CN118013245A - Matching degree evaluation method based on deep learning model - Google Patents
Matching degree evaluation method based on deep learning model Download PDFInfo
- Publication number
- CN118013245A CN118013245A CN202410421346.2A CN202410421346A CN118013245A CN 118013245 A CN118013245 A CN 118013245A CN 202410421346 A CN202410421346 A CN 202410421346A CN 118013245 A CN118013245 A CN 118013245A
- Authority
- CN
- China
- Prior art keywords
- information
- items
- work skill
- matching
- user
- 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
- 238000011156 evaluation Methods 0.000 title claims abstract description 41
- 238000013136 deep learning model Methods 0.000 title claims abstract description 37
- 238000012216 screening Methods 0.000 claims abstract description 144
- 238000000034 method Methods 0.000 claims abstract description 30
- 238000012549 training Methods 0.000 claims description 3
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 238000012937 correction Methods 0.000 description 11
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004590 computer program Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a matching degree evaluation method based on a deep learning model, which belongs to the technical field of artificial intelligence and specifically comprises the following steps: the method comprises the steps of extracting working skill items of a user through a recognition result of a resume, determining association coefficients among the working skill items of the user according to the working skill items of a platform user, determining basic credibility of the working skill items of the user and screening the working skill items based on the association coefficients, obtaining matching information dimensions of the screening working skill items and dimension evaluation coefficients of different information dimensions, determining comprehensive credibility and credible working skill items of the screening working skill items, constructing user portrait information of the user according to the credible working skill items of the user and information data of different information dimensions, determining matching degree scores of the user and different posts by adopting a deep learning model in combination with resume portrait information of different posts, and guaranteeing evaluation accuracy of the matching degree scores.
Description
Technical Field
The invention belongs to the technical field of artificial intelligence, and particularly relates to a matching degree evaluation method based on a deep learning model.
Background
In order to realize the recommendation of the matching post of the user, in the prior art, the user's resume and post are analyzed to realize the recommendation of the matching post of the user, and the following problems are found by analysis, particularly in the invention patent application CN202311604954.9, a personal post matching system based on intelligent data analysis, CN202311239086.9, post item matching method, device, equipment and storage medium, which all provide similar technical schemes:
When the resume information is filled in, the user may not process the skill information, so if the reliability of the related skill information cannot be evaluated, the accuracy of the evaluation result of the final post matching degree may be difficult to meet the requirement.
Aiming at the technical problems, the invention provides a matching degree evaluation method based on a deep learning model.
Disclosure of Invention
In order to achieve the purpose of the invention, the invention adopts the following technical scheme:
according to one aspect of the invention, a matching degree evaluation method based on a deep learning model is provided.
The matching degree evaluation method based on the deep learning model is characterized by comprising the following steps of:
S1, extracting working skill items of a user through a recognition result of a resume, determining association coefficients among the working skill items of the user according to the working skill items of a platform user, determining basic credibility of the working skill items of the user based on the association coefficients, and screening the working skill items;
S2, dividing the resume of the user into different information dimensions according to the identification result, determining dimension matching coefficients of the different information dimensions and the screening work skill items of the user according to different information items of the different information dimensions and matching data of the screening work skill items, and determining matching information dimensions of the screening work skill items based on the dimension matching coefficients;
S3, acquiring matching information dimensionality of the screening work skill items and dimensionality evaluation coefficients of the information dimensionality and the different information dimensionalities, and determining comprehensive credibility and credible work skill items of the screening work skill items by combining the basic credibility of the screening work skill items;
s4, constructing user portrait information of the user according to the trusted work skill items of the user and information data of different information dimensions, and determining matching degree scores of the user and different posts by adopting a deep learning model in combination with resume portrait information of different posts.
The invention has the beneficial effects that:
1. The method has the advantages that the dimension matching coefficients of different information dimensions and screening work skill items are determined through the matching data of the different information items of the different information dimensions and the screening work skill items of the user, so that the matching conditions of the information items of the different information dimensions and the screening work skill items are realized, the accurate evaluation of the matching coefficients of the different information dimensions and the screening work skill items is realized, a foundation is laid for further realizing the accurate evaluation of the screening work skill items with higher credibility in the screening work skill items, and the technical problem that the evaluation results of the matching degree scores with different posts are inaccurate due to low accuracy of the work skill items is avoided.
2. And determining the matching degree scores of the user and different posts by adopting a deep learning model according to the user portrait information of the user and the resume portrait information of different posts, not only considering the working skills of the user with higher credibility, but also ensuring the accuracy of the evaluation result of the matching degree score by constructing the user portrait information and the resume portrait information.
The further technical scheme is that the working skill items are extracted according to matching results of resume data of the user and a preset working skill item database.
The further technical scheme is that the association coefficient between the work skill items of the users is determined according to the distribution condition of the work skill items in all the platform users, and the number of the platform users with different work skill items in the same platform is specifically determined.
The further technical scheme is that the method for constructing the user portrait information of the user comprises the following steps:
Extracting information items of different information dimensions of the user according to the information data of the different information dimensions of the user, and constructing user portrait information of the user based on the trusted work skill items and the information items of different information dimensions.
The further technical scheme is that the method for determining the matching degree score of the user and the post comprises the following steps:
And taking the user portrait information of the user and the post portrait information of the post as input items of the deep learning model, and determining a matching degree score of the user and the post based on an output result of the deep learning model.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be apparent from the description, or may be learned by practice of the invention as set forth hereinafter.
In order to make the above objects, features and advantages of the present invention more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
The above and other features and advantages of the present invention will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings;
FIG. 1 is a flow chart of a matching degree evaluation method based on a deep learning model;
FIG. 2 is a flow chart of a method of determining a base trustworthiness of a user's work skill items;
FIG. 3 is a flow chart of a method of determining the dimension matching coefficients of an information dimension to the screening work skill item;
FIG. 4 is a flow chart of a method of screening for a determination of the integrated trustworthiness of a work skill item;
FIG. 5 is a block diagram of a computer system.
Detailed Description
In order to make the technical solutions in the present specification better understood by those skilled in the art, the technical solutions in the embodiments of the present specification will be clearly and completely described below with reference to the drawings in the embodiments of the present specification, and it is obvious that the described embodiments are only some embodiments of the present specification, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, shall fall within the scope of the present disclosure.
In the prior art, when the position matching degree is evaluated, the position matching is often performed according to job seeking intention of a user or position browsing records and the like, but the difference of the requirements of different positions on the working skills is ignored, and if the working skills of the user are not matched with the positions, more invalid recommendations can be caused.
S1, extracting working skill items of a user through a recognition result of a resume, determining association coefficients among the working skill items of the user according to the working skill items of a platform user, determining basic credibility of the working skill items of the user based on the association coefficients, and screening the working skill items;
S2, dividing the resume of the user into different information dimensions according to the identification result, determining dimension matching coefficients of the different information dimensions and the screening work skill items of the user according to different information items of the different information dimensions and matching data of the screening work skill items, and determining matching information dimensions of the screening work skill items based on the dimension matching coefficients;
S3, acquiring matching information dimensionality of the screening work skill items and dimensionality evaluation coefficients of the information dimensionality and the different information dimensionalities, and determining comprehensive credibility and credible work skill items of the screening work skill items by combining the basic credibility of the screening work skill items;
s4, constructing user portrait information of the user according to the trusted work skill items of the user and information data of different information dimensions, and determining matching degree scores of the user and different posts by adopting a deep learning model in combination with resume portrait information of different posts.
In order to solve the technical problems, the application adopts the following technical scheme:
1. Screening work skill items for determination;
Because the working skill items in the resume of the user are filled in by themselves, for example, the working skill items are familiar with language programming such as JAVA and C language, and the working skill items such as mechanical design software, and in general, the mechanical design software is different from the industry fields of language programming such as JAVA and C language, so that the working skill items are required to be screened according to the association condition among different working skill items, and the working skill items which do not belong to the industry fields are specifically screened, thereby obtaining the screened working skill items familiar with the language programming such as JAVA and C language.
2. Determining the dimension of the matching information;
The working experience, training experience and the like of the user often have differences with the corresponding relations of different JAVA, C languages and the like, so that the working experience can be used as a matching information dimension through information items in the working experience of the user, such as the use and programming of the JAVA languages in multiple sections of working experiences, and dimension matching coefficients are determined according to the duty ratio of the working quantity.
3. Determining trusted work skill items;
And then determining the dimension credibility according to the sum of the dimension evaluation coefficients of the matching information dimension of the screening work skill item, and determining the comprehensive credibility of the screening work skill item according to the weight sum of the dimension credibility and the basic credibility, wherein the comprehensive credibility is used as the credible work skill item.
4. And (3) matching degree score determination:
Extracting information items of different information dimensions of the user according to information data of different information dimensions of the user, constructing user portrait information of the user based on the trusted work skill items and the information items of different information dimensions, taking the user portrait information of the user and the post portrait information of the post as input items of the deep learning model, and determining a matching degree score of the user and the post based on an output result of the deep learning model, wherein the deep learning model comprises one or more combinations in a GRU, RNN, CNN and other neural networks for construction.
Further explanation will be made below from two perspectives of the method class embodiment and the system class embodiment.
In order to solve the above-mentioned problems, according to an aspect of the present invention, as shown in fig. 1, there is provided a matching degree evaluation method based on a deep learning model, which is characterized by comprising:
S1, extracting working skill items of a user through a recognition result of a resume, determining association coefficients among the working skill items of the user according to the working skill items of a platform user, determining basic credibility of the working skill items of the user based on the association coefficients, and screening the working skill items;
further, the work skill items are extracted according to matching results of the resume data of the user and a preset work skill item database.
Specifically, the association coefficient between the work skill items of the users is determined according to the distribution condition of the work skill items in all the platform users, and specifically, the number of the platform users with different work skill items in the same platform user is determined.
Specifically, as shown in fig. 2, the method for determining the basic credibility of the work skill item of the user is as follows:
other work skills except the work skills of the user are taken as other work skills, and the other work skills are divided into associated work skills and deviation work skills according to the association coefficients of the work skills and the other work skills;
Determining the comprehensive association degree of the associated work skill items according to the number of the associated work skill items and the association coefficients of different associated work skill items, and determining the comprehensive association degree of the deviation work skill items based on the number of the deviation work skill items and the association coefficients of different deviation work skill items;
And acquiring the number of the work skill items of the user, and determining the basic credibility of the work skill items of the user by combining the comprehensive association degree of the associated work skill items and the comprehensive association degree of the deviation work skill items.
In another embodiment, the method for determining the basic credibility of the work skill item of the user is as follows:
Taking other work skill items except the work skill item of the user as other work skill items, determining the sum of the association coefficients according to the association coefficients of the work skill item and the other work skill items, judging whether the sum of the association coefficients meets the requirement, if so, entering the next step, and if not, determining that the work skill item does not belong to the screening work skill item;
Dividing the other work skill items into associated work skill items and deviation work skill items according to the association coefficients of the work skill items and the other work skill items, judging whether the number of the associated work skill items is smaller than the number of preset skill items, if so, determining that the work skill items do not belong to screening work skill items, and if not, entering the next step;
Determining the comprehensive association degree of the associated work skill items according to the number of the associated work skill items and the association coefficients of different associated work skill items, judging whether the comprehensive association degree of the associated work skill items is smaller than a preset association degree, if so, determining that the work skill items do not belong to screening work skill items, and if not, entering the next step;
determining the comprehensive association degree of the deviation work skill items based on the number of the deviation work skill items and association coefficients of different deviation work skill items;
And acquiring the number of the work skill items of the user, and determining the basic credibility of the work skill items of the user by combining the comprehensive association degree of the associated work skill items and the comprehensive association degree of the deviation work skill items.
Further, when the basic credibility of the user's work skill items is greater than a preset credibility, determining the user's work skill items as screening work skill items.
In another embodiment, the method for determining the basic credibility of the work skill item of the user is as follows:
S11, taking other work skill items except the work skill items of the user as other work skill items, determining the sum of the association coefficients according to the association coefficients of the work skill items and the other work skill items, determining the basic association degree of the user by combining the number of the work skill items of the user, judging whether the basic association degree of the user meets the requirement, if so, entering a step S14, and if not, entering a next step;
S12, dividing the other work skill items into associated work skill items and deviation work skill items according to the association coefficients of the work skill items and the other work skill items, judging whether the number of the associated work skill items is smaller than the number of preset skill items, if so, determining that the work skill items do not belong to screening work skill items, and if not, entering the next step;
S13, determining the comprehensive association degree of the associated work skill items according to the number of the associated work skill items and the association coefficients of different associated work skill items, judging whether the comprehensive association degree of the associated work skill items is smaller than a preset association degree, if so, determining that the work skill items do not belong to screening work skill items, and if not, entering the next step;
S14, determining the comprehensive association degree of the deviation work skill items based on the number of the deviation work skill items and association coefficients of different deviation work skill items, and determining the basic credibility of the work skill items of the user by combining the comprehensive association degree of the association work skill items and the basic association degree of the user.
S2, dividing the resume of the user into different information dimensions according to the identification result, determining dimension matching coefficients of the different information dimensions and the screening work skill items of the user according to different information items of the different information dimensions and matching data of the screening work skill items, and determining matching information dimensions of the screening work skill items based on the dimension matching coefficients;
further, the information dimension comprises academic information, work experience information and training experience information.
Specifically, as shown in fig.3, the method for determining the matching coefficient between the information dimension and the dimension of the screening work skill item is as follows:
determining matching information items and unmatched information items of the screening work skill items and the information dimension based on the information data of the information dimension;
Determining comprehensive matching coefficients of the matching information items of the information dimension according to the number of the matching information items of the information dimension and the matching coefficients of different matching information items and the screening work skill items, and determining comprehensive matching coefficients of the non-matching information items of the information dimension according to the number of the non-matching information items of the information dimension and the matching coefficients of different non-matching information items and the screening work skill items;
and determining the dimension matching coefficient of the information dimension and the screening work skill item through the comprehensive matching coefficient of the matching information item of the information dimension and the comprehensive matching coefficient of the non-matching information item.
Further, when the dimension matching coefficient of the information dimension and the screening work skill item meets the requirement, determining that the information dimension is a matching information dimension.
In another embodiment, the method for determining the dimension matching coefficient of the information dimension and the dimension matching coefficient of the screening work skill item is as follows:
Determining the number of matching information items and the number of unmatched information items of the screening work skill items and the information dimension based on the information data of the information dimension, and determining a basic dimension matching coefficient of the information dimension and the screening work skill items according to the number of matching information items and the number of unmatched information items of the information dimension;
judging whether a matching information item with a matching coefficient larger than a preset matching coefficient exists or not according to the matching coefficient of the matching information item of the information dimension and the screening work skill item, if so, entering the next step, and if not, determining a dimension matching coefficient of the information dimension and the screening work skill item based on the basic dimension matching coefficient;
Determining comprehensive matching coefficients of the information dimension matching information items according to the number of the information dimension matching information items and the matching coefficients of different matching information items and the screening work skill items, judging whether the comprehensive matching coefficients of the information dimension matching information items are larger than a preset coefficient threshold, if so, entering the next step, and if not, determining dimension matching coefficients of the information dimension and the screening work skill items based on the basic dimension matching coefficients;
And determining the comprehensive matching coefficient of the unmatched information items of the information dimension according to the number of the unmatched information items of the information dimension and the matching coefficients of different unmatched information items and the screening work skill item, and determining the dimension matching coefficient of the information dimension and the screening work skill item through the comprehensive matching coefficient of the matched information items of the information dimension and the comprehensive matching coefficient of the unmatched information items.
S3, acquiring matching information dimensionality of the screening work skill items and dimensionality evaluation coefficients of the information dimensionality and the different information dimensionalities, and determining comprehensive credibility and credible work skill items of the screening work skill items by combining the basic credibility of the screening work skill items;
Specifically, as shown in fig. 4, the method for determining the comprehensive credibility of the screening work skill items is as follows:
Determining the basic credibility of the screening work skill items by the number of the information dimensions of the screening work skill items and the average value of the dimension evaluation coefficients of different information dimensions;
determining the credibility correction amount of the screening work skill item based on the number of the matching information dimensions of the screening work skill item and the dimension evaluation coefficients of different matching information dimensions and the screening work skill item;
And determining the comprehensive credibility of the screening work skill item according to the credibility correction amount of the screening work skill item and the basic credibility of the screening work skill item.
Further, when the integrated credibility of the screening work skill item is greater than a preset credibility threshold, determining the screening work skill item as a credible work skill item.
In another embodiment, the method for determining the comprehensive credibility of the screening work skill items is as follows:
Determining the basic credibility of the screening work skill items through the number of the information dimensionalities of the screening work skill items and the average value of the dimensionality evaluation coefficients of different information dimensionalities, judging whether the basic credibility of the screening work skill items meets the requirement, if so, determining the screening work skill items as credible work skill items, and if not, entering the next step;
Judging whether the screening work skill item has a matching information dimension, if so, determining that the screening work skill item does not belong to a trusted work skill item, and if not, entering the next step;
acquiring the number of the matching information dimensionalities of the screening work skill items, judging whether the number of the matching information dimensionalities of the screening work skill items is larger than the preset dimensionality number, if so, determining that the screening work skill items are trusted work skill items, and if not, entering the next step;
Determining the credibility correction amount of the screening work skill item according to the number of the matching information dimensionalities and the dimensionality evaluation coefficients of different matching information dimensionalities and the screening work skill item, judging whether the credibility correction amount of the screening work skill item is larger than a preset threshold value, if so, determining that the screening work skill item is the credibility work skill item, and if not, entering the next step;
And determining the comprehensive credibility of the screening work skill item according to the credibility correction amount of the screening work skill item and the basic credibility of the screening work skill item.
In another embodiment, the method for determining the comprehensive credibility of the screening work skill items is as follows:
When the screening work skill item does not have the matching information dimension, determining that the screening work skill item does not belong to the credible work skill item;
when the screening work skill item has a matching information dimension:
when the number of the matching information dimensionalities of the screening work skill items is larger than the number of the preset dimensionalities, determining the screening work skill items as credible work skill items;
When the number of the matching information dimensionalities of the screening work skill items is not more than the number of the preset dimensionalities, judging whether the matching information dimensionalities with the dimensionality matching degree being more than the preset dimensionality matching coefficient exist or not, if so, determining that the screening work skill items are credible work skill items, and if not, entering the next step;
Determining the credibility correction amount of the screening work skill item according to the number of the matching information dimensionalities and the dimensionality evaluation coefficients of different matching information dimensionalities and the screening work skill item, judging whether the credibility correction amount of the screening work skill item is larger than a preset threshold value, if so, determining that the screening work skill item is the credibility work skill item, and if not, entering the next step;
And determining the basic credibility of the screening work skill item according to the credibility correction amount of the screening work skill item and the basic credibility of the screening work skill item.
S4, constructing user portrait information of the user according to the trusted work skill items of the user and information data of different information dimensions, and determining matching degree scores of the user and different posts by adopting a deep learning model in combination with resume portrait information of different posts.
Specifically, the method for constructing the user portrait information of the user comprises the following steps:
Extracting information items of different information dimensions of the user according to the information data of the different information dimensions of the user, and constructing user portrait information of the user based on the trusted work skill items and the information items of different information dimensions.
Furthermore, the deep learning model is constructed by adopting any one or more models of a GAN neural network, an RNN neural network and a deep residual error network.
Further, the method for determining the matching degree score of the user and the post comprises the following steps:
And taking the user portrait information of the user and the post portrait information of the post as input items of the deep learning model, and determining a matching degree score of the user and the post based on an output result of the deep learning model.
In another aspect, as shown in FIG. 5, the present invention provides a computer system comprising: a communicatively coupled memory and processor, and a computer program stored on the memory and capable of running on the processor, characterized by: the processor executes the matching degree evaluation method based on deep learning when running the computer program.
The matching degree evaluation method based on deep learning specifically comprises the following steps:
extracting the working skill items of the user through the recognition result of the resume, determining the association coefficient between the working skill items of the user according to the working skill items of the platform user, determining the basic credibility of the working skill items of the user based on the association coefficient, and screening the working skill items;
Determining the number of matching information items and the number of unmatched information items of the screening work skill items and the information dimension based on the information data of the information dimension, and determining a basic dimension matching coefficient of the information dimension and the screening work skill items according to the number of matching information items and the number of unmatched information items of the information dimension;
judging whether a matching information item with a matching coefficient larger than a preset matching coefficient exists or not according to the matching coefficient of the matching information item of the information dimension and the screening work skill item, if so, entering the next step, and if not, determining a dimension matching coefficient of the information dimension and the screening work skill item based on the basic dimension matching coefficient;
Determining comprehensive matching coefficients of the information dimension matching information items according to the number of the information dimension matching information items and the matching coefficients of different matching information items and the screening work skill items, judging whether the comprehensive matching coefficients of the information dimension matching information items are larger than a preset coefficient threshold, if so, entering the next step, and if not, determining dimension matching coefficients of the information dimension and the screening work skill items based on the basic dimension matching coefficients;
Determining comprehensive matching coefficients of the unmatched information items of the information dimension according to the number of the unmatched information items of the information dimension and the matching coefficients of different unmatched information items and the screening work skill item, determining dimension matching coefficients of the information dimension and the screening work skill item through the comprehensive matching coefficients of the matched information items of the information dimension and the comprehensive matching coefficients of the unmatched information items, and determining the matching information dimension of the screening work skill item based on the dimension matching coefficients;
When the screening work skill item does not have the matching information dimension, determining that the screening work skill item does not belong to the credible work skill item;
when the screening work skill item has a matching information dimension:
when the number of the matching information dimensionalities of the screening work skill items is larger than the number of the preset dimensionalities, determining the screening work skill items as credible work skill items;
When the number of the matching information dimensionalities of the screening work skill items is not more than the number of the preset dimensionalities, judging whether the matching information dimensionalities with the dimensionality matching degree being more than the preset dimensionality matching coefficient exist or not, if so, determining that the screening work skill items are credible work skill items, and if not, entering the next step;
Determining the credibility correction amount of the screening work skill item according to the number of the matching information dimensionalities and the dimensionality evaluation coefficients of different matching information dimensionalities and the screening work skill item, judging whether the credibility correction amount of the screening work skill item is larger than a preset threshold value, if so, determining that the screening work skill item is the credibility work skill item, and if not, entering the next step;
determining the basic credibility of the screening work skill item according to the credibility correction amount of the screening work skill item and the basic credibility of the screening work skill item;
Constructing user portrait information of the user according to the trusted work skill items of the user and the information data of different information dimensions, and determining matching degree scores of the user and different posts by adopting a deep learning model in combination with resume portrait information of different posts.
Through the above embodiments, the present invention has the following beneficial effects:
1. The method has the advantages that the dimension matching coefficients of different information dimensions and screening work skill items are determined through the matching data of the different information items of the different information dimensions and the screening work skill items of the user, so that the matching conditions of the information items of the different information dimensions and the screening work skill items are realized, the accurate evaluation of the matching coefficients of the different information dimensions and the screening work skill items is realized, a foundation is laid for further realizing the accurate evaluation of the screening work skill items with higher credibility in the screening work skill items, and the technical problem that the evaluation results of the matching degree scores with different posts are inaccurate due to low accuracy of the work skill items is avoided.
2. And determining the matching degree scores of the user and different posts by adopting a deep learning model according to the user portrait information of the user and the resume portrait information of different posts, not only considering the working skills of the user with higher credibility, but also ensuring the accuracy of the evaluation result of the matching degree score by constructing the user portrait information and the resume portrait information.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for apparatus, devices, non-volatile computer storage medium embodiments, the description is relatively simple, as it is substantially similar to method embodiments, with reference to the section of the method embodiments being relevant.
The foregoing describes specific embodiments of the present disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
The foregoing is merely one or more embodiments of the present description and is not intended to limit the present description. Various modifications and alterations to one or more embodiments of this description will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, or the like, which is within the spirit and principles of one or more embodiments of the present description, is intended to be included within the scope of the claims of the present description.
Claims (10)
1. The matching degree evaluation method based on the deep learning model is characterized by comprising the following steps of:
extracting the working skill items of the user through the recognition result of the resume, determining the association coefficient between the working skill items of the user according to the working skill items of the platform user, determining the basic credibility of the working skill items of the user based on the association coefficient, and screening the working skill items;
Dividing the resume of the user into different information dimensions according to the identification result, determining dimension matching coefficients of the different information dimensions and the screening work skill items of the user according to different information items of the different information dimensions and matching data of the screening work skill items, and determining the matching information dimensions of the screening work skill items based on the dimension matching coefficients;
Acquiring matching information dimensionality of the screening work skill items and dimensionality evaluation coefficients of different information dimensionalities, and determining comprehensive credibility and credible work skill items of the screening work skill items by combining the basic credibility of the screening work skill items;
Constructing user portrait information of the user according to the trusted work skill items of the user and the information data of different information dimensions, and determining matching degree scores of the user and different posts by adopting a deep learning model in combination with resume portrait information of different posts.
2. The deep learning model-based matching degree evaluation method of claim 1, wherein the work skill items are extracted according to matching results of resume data of the user and a preset work skill item database.
3. The matching degree evaluation method based on a deep learning model according to claim 1, wherein the correlation coefficient between the work skill items of the users is determined according to the distribution situation of the work skill items in all platform users, and specifically the number of the platform users with different work skill items in the same platform user is determined.
4. The matching degree evaluation method based on a deep learning model according to claim 1, wherein the method for determining the basic credibility of the user's work skill items is as follows:
other work skills except the work skills of the user are taken as other work skills, and the other work skills are divided into associated work skills and deviation work skills according to the association coefficients of the work skills and the other work skills;
Determining the comprehensive association degree of the associated work skill items according to the number of the associated work skill items and the association coefficients of different associated work skill items, and determining the comprehensive association degree of the deviation work skill items based on the number of the deviation work skill items and the association coefficients of different deviation work skill items;
And acquiring the number of the work skill items of the user, and determining the basic credibility of the work skill items of the user by combining the comprehensive association degree of the associated work skill items and the comprehensive association degree of the deviation work skill items.
5. The deep learning model based matching degree evaluation method of claim 1, wherein the information dimension includes academic information, work experience information, training experience information.
6. The method for evaluating the matching degree based on the deep learning model according to claim 1, wherein the method for determining the matching coefficient of the information dimension and the dimension of the screening work skill item is as follows:
determining matching information items and unmatched information items of the screening work skill items and the information dimension based on the information data of the information dimension;
Determining comprehensive matching coefficients of the matching information items of the information dimension according to the number of the matching information items of the information dimension and the matching coefficients of different matching information items and the screening work skill items, and determining comprehensive matching coefficients of the non-matching information items of the information dimension according to the number of the non-matching information items of the information dimension and the matching coefficients of different non-matching information items and the screening work skill items;
and determining the dimension matching coefficient of the information dimension and the screening work skill item through the comprehensive matching coefficient of the matching information item of the information dimension and the comprehensive matching coefficient of the non-matching information item.
7. The deep learning model based matching degree evaluation method of claim 6, wherein the information dimension is determined to be a matching information dimension when the information dimension meets the requirement with the dimension matching coefficient of the screening work skill item.
8. The deep learning model-based matching degree evaluation method of claim 1, wherein the screening work skill item is determined to be a trusted work skill item when the integrated confidence level of the screening work skill item is greater than a preset confidence threshold.
9. The matching degree evaluation method based on a deep learning model according to claim 1, wherein the method for constructing user portrait information of the user is as follows:
Extracting information items of different information dimensions of the user according to the information data of the different information dimensions of the user, and constructing user portrait information of the user based on the trusted work skill items and the information items of different information dimensions.
10. The method for evaluating the matching degree based on the deep learning model according to claim 1, wherein the method for determining the matching degree score of the user and the post is as follows:
And taking the user portrait information of the user and the post portrait information of the post as input items of the deep learning model, and determining a matching degree score of the user and the post based on an output result of the deep learning model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410421346.2A CN118013245B (en) | 2024-04-09 | 2024-04-09 | Matching degree evaluation method based on deep learning model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410421346.2A CN118013245B (en) | 2024-04-09 | 2024-04-09 | Matching degree evaluation method based on deep learning model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118013245A true CN118013245A (en) | 2024-05-10 |
| CN118013245B CN118013245B (en) | 2024-06-18 |
Family
ID=90954308
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410421346.2A Active CN118013245B (en) | 2024-04-09 | 2024-04-09 | Matching degree evaluation method based on deep learning model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118013245B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104318379A (en) * | 2014-10-20 | 2015-01-28 | 沈阳化工大学 | Talent recruitment management system |
| CN109409831A (en) * | 2018-10-18 | 2019-03-01 | 深圳前海乐寻坊区块链科技有限公司 | Resume authentication method, device, computer equipment and storage medium |
| CN111078971A (en) * | 2019-11-19 | 2020-04-28 | 平安金融管理学院(中国·深圳) | Resume file screening method and device, terminal and storage medium |
| KR20200141919A (en) * | 2019-06-11 | 2020-12-21 | 주식회사 에이아이앤잡 | Method for machine learning train set and recommendation systems to recommend the scores to match between the recruiter and job seekers, and to give the scores of matching candidates to recruiters and to give the pass scores to job seekers respectively |
| CN112559726A (en) * | 2020-12-22 | 2021-03-26 | 深圳市易博天下科技有限公司 | Resume information filtering method, model training method, device, equipment and medium |
| US20210097496A1 (en) * | 2018-06-11 | 2021-04-01 | Dingtalk Holding (Cayman) Limited | Method and Apparatus for Managing Resume Information, and Method and Apparatus for Managing Recruitment Information |
| CN116362703A (en) * | 2023-03-14 | 2023-06-30 | 成都商高智能科技有限公司 | Post screening method and system based on knowledge graph |
| CN116401464A (en) * | 2023-06-02 | 2023-07-07 | 深圳市一览网络股份有限公司 | Professional user portrait construction method, device, equipment and storage medium |
| CN116596494A (en) * | 2023-05-18 | 2023-08-15 | 数应科技(浙江)有限公司 | Person post matching method and system based on knowledge map deep learning |
| CN116596493A (en) * | 2023-04-04 | 2023-08-15 | 平安科技（深圳）有限公司 | Resume evaluation method, resume evaluation device, computer equipment and storage medium |
| CN116630033A (en) * | 2023-07-20 | 2023-08-22 | 杭银消费金融股份有限公司 | Information auditing method, system and storage medium |
| CN117093778A (en) * | 2023-08-23 | 2023-11-21 | 江苏徐工国重实验室科技有限公司 | Engineering machinery software task and personnel recommendation method and system based on word sense weighting |
-
2024
- 2024-04-09 CN CN202410421346.2A patent/CN118013245B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104318379A (en) * | 2014-10-20 | 2015-01-28 | 沈阳化工大学 | Talent recruitment management system |
| US20210097496A1 (en) * | 2018-06-11 | 2021-04-01 | Dingtalk Holding (Cayman) Limited | Method and Apparatus for Managing Resume Information, and Method and Apparatus for Managing Recruitment Information |
| CN109409831A (en) * | 2018-10-18 | 2019-03-01 | 深圳前海乐寻坊区块链科技有限公司 | Resume authentication method, device, computer equipment and storage medium |
| KR20200141919A (en) * | 2019-06-11 | 2020-12-21 | 주식회사 에이아이앤잡 | Method for machine learning train set and recommendation systems to recommend the scores to match between the recruiter and job seekers, and to give the scores of matching candidates to recruiters and to give the pass scores to job seekers respectively |
| CN111078971A (en) * | 2019-11-19 | 2020-04-28 | 平安金融管理学院(中国·深圳) | Resume file screening method and device, terminal and storage medium |
| CN112559726A (en) * | 2020-12-22 | 2021-03-26 | 深圳市易博天下科技有限公司 | Resume information filtering method, model training method, device, equipment and medium |
| CN116362703A (en) * | 2023-03-14 | 2023-06-30 | 成都商高智能科技有限公司 | Post screening method and system based on knowledge graph |
| CN116596493A (en) * | 2023-04-04 | 2023-08-15 | 平安科技（深圳）有限公司 | Resume evaluation method, resume evaluation device, computer equipment and storage medium |
| CN116596494A (en) * | 2023-05-18 | 2023-08-15 | 数应科技(浙江)有限公司 | Person post matching method and system based on knowledge map deep learning |
| CN116401464A (en) * | 2023-06-02 | 2023-07-07 | 深圳市一览网络股份有限公司 | Professional user portrait construction method, device, equipment and storage medium |
| CN116630033A (en) * | 2023-07-20 | 2023-08-22 | 杭银消费金融股份有限公司 | Information auditing method, system and storage medium |
| CN117093778A (en) * | 2023-08-23 | 2023-11-21 | 江苏徐工国重实验室科技有限公司 | Engineering machinery software task and personnel recommendation method and system based on word sense weighting |
Non-Patent Citations (3)
| Title |
|---|
| ZHANG QI.ETC: "Analysis of resumption of work based on load curve data", IEEE, 31 December 2021 (2021-12-31) * |
| 李思杰;刘鑫;杨梦慈;林炎锋;赖熙蒙;: "互联网+大学生求职简历信息服务平台建设", 数码世界, no. 06, 1 June 2018 (2018-06-01) * |
| 杨真;陈建安;: "招聘面试人工智能系统的框架与模块研究", 江苏大学学报(社会科学版), no. 06, 30 November 2017 (2017-11-30) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118013245B (en) | 2024-06-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110175851B (en) | Cheating behavior detection method and device | |
| CN113407694A (en) | Customer service robot knowledge base ambiguity detection method, device and related equipment | |
| US10467221B2 (en) | Software function verification system and software function verification method | |
| CN111738589A (en) | Content recommendation-based big data project workload assessment method, device and equipment | |
| CN116719520B (en) | Code generation method and device | |
| CN113256383A (en) | Recommendation method and device for insurance products, electronic equipment and storage medium | |
| CN115082041A (en) | User information management method, device, equipment and storage medium | |
| CN115456166A (en) | Knowledge distillation method for neural network classification model of passive domain data | |
| CN111190973A (en) | Method, device, equipment and storage medium for classifying statement forms | |
| CN112598405B (en) | Business project data management method and system based on big data | |
| CN118013245B (en) | Matching degree evaluation method based on deep learning model | |
| CN117151222B (en) | Domain knowledge guided emergency case entity attribute and relation extraction method thereof, electronic equipment and storage medium | |
| WO2021139255A1 (en) | Model based method and apparatus for predicting data change frequency, and computer device | |
| US9348810B2 (en) | Model learning method | |
| CN117312514A (en) | Consultation reply method, consultation reply device and computer readable storage medium | |
| JP2010272004A (en) | Discriminating apparatus, discrimination method, and computer program | |
| CN113486169B (en) | Synonymous statement generation method, device, equipment and storage medium based on BERT model | |
| CN112308294B (en) | Method and device for predicting default probability | |
| CN114757205A (en) | Text evaluation method, device, equipment and storage medium | |
| CN113469237A (en) | User intention identification method and device, electronic equipment and storage medium | |
| CN113191527A (en) | Prediction method and device for population prediction based on prediction model | |
| CN111400413A (en) | Method and system for determining category of knowledge points in knowledge base | |
| US20230289531A1 (en) | Deep Learning Systems and Methods to Disambiguate False Positives in Natural Language Processing Analytics | |
| JP6879983B2 (en) | Information extraction device, information extraction method and information extraction program | |
| WO2024098282A1 (en) | Geometric problem-solving method and apparatus, and device and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |