CN115526589A - Employment information processing method and equipment for people to be cared - Google Patents
Employment information processing method and equipment for people to be cared Download PDFInfo
- Publication number
- CN115526589A CN115526589A CN202211128537.7A CN202211128537A CN115526589A CN 115526589 A CN115526589 A CN 115526589A CN 202211128537 A CN202211128537 A CN 202211128537A CN 115526589 A CN115526589 A CN 115526589A
- Authority
- CN
- China
- Prior art keywords
- information
- post
- evaluation score
- determining
- text
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 230000010365 information processing Effects 0.000 title claims abstract description 17
- 238000003672 processing method Methods 0.000 title claims abstract description 11
- 238000011156 evaluation Methods 0.000 claims abstract description 132
- 238000000034 method Methods 0.000 claims abstract description 39
- 239000013598 vector Substances 0.000 claims description 79
- 239000011159 matrix material Substances 0.000 claims description 43
- 230000007115 recruitment Effects 0.000 claims description 23
- 238000012549 training Methods 0.000 claims description 14
- 238000000605 extraction Methods 0.000 claims description 11
- 230000006870 function Effects 0.000 claims description 11
- 230000002457 bidirectional effect Effects 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 6
- 238000012216 screening Methods 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000000354 decomposition reaction Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 230000036541 health Effects 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 241001442234 Cosa Species 0.000 description 1
- 241000238413 Octopus Species 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000019771 cognition Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 238000013441 quality evaluation Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/105—Human resources
- G06Q10/1053—Employment or hiring
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Game Theory and Decision Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application provides a employment information processing method and equipment for people to be cared for, wherein the method acquires a plurality of post information; the position information at least includes: a post name, a post responsibility text, a post requirement text. And determining a first evaluation score corresponding to each post information based on the recruiter information corresponding to each post information and the resume information of the currently pending person. And the first evaluation score represents the matching degree of the currently undetermined person and each post information. And determining a second evaluation score corresponding to the undetermined person corresponding to the first evaluation score according to the first evaluation score and the preset implicit semantic model. And determining the corresponding matching probability values of the resume information and the post information through a pre-trained deep forest model. And determining at least one post information matched with the currently undetermined person based on the second evaluation score and the matching probability value so as to send the post information to the corresponding user terminal.
Description
Technical Field
The application relates to the technical field of internet, in particular to a employment information processing method and equipment for people waiting for employment.
Background
With the development of science and technology and the progress of society, for enterprises, the working posts need to be equipped with proper talents to improve the industrial benefit. Even under the background of the era of rapid development of big data technology, the matching between enterprises and talents has information disjointed, for example, the enterprises cannot really know the talents' abilities, the talents cannot accurately obtain the requirements of the enterprise posts, and whether the enterprises are matched with the abilities of the talents is determined in time. Under the conditions that the current recruitment means is limited, graduates increase dramatically year by year and the rate of leaving jobs of young employees is high, enterprises only adopt to recruit talents continuously by the traditional means, and the time and the labor are consumed.
In addition, among the people waiting for employment, there are people waiting for employment who have working experience and technical ability, and the people in this part are difficult to find out proper work in a short time by depending on the cognition of the people on the post and the application means mastered by the people. Moreover, at present, the recruitment information sent by different enterprises is difficult to understand, and the personnel waiting for employment are difficult to accurately deliver resumes to proper positions, so that the normal employment and long-term professional planning of the personnel waiting for employment are influenced.
And the resume of the job-to-be-cared personnel is manually identified and screened, and then the recruitment requirements of the enterprise are matched, so that the work is complex, the workload is large, and the labor cost is wasted. Therefore, there is a need for a method for processing, identifying and screening employment information of employment personnel for human reference to solve most of the labor cost.
Disclosure of Invention
The embodiment of the application provides a employment information processing method and equipment for people waiting for employment, and solves the problems that employment information is mainly processed, identified and screened by manpower, the work is complex, the workload is large, and the labor cost is wasted.
In one aspect, an embodiment of the present application provides a employment information processing method for people to be cared for, where the method includes:
and acquiring a plurality of position information. The position information at least includes: a post name, a post responsibility text, a post requirement text. And determining a first evaluation score corresponding to each post information based on the recruiter information corresponding to each post information and the resume information of the currently pending person. And the first evaluation score represents the matching degree of the currently undetermined person and each post information. And determining a second evaluation score corresponding to the undetermined person corresponding to the first evaluation score according to the first evaluation score and the preset implicit semantic model. And determining the corresponding matching probability value of the resume information and each post information through a pre-trained deep forest model. And determining at least one post information matched with the currently undetermined person based on the second evaluation score and the matching probability value, so as to send the post information to a corresponding user terminal. Wherein, the corresponding text is an evaluation text for the post. And generating a similarity matrix of the currently undetermined personnel and the recruiters according to the cosine similarity. And determining the corresponding post score of the currently undetermined person as a first evaluation score based on the evaluation text of each post by each recruiter and the corresponding cosine similarity in the similarity matrix.
In an implementation manner of the application, resume information of the currently undetermined person and an evaluation text corresponding to the first evaluation score are input into a latent semantic model, so that the latent semantic model determines a scoring matrix corresponding to the currently undetermined person and a recruiter corresponding to the first evaluation score. Factorization processing is carried out on the scoring matrix through a latent semantic model and a preset loss function until the preset loss function meets a preset condition, so that the scoring matrix is decomposed into a user scoring matrix and a post weight matrix. And determining the predictive evaluation score of the currently undetermined person to each post based on the user scoring matrix and the post weight matrix. The weighted average of the first rating score and the predicted rating score is taken as the second rating score.
In one implementation of the present application, a number of resume samples and a number of position information samples are obtained. And respectively carrying out short text coding processing on each resume sample and each post information sample through a deep forest model to determine a first feature vector set corresponding to each resume sample and a second feature vector set corresponding to each post information sample, calculating the matching probability value of each feature vector in the first feature vector set and each feature vector in the second feature vector set until the matching probability value meets a preset sample probability table, and determining that the deep forest model is trained.
In one implementation manner of the application, the second evaluation score larger than the first preset threshold is determined as a score of a pending position. And determining the information of the position to be determined corresponding to the value of the position to be determined. And determining the corresponding matching probability value of the information of the position to be positioned. And determining a third evaluation score corresponding to each to-be-positioned information according to the matching probability value corresponding to the to-be-positioned information, the to-be-positioned score corresponding to the to-be-positioned information and the preset matching weight. And taking the position information corresponding to the third evaluation score which is larger than the second preset threshold value as the position information matched with the currently undetermined person.
In one implementation of the present application, a data dictionary of post information and corresponding resume information is generated. The data dictionary includes a number of triples of text. The triplet text includes: and the employment ability text and the post name correspond to the personnel name and the resume information. And respectively encoding each triple text into vector groups through a preset TransH model, projecting each vector group to a corresponding relation hyperplane, and calculating the recommended value of each triple text in the data dictionary according to a preset formula and the projection vector of each vector group in the relation hyperplane. Wherein, the vector group comprises a personnel name vector, a employment ability text vector and a position name vector; the relation hyperplane is the hyperplane corresponding to the employment ability text vector in the vector group. And rejecting the triple texts with the recommended values smaller than the preset value, and updating the data dictionary to determine the position information according to the updated data dictionary.
In one implementation of the application, a plurality of recruitment information and post keywords corresponding to the recruitment information are obtained. And sequentially combining the post keywords into post key sentences according to the position sequence of the post keywords in the recruitment information, inputting the post key sentences into a bidirectional LSTM-CRF model as training samples, and training the bidirectional LSTM-CRF model.
In one implementation manner of the application, a plurality of post keywords in the recruitment information are screened through a bidirectional LSTM-CRF model, and the screened post keywords are used as post responsibility sample words. And determining post demand sample words corresponding to the post responsibility texts in the recruitment information. And inputting the post responsibility sample words and the post demand sample words into a keyword extraction model for training so as to obtain post information through the trained keyword extraction model. The keyword extraction model is a TextRank algorithm model.
In one implementation manner of the present application, feedback information of a user terminal is obtained. And determining the accuracy and the recall rate of the post information according to the feedback information. And determining whether the post information is the post information to be updated or not according to the accuracy and the recall rate. And if so, eliminating the recruiter information corresponding to the post information to be updated, and updating the first evaluation score and the second evaluation score according to the updated recruiter information so as to update the post information to be updated.
On the other hand, the embodiment of the present application further provides an employment information processing apparatus for people to be cared for, the apparatus including:
at least one processor; and a memory communicatively coupled to the at least one processor. Wherein the memory stores instructions executable by the at least one processor to cause the at least one processor to:
and acquiring a plurality of position information. The position information at least includes: a post name, a post responsibility text, and a post requirement text. And determining a first evaluation score corresponding to each post information based on the recruiter information corresponding to each post information and the resume information of the currently pending person. And the first evaluation score represents the matching degree of the currently undetermined person and each post information. And determining a second evaluation score corresponding to the undetermined person corresponding to the first evaluation score according to the first evaluation score and the preset implicit semantic model. And determining the corresponding matching probability values of the resume information and the post information through a pre-trained deep forest model. And determining at least one post information matched with the currently undetermined person based on the second evaluation score and the matching probability value so as to send the post information to the corresponding user terminal.
According to the technical scheme, the first evaluation scores corresponding to the position information of the pending person to be cared and the historical recruiters are determined, and then the position information matched with the pending person is obtained through the latent semantic model and the deep forest model, so that the employment information is accurately recommended. And employment information can be processed, identified and screened for people to refer to so as to solve most of labor cost.
According to the application, the proper position can be quickly found by utilizing the working experience and the technical capability mastered by the personnel to be cared, so that the resume delivery of the personnel to be cared is not delivered blindly, and the normal employment of the personnel to be cared and the long-distance professional planning are ensured. In addition, the method and the system realize convenient and accurate employment information processing by combining various algorithms and meet the post information recommendation of the characteristics of the personnel to be cared.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
FIG. 1 is a schematic flow chart of a employment information processing method for a group of persons to be cared for in the embodiment of the present application;
fig. 2 is a schematic structural diagram of a employment information processing apparatus for a group of persons to be cared for in an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be described in detail and completely with reference to the following specific embodiments of the present application and the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
At present, the complicated and various post information makes a lot of job hunting personnel difficult to find the post information suitable for the personnel. The group to be cared for relates to college students, personnel to be cared for unemployment, young unemployment personnel, elder employment group, retired soldier and the like. The employment people are large, the employment situation is severe, the staff change greatly under the influence of social factors, the accuracy of the traditional recruitment information recruitment mode is poor in subjectivity and accuracy of post matching, and the description of each enterprise on the same post is different and the positioning is not clear. Therefore, it is difficult for the people waiting for employment to quickly find the proper position. The recruitment post is difficult to quickly screen out proper post response personnel among a large number of people waiting for employment.
Based on this, the embodiment of the application provides a employment information processing method and equipment for people waiting for employment, which are used for solving the problems that the employment information is processed, identified and screened mainly by manpower, the work is complex, the workload is large, and the labor cost is wasted, and the current personnel waiting for employment cannot quickly find out appropriate work according to the work experience and the technical capability which are mastered by the current personnel, so that blind delivery exists in resume delivery, and the normal employment and long-term professional planning of the personnel waiting for employment are influenced.
Various embodiments of the present application are described in detail below with reference to the accompanying drawings.
The embodiment of the application provides a employment information processing method for a group to be cared for, as shown in fig. 1, the method may include steps S101-S105:
s101, the server obtains a plurality of position information.
The position information at least comprises: a post name, a post responsibility text, and a post requirement text.
In the embodiment of the present application, the station information may include a station name, a station responsibility text, a station requirement text, for example, an attendant, a station responsibility text: welcome/deliver/serve/hygiene; post requirement text: health/good appearance/quality/18-30 years/starred hotel/kendyji, mcdonald's duty, must win guest/health card/shift system.
It should be noted that the server is only exemplary and the execution subject of the employment information processing method for the group of persons to be cared for exists, and the execution subject is not limited to the server, and the present application is not particularly limited thereto.
In this embodiment of the present application, before the server obtains the position information, the method further includes:
the server acquires the recruitment information and the post keywords corresponding to the recruitment information.
The server can acquire the recruitment related text by using a crawler program or a octopus and a hough collector, and perform data governance on the related text, for example: removing br, div, span and other special character data; removing repeated data (one-job polysemy condition), missing data and supplementing incomplete data; and (4) managing the place where the work description field is inconsistent with other fields of the post, removing misleading post description and eliminating ambiguous data. Translating the English data into Chinese data.
And then, the server sequentially combines the post keywords into post key sentences according to the position sequence of the post keywords in the recruitment information, and inputs the post key sentences into a bidirectional LSTM-CRF model as training samples so as to train the bidirectional LSTM-CRF model.
Specifically, a post description related text is segmented according to punctuation marks, special characters pattern = | post description | post requirements are filtered regularly, then a long sentence is segmented into a short sentence, then an entity dictionary and word boundaries are obtained, and words at each position in the short sentence are converted into word vectors. Combining a plurality of word vectors to obtain a short sentence as the input of the model, and providing the short sentence for the model to learn
The server specifically acquires the post information and comprises the following steps:
firstly, the server screens a plurality of post keywords in the recruitment information through a bidirectional LSTM-CRF model, and the screened post keywords are used as post responsibility sample words.
For example, the recruitment information includes "cases submitted by law, legal consultation, complaint, evidence arrangement, court complaints, internal training of regular organizations of law, and courses of related activities, forums, training and the like held by external industry associations and institutions together", so that post responsibility sample words can be obtained through the bidirectional LSTM-CRF model: case law consultation drafting and appetizing participation training … … of handling law
And then, the server determines post requirement sample words corresponding to the post responsibility texts in the recruitment information.
And the post requirement sample word generation uses an automatic summary generation technology. The automated Generation of the position demand is described as given a sample set of position demand words B with | B | different positions, i.e.
Wherein, X i Is a station role which describes the role of the ith station, Y i Is a station requirement that describes the various capability requirements of this station. In particular, X is responsibility for each station i Provided that it contains H d A word, i.e.
A position demand typically contains multiple sentences to describe different capability requirements, so each position demand Y will be i Is shown as
Y i ＝{y 1 ，y 2 ，...y N }。
Wherein, y i Is the jth sentence. For example, 5 post requirement sentences are given, i.e., N =5, which corresponds to the introduction of professional colleges, academic calendars, work experiences, communication abilities, and thinking abilities. Further assume that each y i Included
A word, i.e.
For example, the station responsibility sample word in the above example may correspond to the following station demand sample word: law qualification certificate in Law, above this department.
Then, the server inputs the post responsibility sample words and the post demand sample words into a keyword extraction model for training, so as to obtain post information through the trained keyword extraction model. The keyword extraction model is a TextRank algorithm model.
The server can perform voting mechanism sequencing on the post responsibility sample words and the post demand sample words, the sequenced sentences are post information samples, training of the keyword extraction model is further completed, and post information is extracted through the trained keyword extraction model.
The post information is generated by the short text description identification mode, so that the situation that the capability and the post of employment personnel cannot be normally matched due to the description difference of enterprises on different posts can be avoided. And the text is processed, so that the calculation amount of the server can be saved, and the employment information processing time is shortened.
And S102, the server determines a first evaluation score corresponding to each post information based on the recruiter information corresponding to each post information and the resume information of the currently undetermined person.
And the first evaluation score represents the matching degree of the currently undetermined person and each post information.
In the embodiment of the application, the server determines a first evaluation score corresponding to each post information based on the recruiter information corresponding to each post information and the resume information of the currently pending person, and specifically includes:
firstly, the server determines the cosine similarity of a text corresponding to the resume information and a text corresponding to each recruiter information according to the resume information of the currently undetermined person and the information of each recruiter.
Wherein, the corresponding text is an evaluation text for the post.
The resume information of the to-be-determined personnel can be resume related information uploaded by the to-be-determined personnel, and the server determines cosine similarity between the information of the plurality of recruiters and the resume information. The server can determine the similarity between the resume of the to-be-determined person and the information of each recruiter through a collaborative filtering recommendation algorithm.
Specifically, the server respectively vector-encodes the resume information, the post evaluation text and the recruiter evaluation text of each post in the recruiter information, for example: the work overtime is more, and the work overtime is coded into two vector work overtime [1,0,0], and more [2,0,0]; the work competition pressure is large, and the work competition is coded as work competition [1,2,0], the pressure is large [1,2,1] … …, and then the cosine similarity of the corresponding text is calculated through the following cosine similarity calculation companies:
wherein cosA is the cosine similarity between the evaluation text J corresponding to the resume information and the evaluation text Z corresponding to the recruiter information, and J is x To evaluate the xth word vector, Z, of text J x To evaluate the xth word vector of text Z, y is the total number of word vectors.
And then, the server generates a similarity matrix of the currently undetermined person and each recruiter according to the cosine similarity.
The server may establish a similarity matrix, thereby simplifying the amount of computation in calculating the evaluation score. Specifically, after the server obtains the cosine similarity of the evaluation text of the currently undetermined person to the evaluation text of the same post by the recruiter according to the resume information of the currently undetermined person, the currently undetermined person and the recruiter can be used as the rows of the matrix, the post can be used as the column of the matrix, and the cosine similarity matrix of the evaluation text of the currently undetermined person and the recruiter to each post is obtained. The recruiters in the similarity matrix can be the recruiters which are obtained after the recruiters with the cosine similarity with the currently undetermined person smaller than a preset value are removed, for example, the preset value is a, the information of the recruiters with the cosine similarity smaller than a is removed, and the recruiters which are closer to the currently undetermined person and have similar preference are reserved. According to the method and the device, the cosine similarity between all recruiters and the currently undetermined person can be written in the similarity matrix, and the information of the recruiters with the cosine similarity smaller than a can be removed before the first evaluation score is generated subsequently.
And then, the server determines the corresponding post score of the currently undetermined person as a first evaluation score based on the evaluation text of each post by each recruiter and the corresponding cosine similarity in the similarity matrix.
The server can determine the evaluation score corresponding to the evaluation text according to the evaluation text of each post by each recruiter, and the server can determine the evaluation score corresponding to the evaluation text by using a preset model or a preset text comparison table (text and evaluation score comparison table). The server determines a similarity matrix generated by each historical recruiter and the pending person, and then calculates the similarity between the current pending person and the historical recruiter when evaluating each post according to the following formula.
Wherein A is 1 For similarity of the pending person's evaluation of the third post with the first historical recruiter, a 1 Score of evaluation of the first position for the person to be determined, a 2 Score of the second post for the person in question, b 1 A valuation score for the first historical recruiter to the first position, b 2 A score for the first historical recruiter's appraisal of the second position.
The first appraisal score is calculated according to the following formula, taking three historical recruiters as an example:
wherein, T 3 A first rating score for the pending person for the third post,
for the similarity of the pending person's evaluation of the third post with the second historical recruiter,
similarity of evaluation of the third post for the pending person and the third historical recruiter, b 3 A rating score for the first historical recruiter for the third post, c 3 A rating score for the second historical recruiter for the third post, d 3 A score for the third historical recruiter's appraisal of the third post.
And after the first evaluation scores of the undetermined persons on each post are obtained through collaborative filtering recommendation, if a certain difference still exists between the recommended post and an expected result, the server further processes employment information through the following method.
S103, the server determines a second evaluation score corresponding to the undetermined person corresponding to the first evaluation score according to the first evaluation score and the preset implicit semantic model.
In this embodiment of the application, the server determines, according to the first evaluation score and the preset latent semantic model, a second evaluation score corresponding to the undetermined person corresponding to the first evaluation score, and specifically includes:
firstly, inputting resume information of the currently undetermined person and an evaluation text corresponding to a first evaluation score into a latent semantic model by the server, so that the latent semantic model determines a grading matrix corresponding to the currently undetermined person and a recruiter corresponding to the first evaluation score.
The server can input a latent semantic model into the evaluation text corresponding to the resume information and the first evaluation score for the scores of the posts, namely the evaluation text of the undetermined person corresponding to the resume information for each post, and the evaluation text can be generated through the resume information and the obtained first evaluation score. For example, when there is no evaluation text for the X position pair in the resume information of the person to be determined, the first evaluation score may be 5 for the X position, the corresponding history recruiter m may be 5 for the X position, and the evaluation text for m may be used as the evaluation text for the person to be determined. And then the evaluation text is processed through the latent semantic model, and the scores of the to-be-determined personnel on all the posts can be obtained through the pre-trained latent semantic model.
The hidden semantic model can determine the interest degree of the person to be determined in the hidden class of the post and the hidden class weight of the post, and the target function of the hidden semantic model is continuously optimized through a loss function constructed by the penalty factor of the hidden semantic model, so that a high-quality evaluation score output by the hidden semantic model is obtained.
And then, the server carries out factorization processing on the scoring matrix through the hidden semantic model and a preset loss function until the preset loss function meets a preset condition, so that the scoring matrix is decomposed into a user scoring matrix and a post weight matrix.
After the scoring matrix is obtained, the hidden semantic model decomposes the scoring matrix into a user scoring matrix and a post weight matrix according to the interest degree of the undetermined person in the post hidden class and the hidden class weight of the post. In order to ensure the minimum error in the decomposition, the decomposition is carried out by setting a loss function, for example, the minimization is carried out by a random gradient descent method, the loss function is solved, the scoring matrix is decomposed, and the completion of the decomposition is determined under the condition that the loss function is smaller than a set value.
And then, the server determines the prediction evaluation score of the currently undetermined person on each post based on the user scoring matrix and the post weight matrix.
The server can calculate the product by using the user scoring matrix and the post weight matrix, and calculate the predicted evaluation score by the following formula:
wherein r is ui A predictive appraisal score, p, for the recruit post i for the person u uk Scoring the interest degree user of the kth hidden class of the position for the person u, wherein the kth hidden class corresponds to a user scoring matrix; q. q.s ki The weight is scored for the impact of the kth hidden class on position i, which corresponds to the position weight matrix.
The server then takes the weighted average of the first rating score and the predicted rating score as the second rating score.
After the server obtains the predicted evaluation score output by the latent semantic model, a weighted average value is obtained by the first evaluation score and the corresponding predicted evaluation score, and the weighted average value is used as a second evaluation score. The weights corresponding to the first evaluation score and the predicted evaluation score in the weighted average may be set by the user, which is not specifically limited in this application. The hidden behaviors of the user are introduced into the evaluation consideration of the posts, so that a more accurate recommendation result can be obtained when the posts are recommended.
And S104, the server determines the corresponding matching probability value of the resume information and each post information through a pre-trained deep forest model.
In this embodiment of the application, before the server determines the matching probability values corresponding to the resume information and the post information through the pre-trained deep forest model, the method further includes:
the server obtains a plurality of resume samples and a plurality of position information samples. And then, the server performs short text coding processing on each resume sample and each post information sample through a deep forest model respectively to determine a first feature vector set corresponding to each resume sample and a second feature vector set corresponding to each post information sample, calculates the matching probability value of each feature vector in the first feature vector set and each feature vector in the second feature vector set until the matching probability value meets a preset sample probability table, and determines that the deep forest model training is completed.
The server needs to train the deep forest model to obtain the deep forest model fused with the implicit features, and the first feature vector set is a feature vector obtained from the resume sample and contains the implicit features of the user. The second feature vector set is composed of feature vectors of the position information samples and contains position latent semantic features. And respectively matching the feature vectors in the first feature vector set with the feature vectors in the second feature vector set, wherein the matching can be performed by vector cross product calculation, and the calculation result is used as the matching probability value of each feature vector.
The preset sample probability table can be a probability table corresponding to the resume samples and the post information samples, and if the matching probability values of the feature vectors in the resume samples and the feature vectors in the post information samples are preset, the deep forest model is trained until the matching probability values of the feature vectors in the resume samples and the feature vectors in the post information samples obtained through training are matched with the probability values in the preset sample probability table, and the trained deep forest model is obtained.
And S105, the server determines at least one post information matched with the currently undetermined person based on the second evaluation score and the matching probability value so as to send the post information to the corresponding user terminal.
In this embodiment of the application, the server determines, based on the second evaluation score and the matching probability value, at least one position information matched with the currently undetermined person, and specifically includes:
firstly, the server determines a second evaluation score which is larger than a first preset threshold value and is a score of a position to be determined.
The server can eliminate the second evaluation score which is too small, and the recommendation accuracy when the employment information is recommended is guaranteed.
And then, the server determines the information of the position to be determined corresponding to the value of the position to be determined.
The score value of the position to be determined corresponds to position information, and the position information is the information of the position to be determined.
And meanwhile, the server determines the corresponding matching probability value of the information of the position to be positioned.
And then, the server determines a third evaluation score corresponding to each to-be-positioned information according to the matching probability value corresponding to the to-be-positioned information, the to-be-positioned score corresponding to the to-be-positioned information and the preset matching weight.
The preset matching weight is set by the user, for example, the matching probability value and the weight of the score of the undetermined position are 0.6 and 0.4 respectively, and then the third evaluation score is: y = 0.6U +0.4P, wherein the third evaluation score is Y, the matching probability value is U, and the pending position score is P.
And finally, the server takes the position information corresponding to the third evaluation score which is larger than the second preset threshold value as the position information matched with the currently undetermined person.
The second preset threshold is set by the user, which is not specifically limited in this application.
In this embodiment of the present application, before the server sends the post information to the corresponding user terminal, the method further includes:
and the server generates a data dictionary of the post information and the corresponding resume information.
The data dictionary includes a number of triples of text. The triple text includes: and the employment ability text and the post name correspond to the personnel name and the resume information.
The data dictionary may be as follows: zhang san \ t education experience \ t legal affair supervisor; zhang three \ t professional skill \ t lawyer.
And then, the server encodes each triple text into a vector group through a preset TransH model, projects each vector group to a corresponding relation hyperplane, and calculates the recommended value of each triple text in the data dictionary according to a preset formula and the projection vector of each vector group in the relation hyperplane.
Wherein, the vector group comprises a personnel name vector, a employment ability text vector and a position name vector; the relation hyperplane is the hyperplane corresponding to the employment ability text vector in the vector group.
1, separating a head part, a relation part and a tail part; step 2, transforming the index tensor into an embedding tensor; step 3, embedding the head and the tail of the project entity into the associated hyperplane; and 4, calculating the associated hyperplane similarity score, namely the recommendation score. In the embodiment of the present application, the person name vector and the position name vector in the data dictionary are projected onto the hyperplane corresponding to the employment ability text vector along the normal vector of the hyperplane, wherein the projected vectors can be represented as,
h ⊥ ＝h-w r T *h*w r ，t ⊥ ＝t-w r T *t*w r
wherein h is ⊥ Is the name vector of the person after projection, h is the name vector of the person, w r Is a hyperplane normal vector, t ⊥ Is the post name vector after projection, and t is the post name vector. Recommendation score calculation formula:
wherein f is a recommendation score, and r is an employment ability text vector.
And then, the server eliminates the triple texts with the recommended values smaller than the preset value, and updates the data dictionary to determine the position information according to the updated data dictionary.
In this embodiment of the present application, after the server sends the post information to the corresponding user terminal, the method further includes:
first, the server acquires feedback information of the user terminal.
The user terminal may be a mobile phone, a tablet computer, or other devices of the user, which is not specifically limited in this application. The feedback information may be an accurate result of the recommended position information fed back by the user through the user terminal, such as "the position information does not conform to career planning", "the position information is not suitable for my interest", and so on.
And then, the server determines the accuracy and the recall rate of the position information according to the feedback information.
Wherein, TP: TP is positive type and is judged as positive type; FP: FP is negative and is determined to be positive; FN: FN is positive class and determined as negative class; TN: TN is negative and judged as negative; accuracy = TP/(TP + FP); recall = TP/(TP + FN).
And then, the server determines whether the post information is the post information to be updated according to the accuracy and the recall rate.
And under the condition that the post information is determined to be the post information to be updated, eliminating the recruiter information corresponding to the post information to be updated, and updating the first evaluation score and the second evaluation score according to the updated recruiter information so as to update the post information to be updated.
And under the condition that the F1 is smaller than a fourth preset threshold, determining the post information as the post information to be updated.
According to the technical scheme, the first evaluation scores corresponding to the position information of the pending person to be cared and the historical recruiters are determined, and then the position information matched with the pending person is obtained through the latent semantic model and the deep forest model, so that the employment information is accurately recommended. According to the application, the proper position can be quickly found by utilizing the working experience and the technical capability mastered by the personnel to be cared, so that the resume delivery of the personnel to be cared is not delivered blindly, and the normal employment of the personnel to be cared and the long-distance professional planning are ensured.
The employment information can be processed, identified and screened for people to refer to so as to solve most of labor cost. In addition, the method and the system can provide corresponding training courses according to resume information of the undetermined personnel, and the employment of the undetermined personnel is accelerated.
Fig. 2 is a schematic structural diagram of an employment information processing apparatus for persons to be cared for according to an embodiment of the present application, the apparatus including:
at least one processor; and a memory communicatively coupled to the at least one processor. Wherein the memory stores instructions executable by the at least one processor to cause the at least one processor to:
and acquiring a plurality of position information. The position information at least includes: a post name, a post responsibility text, a post requirement text. And determining a first evaluation score corresponding to each post information based on the recruiter information corresponding to each post information and the resume information of the currently pending person. And the first evaluation score represents the matching degree of the currently undetermined person and each post information. And determining a second evaluation score corresponding to the undetermined person corresponding to the first evaluation score according to the first evaluation score and the preset implicit semantic model. And determining the corresponding matching probability value of the resume information and each post information through a pre-trained deep forest model. And determining at least one post information matched with the currently undetermined person based on the second evaluation score and the matching probability value so as to send the post information to the corresponding user terminal.
The embodiments in the present application are described in a progressive manner, and the same and similar parts among the embodiments can be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the apparatus embodiment, since it is substantially similar to the method embodiment, the description is relatively simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The device and the method provided by the embodiment of the application are in one-to-one correspondence, so the device also has the beneficial technical effects similar to the corresponding method, and the beneficial technical effects of the method are explained in detail above, so the beneficial technical effects of the device are not described in detail here.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising a … …" does not exclude the presence of another identical element in a process, method, article, or apparatus that comprises the element.
The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement or the like made within the spirit and principle of the present application shall be included in the scope of the claims of the present application.
Claims (10)
1. A employment information processing method for a group of persons to be cared for, the method comprising:
acquiring a plurality of post information; the post information at least includes: the method comprises the following steps of (1) a post name, a post responsibility text and a post requirement text;
determining a first evaluation score corresponding to each post information based on the recruiter information corresponding to each post information and the resume information of the currently pending person; the first evaluation score represents the matching degree of the currently undetermined person and each position information;
determining a second evaluation score corresponding to the undetermined person corresponding to the first evaluation score according to the first evaluation score and a preset implicit model;
determining the matching probability values of the resume information and the post information through a pre-trained deep forest model;
and determining at least one of the post information matched with the currently pending person based on the second evaluation score and the matching probability value so as to send the post information to a corresponding user terminal.
2. The method of claim 1, wherein determining a first appraisal score corresponding to each of the position information based on the recruiter information corresponding to each of the position information and the resume information of the currently pending person comprises:
determining cosine similarity between a text corresponding to the resume information and a text corresponding to each recruiter information according to the resume information of the currently undetermined person and the information of each recruiter; wherein, the corresponding text is an evaluation text of a post;
generating a similarity matrix of the currently undetermined person and each recruiter according to the cosine similarity;
and determining the corresponding position score of the currently undetermined person as the first evaluation score based on the evaluation text of each position by each recruiter and the corresponding cosine similarity in the similarity matrix.
3. The method according to claim 1, wherein determining a second evaluation score corresponding to the undetermined person corresponding to the first evaluation score according to the first evaluation score and a preset implicit model specifically comprises:
inputting the resume information of the currently pending person and the evaluation text corresponding to the first evaluation score into the latent semantic model, so that the latent semantic model determines a scoring matrix corresponding to the currently pending person and the recruiter corresponding to the first evaluation score;
factorizing the scoring matrix through the hidden semantic model and a preset loss function until the preset loss function meets a preset condition so as to decompose the scoring matrix into a user scoring matrix and a post weight matrix;
determining the predictive evaluation score of the currently undetermined person for each post based on the user scoring matrix and the post weight matrix;
and taking the weighted average of the first evaluation score and the predicted evaluation score as the second evaluation score.
4. The method as claimed in claim 1, wherein before determining the matching probability values of the resume information and the post information through a pre-trained deep forest model, the method further comprises:
acquiring a plurality of resume samples and a plurality of position information samples;
and respectively carrying out short text coding processing on each resume sample and each post information sample through the deep forest model to determine a first feature vector set corresponding to each resume sample and a second feature vector set corresponding to each post information sample, calculating the matching probability value of each feature vector in the first feature vector set and each feature vector in the second feature vector set until the matching probability value meets a preset sample probability table, and determining that the deep forest model is trained.
5. The method according to claim 4, wherein determining at least one of the post information matched by the currently pending person based on the second evaluation score and the match probability value comprises:
determining the second evaluation score which is larger than a first preset threshold value as a score of a position to be determined;
determining the information of the pending position corresponding to the score of the pending position; and
determining the matching probability value corresponding to the information of the position to be determined;
determining a third evaluation score corresponding to each piece of information to be positioned according to the matching probability value corresponding to the information to be positioned, the score of the information to be positioned corresponding to the information to be positioned and a preset matching weight;
and taking the position information corresponding to the third evaluation score which is larger than a second preset threshold value as the position information matched with the currently undetermined person.
6. The method according to claim 5, wherein before sending the position information to the corresponding user terminal, the method further comprises:
generating a data dictionary of the post information and the corresponding resume information; the data dictionary comprises a plurality of triple texts; the triplet text includes: the names of the persons, employment ability texts and post names corresponding to the resume information;
respectively encoding each triple text into vector groups through a preset TransH model, projecting each vector group to a corresponding relation hyperplane, and calculating the recommended value of each triple text in the data dictionary according to a preset formula and the projection vector of each vector group in the relation hyperplane; wherein the vector group comprises a person name vector, a employment capability text vector and a position name vector; the relation hyperplane is a hyperplane corresponding to the employment ability text vector in the vector group;
and removing the triple text with the recommended value smaller than a preset value, updating the data dictionary, and determining the post information according to the updated data dictionary.
7. The method of claim 1, wherein prior to obtaining the position information, the method further comprises:
acquiring a plurality of recruitment information and post keywords corresponding to the recruitment information;
and sequentially combining the post keywords into post key sentences according to the position sequence of the post keywords in the recruitment information, taking the post key sentences as training samples, and inputting the post key sentences into a bidirectional LSTM-CRF model to train the bidirectional LSTM-CRF model.
8. The method of claim 7, further comprising:
screening a plurality of post keywords in the recruitment information through the bidirectional LSTM-CRF model, and taking the screened post keywords as post responsibility sample words;
determining post demand sample words corresponding to the post responsibility texts in the recruitment information;
inputting the post responsibility sample words and the post demand sample words into a keyword extraction model for training so as to obtain the post information through the trained keyword extraction model; the keyword extraction model is a TextRank algorithm model.
9. The method according to claim 1, wherein after sending the position information to the corresponding user terminal, the method further comprises:
acquiring feedback information of the user terminal;
according to the feedback information, determining the accuracy and the recall rate of the post information;
determining whether the post information is to-be-updated post information or not according to the accuracy and the recall rate;
and if so, eliminating the recruiter information corresponding to the post information to be updated, and updating the first evaluation score and the second evaluation score according to the updated recruiter information so as to update the post information to be updated.
10. An employment information processing apparatus for a group of persons to be cared for, characterized by comprising:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein,
the memory stores instructions executable by the at least one processor to enable the at least one processor to:
acquiring a plurality of post information; the post information at least includes: the method comprises the following steps of (1) a post name, a post responsibility text and a post requirement text;
determining a first evaluation score corresponding to each post information based on the historical recruiter information corresponding to each post information and the resume information of the currently undetermined person; the first evaluation score represents the matching degree of the currently undetermined person and each position information;
determining a second evaluation score corresponding to the currently undetermined person according to the first evaluation score and a preset implicit model;
determining the matching probability values of the resume information and the post information through a pre-trained deep forest model;
and determining at least one of the post information matched with the currently pending person based on the second evaluation score and the matching probability value so as to send the post information to a corresponding user terminal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211128537.7A CN115526589B (en) | 2022-09-16 | 2022-09-16 | Employment information processing method and equipment for crowd to be employment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211128537.7A CN115526589B (en) | 2022-09-16 | 2022-09-16 | Employment information processing method and equipment for crowd to be employment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115526589A true CN115526589A (en) | 2022-12-27 |
| CN115526589B CN115526589B (en) | 2023-06-06 |
Family
ID=84697873
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211128537.7A Active CN115526589B (en) | 2022-09-16 | 2022-09-16 | Employment information processing method and equipment for crowd to be employment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115526589B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117196550A (en) * | 2023-09-22 | 2023-12-08 | 蔓悦科技(宁波)有限公司 | Talent and enterprise supply and demand matching method and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110633960A (en) * | 2019-09-25 | 2019-12-31 | 重庆市重点产业人力资源服务有限公司 | Human resource intelligent matching and recommending method based on big data |
| CN111105209A (en) * | 2019-12-17 | 2020-05-05 | 上海沃锐企业发展有限公司 | Job resume matching method and device suitable for post matching recommendation system |
| CN111144723A (en) * | 2019-12-17 | 2020-05-12 | 埃摩森网络科技(上海)有限公司 | Method and system for recommending people's job matching and storage medium |
| CN113435841A (en) * | 2021-06-24 | 2021-09-24 | 浙江工贸职业技术学院 | Talent intelligent matching recruitment system based on big data |
| CN114819924A (en) * | 2022-06-28 | 2022-07-29 | 杭银消费金融股份有限公司 | Enterprise information push processing method and device based on portrait analysis |
-
2022
- 2022-09-16 CN CN202211128537.7A patent/CN115526589B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110633960A (en) * | 2019-09-25 | 2019-12-31 | 重庆市重点产业人力资源服务有限公司 | Human resource intelligent matching and recommending method based on big data |
| CN111105209A (en) * | 2019-12-17 | 2020-05-05 | 上海沃锐企业发展有限公司 | Job resume matching method and device suitable for post matching recommendation system |
| CN111144723A (en) * | 2019-12-17 | 2020-05-12 | 埃摩森网络科技(上海)有限公司 | Method and system for recommending people's job matching and storage medium |
| CN113435841A (en) * | 2021-06-24 | 2021-09-24 | 浙江工贸职业技术学院 | Talent intelligent matching recruitment system based on big data |
| CN114819924A (en) * | 2022-06-28 | 2022-07-29 | 杭银消费金融股份有限公司 | Enterprise information push processing method and device based on portrait analysis |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117196550A (en) * | 2023-09-22 | 2023-12-08 | 蔓悦科技(宁波)有限公司 | Talent and enterprise supply and demand matching method and system |
| CN117196550B (en) * | 2023-09-22 | 2024-05-28 | 蔓悦科技(宁波)有限公司 | Talent and enterprise supply and demand matching method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115526589B (en) | 2023-06-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Neculoiu et al. | Learning text similarity with siamese recurrent networks | |
| US11238394B2 (en) | Assessment-based qualified candidate delivery | |
| US20150317610A1 (en) | Methods and system for automatically obtaining information from a resume to update an online profile | |
| US20140278633A1 (en) | Skill-based candidate matching | |
| CN111241361B (en) | Intelligent referral system and method for enterprises and colleges based on cloud platform | |
| CN111192024A (en) | Method and device for matching human posts based on reinforcement learning | |
| CN111626616A (en) | Crowdsourcing task recommendation method | |
| US20200302397A1 (en) | Screening-based opportunity enrichment | |
| CN113656687B (en) | Teacher portrait construction method based on teaching and research data | |
| CN110866850A (en) | Accurate online education system of enterprise | |
| CN116362699A (en) | Post matching report generation method | |
| KR100797225B1 (en) | E-Class Management system and Evaluation method for grade | |
| Faliagka et al. | An Integrated e-Recruitment System for CV Ranking based on AHP. | |
| CN115526589B (en) | Employment information processing method and equipment for crowd to be employment | |
| Kortum et al. | Dissection of AI job advertisements: A text mining-based analysis of employee skills in the disciplines computer vision and natural language processing | |
| Aloqayli et al. | Intelligent Chatbot for Admission in Higher Education | |
| CN111782803B (en) | Work order processing method and device, electronic equipment and storage medium | |
| Okhdar et al. | English vocabulary learning through recommender system based on sentence complexity and vocabulary difficulty | |
| CN112507082B (en) | Method and device for intelligently identifying improper text interaction and electronic equipment | |
| Mouchel et al. | Understanding revision behavior in adaptive writing support systems for education | |
| CN115640403A (en) | Knowledge management and control method and device based on knowledge graph | |
| Tennakoon et al. | An interactive application for university students to reduce the industry-academia skill gap in the software engineering field | |
| Ghadekar et al. | A Semantic Approach for Automated Hiring using Artificial Intelligence & Computer Vision | |
| CN108153829B (en) | Resume evaluation method and device | |
| CN111177283A (en) | Intelligent personnel model construction method and device based on reinforcement learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |