, a second academic, a second working history, a second skill, and a second intention;

and a third determining unit 408, configured to determine a matching level by sorting the matching degrees from high to low, and determine a target talent according to the matching level.

Optionally, the confirmation unit 402 is specifically configured to determine a first face image of the talents to be analyzed based on the identity information of the talents to be analyzed;

Optionally, the matching unit 407 is specifically configured to determine a first identity, a first academic, a first work history, a first skill, and a first job-seeking intent of the talent corresponding to the talent data information, and set five first characteristics that i= {1,2,3,4,5} respectively represent the first identity, the first academic, the first work history, the first skill, and the first job-seeking intent;

wherein P represents the matching degree, u represents the talent data information, v represents the talent demand intention information of the target enterprise, Weight value representing the corresponding weighting of said second property j +.>A coincidence value of a first characteristic i representing said talent data information u to said second characteristic j,/or->A value representing the importance of the talent demand intent information v to the second characteristic j;

wherein N represents the total number of talents for which the first job hunting intention is the second job hunting intention in the talent information management database,of talents whose first intention is the second intention, whether the first characteristic i of the a-th talent satisfies the first criterionThe j-th characteristic of the two characteristics, if satisfied, is to the +.>And assigning a preset value, and if the preset value is not met, assigning 0.

Optionally, the matching unit 407 is specifically configured to construct a feature-related vocabulary, where the feature-related vocabulary includes the first feature i, the second feature j, and a coincidence value;

Optionally, the talent information analysis management system further includes:

The tag positioning unit 405 is configured to perform tag positioning on talent data information in the talent information management database, and position a tag thereof to an incumbent state and/or a free state according to a current working state of a talent corresponding to the talent data information;

the second determining unit 406 is specifically configured to determine talent data information corresponding to a talent whose current tag is located in a free state from the talent information management database, and obtain the talent data information.

Optionally, the third determining unit 408 is specifically configured to screen talents with matching levels higher than a preset threshold, determine the talents as target talents, and directionally push talent data information of the target talents to a recruitment end of the target enterprise.

Optionally, the storage unit 404 is specifically configured to perform a structuring process on the talent data information, and store the talent data information in a json format in a talent information management database.

Optionally, the storage unit 404 is specifically configured to set a talent information storage limit value to be currently stored in a talent information management database;

In the system of this embodiment, the functions executed by each unit correspond to the steps in the method embodiment shown in fig. 2, and are not described herein again.

In the embodiment, the system acquires talent resume information of talents to be analyzed; carrying out identity confirmation on talent resume information based on the identity information of talents to be analyzed; after the identity is confirmed, inputting talent resume information into an information extraction model to extract keywords, and determining talent data information, wherein the talent data information comprises identity information, learning information, work history information, skill information and job hunting intention information; and directionally storing talent data information to a talent information management database to effectively manage talents.

The application also provides a talent information analysis and management device based on big data, referring to fig. 5, fig. 5 is an embodiment of the talent information analysis and management device based on big data, which comprises:

a processor 501, a memory 502, an input/output unit 503, and a bus 504;

the processor 501 is connected to the memory 502, the input/output unit 503, and the bus 504;

the memory 502 holds a program, and the processor 501 calls the program to execute any of the talent information analysis management methods based on big data as described above.

The present application also relates to a computer-readable storage medium having a program stored thereon, which when run on a computer causes the computer to execute any of the above big data-based talent information analysis management methods.

It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, which are not repeated herein.

In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of elements is merely a logical functional division, and there may be additional divisions of actual implementation, e.g., multiple elements or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.

The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed over a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.

The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be embodied in essence or a part contributing to the prior art or all or part of the technical solution in the form of a software product stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a read-only memory (ROM), a random access memory (RAM, random access memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.

Claims

1. The talent information analysis and management method based on big data is characterized by comprising the following steps of:

acquiring talent resume information of talents to be analyzed;

2. The talent information analysis management method according to claim 1, wherein said identifying the talent resume information based on the identity information of the talents to be analyzed comprises:

3. The talent information analysis management method according to claim 1, wherein said determining a first characteristic of the talent data information and a second characteristic of the talent demand intention information, calculating a compliance value of the first characteristic with the second characteristic, and determining a value of importance of the talent demand intention information to the second characteristic, and calculating a degree of matching of the talent data information and the talent demand intention information using a data matching algorithm based on the first characteristic, the second characteristic, the compliance value, and the value of importance comprises:

wherein N represents the total number of talents for which the first job hunting intention is the second job hunting intention in the talent information management database,for first job-seeking intent to second job-seeking intentIn talents, whether the first characteristic i of the a-th talent meets the j-th characteristic in the second characteristics, if so, the first characteristic i is equal to the j-th characteristic in the first characteristic, and if so, the first characteristic i is equal to the j-th characteristic in the second characteristic>And assigning a preset value, and if the preset value is not met, assigning 0.

4. The talent information analysis management method according to claim 3, wherein said calculating a coincidence value of the first characteristic i in coincidence with the second characteristic j comprises:

5. The talent information analysis management method according to claim 1, wherein after said directionally storing the talent data information to a talent information management database, the talent information analysis management method further comprises:

6. The talent information analysis management method according to claim 1, wherein said determining a target talent according to the matching level comprises:

7. The talent information analysis management method according to any one of claims 1 to 6, wherein said directionally storing the talent data information to a talent information management database includes:

8. The talent information analysis management method according to any one of claims 1 to 6, wherein said directionally storing the talent data information to a talent information management database includes:

9. The talent information analysis and management system based on big data is characterized by comprising:

10. A talent information analysis and management device based on big data, characterized in that the talent information analysis and management device includes:

a processor, a memory, an input-output unit, and a bus;

the processor is connected with the memory, the input/output unit and the bus;

the memory holds a program, and the processor calls the program to execute the talent information analysis management method according to any one of claims 1 to 8.