Intelligent matching system and method for production and education integration post based on natural language processing

Abstract

The invention belongs to the technical field of post matching, and particularly relates to a production and teaching integrated post intelligent matching system and method based on natural language processing, a post uploading interface and a resume uploading interface are provided, a post specification and a personal resume are uploaded by related personnel, and then the system splits and matches the skills of the post specification and the personal resume. A comprehensive post comprehensive matching model is formed by combining dynamic change analysis and learning ability evaluation of post demand skills, and the comprehensive matching model not only can select students with the skills required by the current post, but also can ensure that the students have the ability of adapting to future technical reform. Therefore, more stable and efficient talent support is provided for enterprises, meanwhile, after the non-conforming resumes are screened out based on the comprehensive matching model, the deficient skills of the non-conforming resumes are recognized and labeled, graduates can clearly know which aspects have defects, and learning and improvement are conducted in a targeted mode.

Description

Technical Field

[0001] The present invention belongs to the technical field of job matching, and specifically relates to an intelligent matching system and method for industry-education integration jobs based on natural language processing. Background Art

[0002] With the acceleration of scientific and technological progress and industrial upgrading, enterprises are increasingly in need of professionals who master cutting-edge technologies and have practical operational capabilities. However, due to its inherent hysteresis, the traditional education model is difficult to adapt to these rapidly changing needs in a timely manner, resulting in a mismatch between graduates' skills and the actual needs of enterprises. In this context, industry-education integration has emerged as an innovative talent training model, aiming to solve the information asymmetry problem between the education system and the industry by building a bridge connecting the two.

[0003] At this stage, the core of industry-education integration is the in-depth cooperation between schools and enterprises in professional fields. Through this cooperation model, enterprises can clearly convey specific job skill requirements to students of relevant majors, so that enterprises can identify and select potential talents with the required skills at an early stage, thereby optimizing their recruitment process. In this process, how to accurately select talents that meet the needs of enterprises based on these job skill requirements is the key link in achieving job matching in industry-education integration.

[0004] There are many technical solutions for industry-education integration job matching in the prior art. For example, the Chinese invention patent with publication number CN110992227B proposes a vocational skills talent training system combining school and enterprise, which includes a standard module, an education module and an industry module. The standard module is responsible for generating new vocational skills standard information; the education module designs practical training courses based on this standard information, and collects students' practical training results and assessment information; the industry module matches this information with job requirements and finally generates talent recommendation results.

[0005] Another technical solution is the Chinese invention patent with publication number CN119295016A, which discloses an online multi-terminal interconnected industry-education integration management system. The system conducts a comprehensive analysis of the job career directions of the cooperating enterprises, evaluates the development trend of the target occupations, and uses the students' target preference data and job status indicators for fitting, matching and screening, thereby recommending jobs and occupations that meet the students' preferences.

[0006] However, the above two schemes mainly focus on the matching of hard indicators such as job skill status when matching production-education integration positions, but ignore the consideration of job learning ability requirements. In fact, the skills required for a position are not fixed, especially in technical positions. With the rapid development of technology, the competency skills required for positions are also constantly updated. Students who are matched and screened only based on current job skills may be competent for their current jobs, but if their learning ability is weak, they may find it difficult to adapt to changes in job skills in the future, which will affect their long-term competency and the sustainability of job recruitment results. Summary of the invention

[0007] The purpose of the present invention is to improve the deficiencies in the prior art and to provide an intelligent matching system and method for industry-education integration positions based on natural language processing, so as to achieve stable adaptive matching of talents required for positions by combining students' skills and learning ability with the skills required for positions and the variability of the skills required for positions in the process of industry-education integration matching.

[0008] The purpose of the present invention can be achieved through the following technical solutions: The first aspect of the present invention proposes an intelligent matching system for industry-education integration positions based on natural language processing, including a data upload module, which is used to provide a position upload interface, a resume upload interface and an assessment upload interface. Enterprises cooperating with the school's target major upload job descriptions on the position upload interface, and graduates of the school's target major upload their resumes on the resume upload interface. At the same time, teachers of the school's target major upload students' course assessment data on the assessment upload interface, specifically the students' assessment scores for each course in each year of study.

[0009] The job requirement splitting module is used to split the uploaded job description into skill requirements using natural language processing, thereby constructing a job requirement skill knowledge graph.

[0010] The resume splitting module is used to split the uploaded resume into skills and learning abilities using natural language processing, thereby constructing a skills knowledge graph and a learning ability knowledge graph, and evaluating students' learning abilities based on the learning ability knowledge graph.

[0011] The skill change analysis module is used to retrieve the job descriptions uploaded each year within a specified period and conduct change analysis on the job requirement skills in the job descriptions.

[0012] The job matching module is used to combine the constructed job requirement skill knowledge graph with the possessed skill knowledge graph, the course assessment data uploaded by the teacher, the change analysis results of the job requirement skills and the students' learning ability assessment results to perform job matching scoring.

[0013] The resume screening and marking module is used to screen out suitable resumes based on the job matching score results of the uploaded resumes, and mark the missing skills of the resumes that do not match the job.

[0014] The second aspect of the present invention proposes an intelligent matching method for industry-education integration positions based on natural language processing, comprising the following steps: S1, providing a position upload interface, a resume upload interface and an assessment upload interface to upload job descriptions, student resumes and student course assessment data respectively.

[0015] S2. Use natural language processing to split the uploaded job description into skill requirements, thereby constructing a job requirement skill knowledge graph.

[0016] S3. Use natural language processing to split the uploaded resume into skills and learning abilities, thereby constructing a skills knowledge graph and a learning ability knowledge graph, and evaluate the student's learning ability based on the learning ability knowledge graph.

[0017] S4. Retrieve the job descriptions uploaded each year within the specified period and conduct a change analysis on the job requirement skills in the job descriptions.

[0018] S5. Combine the constructed job requirement skill knowledge graph with the possessed skill knowledge graph, course assessment data, job requirement skill change analysis results and students’ learning ability assessment results to perform job matching scoring.

[0019] S6. Filter out the matching resumes based on the job matching score results of the uploaded resumes, and mark the missing skills of the resumes that do not match the job.

[0020] Combining all the above technical solutions, the positive effects of the present invention are as follows: 1. By providing a job upload interface and a resume upload interface, relevant personnel can upload job descriptions and personal resumes, and then the system will split and match the job descriptions and personal resume skills, and combine the dynamic change analysis of job requirement skills and learning ability assessment to form a comprehensive job comprehensive matching model. This comprehensive matching model can not only select students with the skills required for the current job, but also ensure that they have the ability to adapt to future technological changes, thereby providing enterprises with more stable and efficient talent support.

[0021] 2. In addition to skill matching based on job descriptions and personal resumes, dynamic change analysis of job-required skills, and learning ability assessment, the present invention also introduces course assessment data when conducting comprehensive job matching. The non-professional course assessment scores provide a multi-dimensional assessment direction for learning ability assessment. The professional course assessment scores in the course assessment data can be used to verify the authenticity and proficiency of the professional skills claimed in the personal resume, thereby significantly improving the accuracy and comprehensiveness of comprehensive job matching.

[0022] 3. After screening out resumes that do not match the requirements based on the comprehensive matching model, the present invention also identifies and marks the missing skills of the resumes, so that graduates can clearly understand their shortcomings and learn and improve in a targeted manner. At the same time, the school can design and implement targeted training plans based on the data on missing skills to help graduates meet job requirements faster. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The present invention is further described using the accompanying drawings, but the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative work.

[0024] Figure 1 It is a schematic diagram of system module connection of the present invention.

[0025] Figure 2 It is a schematic diagram of the job requirement skill knowledge map of the present invention.

[0026] Figure 3 This is a schematic diagram of the learning ability knowledge graph of the present invention.

[0027] Figure 4 It is a diagram of the steps for implementing the method of the present invention. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] Example 1

[0030] The present invention provides an intelligent matching system for industry-education integration positions based on natural language processing, comprising a data uploading module, a position requirement splitting module, a resume splitting module, a skill change analysis module, a position matching module and a resume screening and marking module.

[0031] See also Figure 1 As shown, the data upload module is connected to the job requirement splitting module and the resume splitting module respectively, the job requirement splitting module, the resume splitting module and the skill change analysis module are all connected to the job matching module, and the job matching module is connected to the resume screening and marking module.

[0032] The data upload module is used to provide a job upload interface, a resume upload interface and an assessment upload interface. Enterprises cooperating with the school's target majors upload job descriptions on the job upload interface, and graduates of the school's target majors upload their resumes on the resume upload interface. At the same time, teachers of the school's target majors upload students' course assessment data on the assessment upload interface, specifically the students' assessment scores for each course in each year of study.

[0033] In the implementation of the above scheme, the course assessment data uploading process is as follows: link with the school's student record management system to obtain the student's year of study in the school. The specific operation is to extract the student's admission year and current grade from the school's student record management system to obtain the student's year of study in the school.

[0034] As an example of the above implementation: Assuming a student enrolls in 2020 and the current year is 2025, the student's enrollment years are 2020-2021, 2021-2022, 2022-2023, 2023-2024, and 2024-2025.

[0035] The academic affairs management system of the associated school obtains the course list of the students in each year of study and their corresponding examination scores for uploading.

[0036] The job requirement splitting module is used to split the uploaded job description into skill requirements using natural language processing, thereby constructing a job requirement skill knowledge graph.

[0037] Applied to the above solution, the following process is used to construct a job requirement skill knowledge graph: the text content of the job description is cleaned.

[0038] The cleaning operations mentioned above include but are not limited to removing irrelevant characters and formats, unifying codes and character sets, removing stop words, spelling correction, normalizing abbreviations and terms, removing duplicate information, etc., wherein removing irrelevant characters and formats specifically includes removing HTML tags, special symbols and extra spaces, unifying codes and character sets specifically includes ensuring that the text uses a unified character encoding, removing stop words specifically includes removing common meaningless words and reducing noise, spelling correction specifically includes correcting spelling errors and improving text accuracy, normalizing abbreviations and terms specifically includes normalizing different forms of words to the basic form, and removing duplicate information specifically includes removing repeated parts in the text and reducing redundancy.

[0039] The purpose of the cleaning operation is to ensure the quality and consistency of the data, reduce noise and redundant information, and thus provide a reliable basis for subsequent text analysis and processing.

[0040] The cleaned job description is used to identify the skill requirement entities and extract the skill requirement text. Extracting the skill requirement part helps focus on key information and reduce the impact of noise data on subsequent processing.

[0041] The skill requirement entity recognition mentioned above can specifically use named entity recognition technology to identify skill-related words.

[0042] The intercepted skill requirement text is divided into independent sentences, and each sentence is segmented and tagged with parts of speech.

[0043] It is important to understand that sentence segmentation helps break down long paragraphs into smaller, more manageable units (sentences or phrases) for subsequent analysis, word segmentation breaks sentences into words or phrases, and part-of-speech tagging assigns a grammatical category (such as noun, verb, adjective, etc.) to each word. This information is crucial for understanding sentence structure and meaning.

[0044] Based on the part-of-speech tagging results, nouns are captured from each sentence for skill name recognition, thereby extracting the skill names required for the position and recording the sentences in which they are located.

[0045] It should be pointed out that nouns in sentences usually represent specific concepts or entities, so capturing nouns can effectively identify skill names.

[0046] In the above example, assume that the skill requirement text in a resume is: proficient in Java and Python programming languages.

[0047] Familiarity with data structures and algorithms.

[0048] Ability to use mysql and postgresql for database management.

[0049] The skill names extracted in this example are as follows: Skill name 1: java, python programming language.

[0050] Where to say: Proficient in Java and Python programming languages.

[0051] Skill name 2: data structure, algorithm.

[0052] In the statement: Familiar with data structures and algorithms.

[0053] Skill name 3: mysql, postgresql, database management.

[0054] Statement: Can use mysql and postgresql for database management.

[0055] The surrounding participles of the corresponding participle of each skill name are located from the sentence where the skill name is located, and the participles corresponding to the adjective are captured from them as the demand degree of the skill name.

[0056] It should be noted that adjectives are usually used to describe the characteristics or states of nouns, so capturing adjectives can help determine the specific requirements of skill names. This method can infer the importance or proficiency requirements of skills by analyzing the surrounding modifiers.

[0057] Continuing with the above resume example, the skill names would be in demand as follows: Skill Name 1: java, python, programming language.

[0058] Required level: Proficient.

[0059] Skill name 2: data structure, algorithm.

[0060] Degree of need: Familiarity.

[0061] Skill name 3: mysql, postgresql, database management.

[0062] Degree of need: Able to use.

[0063] The identified skill name is used as a node, and the skill name's demand level is used as the node's edge to construct a job demand skill knowledge graph. Figure 2 shown.

[0064] The present invention uses natural language to parse job requirement skills from job descriptions and uses knowledge graphs to display job requirement skills, which can focus on visually displaying job requirements and provide structured data support for the matching system.

[0065] The resume splitting module is used to split the uploaded personal resume into skills and learning abilities using natural language processing, thereby constructing a skill knowledge graph and a learning ability knowledge graph, and evaluating the student's learning ability based on the learning ability knowledge graph.

[0066] Preferably, the process of constructing a skill knowledge graph is as follows: the text content of the personal resume is cleaned.

[0067] The cleaned resume is used to identify the skilled entities and extract the skilled text.

[0068] The intercepted skill text is divided into independent sentences, and each sentence is segmented and tagged with parts of speech.

[0069] Based on the part-of-speech tagging results, nouns are captured from each sentence to identify the skill names, thereby extracting the skill names and recording the sentences in which they are located.

[0070] From each sentence containing a skill, locate the surrounding participles of the skill name, and capture the adjective-like participles as the mastery degree of the skill.

[0071] The identified skills are taken as a node, and the mastery level of the skills is taken as the edge of the node, thereby constructing a knowledge graph of skills.

[0072] It should be noted that the identification of students' skills in their personal resumes and the knowledge map of their skills can be similarly referred to the knowledge map of skills required for the job.

[0073] The present invention can identify skills from the content of students' personal resumes and construct skill knowledge maps, so as to understand students' skill backgrounds in a targeted manner and display this information visually. This method can match students' skill knowledge maps with job requirement skill knowledge maps in the same framework when matching jobs, thereby significantly improving the efficiency and reliability of matching.

[0074] In another preferred embodiment, the learning ability knowledge graph is constructed as follows: the cleaned personal resume content is matched with the personal resume content using a predefined learning ability keyword list, and the sentences where the learning ability keywords are successfully matched are located and marked as learning ability attribution sentences.

[0075] The above-mentioned learning ability keyword list includes "learning", "adaptation", "improvement", etc., and these keywords are identified from personal resumes through text matching technology.

[0076] Capture learning achievements from learning ability attribution statements and extract achievement descriptions.

[0077] The learning ability keywords matched in the learning ability attribution sentence are used as nodes, and the extracted learning achievement descriptions are used as the edges of the nodes to construct a learning ability knowledge graph. Figure 3 shown.

[0078] In the example of the above operation, suppose there is a simplified resume text as follows: Possess the ability to learn quickly and adapt well to changes. Have achieved remarkable achievements in projects and successfully completed multiple complex tasks.

[0079] Excellent teamwork skills and the ability to quickly master new skills and apply them to real work

[0080] First, we need to clean the resume text to ensure the consistency and quality of the data.

[0081] Cleaned resume text: Have the ability to learn quickly and adapt well to changes. Have achieved remarkable results in projects. Successfully completed multiple complex tasks. Perform well in teamwork. Can quickly master new skills and apply them in actual work.

[0082] Next, we need to extract sentences related to learning ability from the resume and mark the sentences containing the learning ability keyword as learning ability attribution sentences.

[0083] Learning ability attribution statement: Have the ability to learn quickly and adapt well to changes, can perform well in teamwork, can quickly master new skills and apply them in actual work.

[0084] Specific achievement descriptions were captured from learning ability attribution statements: [‘Has achieved significant results in projects and successfully completed multiple complex tasks’, ‘Performed well in teamwork and was able to quickly master new skills and apply them to actual work’].

[0085] By capturing learning ability content from students' personal resumes and constructing a learning ability knowledge graph, the present invention can comprehensively evaluate students' learning ability and potential, and provide data support for subsequent quantitative evaluation of learning ability.

[0086] Further preferably, the learning ability of students is evaluated based on the learning ability knowledge graph, referring to the following process: defining learning ability evaluation indicators, specifically including the number of learning ability keywords, the strength of the relationship between learning ability and achievement, and the quality of learning ability achievement.

[0087] It should be pointed out that the number of learning ability keywords is the number of nodes in the learning ability knowledge graph. The strength of the relationship between learning ability and achievement refers to the number of edges that exist in the node to which the learning ability keyword belongs. If a learning ability node is connected to multiple achievement edges, it means that this learning ability has played an important role in multiple different situations, so its relationship strength is high. The learning ability achievement quality refers to the achievement expression quality of the edge corresponding to the node to which the learning ability keyword belongs.

[0088] The number of nodes in the learning ability knowledge graph and the number of edges associated with each node are counted, and the number of words in the achievement description of each edge is counted. The quality of each edge is then calculated by calculating the ratio with the total number of words in the text of the personal resume.

[0089] It is important to understand that the space on a resume is limited, so job seekers tend to show as many of their most important achievements and skills as possible within the limited space. A longer description may mean that the achievement has a higher information density and value. By comparing the number of words in each achievement description with the word count of the entire resume, it can reflect the relative importance of the achievement in the overall resume. Generally speaking, if an achievement is described in detail, it may mean that it is more important or more representative to the job seeker.

[0090] The number of nodes in the learning ability knowledge graph is taken as the number of learning ability keywords, the number of edges associated with each node is taken as the strength of the relationship between learning ability and achievement, and the mean quality of each edge associated with the node is taken as the quality of learning ability achievement.

[0091] Substitute the number of learning ability keywords, the strength of the relationship between learning ability and achievement, and the quality of learning ability achievement into the evaluation formula Get the learning achievement coefficient of your resume , where Indicates the number of learning ability keywords, Indicates the number of learning ability keywords in the learning ability keyword list. Indicates The learning ability keyword corresponds to the strength of the relationship between learning ability and achievement. Indicates the number of the learning ability keyword in the learning ability knowledge graph, , Indicates The quality of learning ability achievement corresponding to the learning ability keyword.

[0092] From the above evaluation formula of learning ability achievement coefficient, it can be seen that the number of learning ability keywords in personal resume, the strength of the relationship between learning ability and achievement, and the quality of learning ability achievement all have a positive impact on the learning ability achievement coefficient.

[0093] The above-mentioned learning ability assessment through the three dimensions of the number of learning ability keywords, the strength of the relationship between learning ability and achievement, and the quality of learning ability achievements can achieve a comprehensive assessment of students' learning ability achievements in their personal resumes.

[0094] The assessment scores corresponding to each course of the students in each year of study are extracted from the uploaded course assessment data, and the assessment scores of the corresponding courses of non-professional skill courses are screened out.

[0095] The non-professional learning ability assessment coefficient is obtained by comparing the assessment score of the corresponding course of the non-professional skill course with the full score.

[0096] The student's learning ability coefficient is obtained by taking the weighted average of the learning ability achievement coefficient of the personal resume and the non-professional learning ability assessment coefficient.

[0097] Exemplarily, the weight values ​​of the learning ability achievement coefficient and the non-professional learning ability assessment coefficient can be 0.7 and 0.3.

[0098] It should be added that non-professional courses are often not directly related to students' majors, so students' performance in these courses can better reflect their learning attitudes and habits. When students have high assessment scores in non-professional courses, it means that students are willing to spend time and energy to learn and master this knowledge. In addition, non-professional courses often involve different ways of thinking and learning methods. Being able to get high scores in these courses shows that students have good adaptability and flexibility, and can quickly adjust their learning strategies to adapt to new learning environments and requirements, thus reflecting strong learning ability.

[0099] The skill change analysis module is used to retrieve the job descriptions uploaded each year within a specified period of time, and perform change analysis on the job requirement skills in the job descriptions. The specific analysis process is as follows: job requirement skills are extracted from the job descriptions uploaded each year, and a job requirement skills list is formed, and the total number of job requirement skills in the job requirement skills list of each year is counted.

[0100] It should be added that the specified period should be limited to the current time. For example, job descriptions that are 5 to 6 years away from the current year can be selected to provide a sufficient sample size to ensure that the results of the analysis of changes in job skill requirements are representative and reliable.

[0101] Compare the job requirement skills uploaded each year one by one, count the number of changed job requirement skills, and obtain the number of years of continuous change for each changed job requirement skill.

[0102] It should be added that the skills required for changing job requirements refer to those skills that change between different years. The changes may include adding, removing or modifying (for example, changing from "skilled" to "proficient"). The number of years of continuous change in the skills required for changing job requirements refers to the span of years from the first change of the skill to the last change (including the start and end years). If a skill has undergone changes in multiple consecutive years, the number of these years is the number of years of continuous change of the skill.

[0103] By formula Get the skill change rate of job requirements , where Indicates The number of years of continuous change in skills required for changing positions, The number indicating the skills required for the changed position. , Indicates the number of years within the specified period. Represents the total number of skills required for a job.

[0104] It should be understood that from the above formula for the variability of job skill requirements, the more the number of changing job skill requirements, the longer the number of years in which the changing job skill requirements continue to change, and the greater the variability of the job skill requirements. The number of changing job skill requirements refers to the number of skill items that are added, removed or modified each year. The larger the number, the more changing the job requirements for employees are, and the longer the number of years in which the changing job skill requirements continue to change, the more important or demanding the skills are evolving. All these reflect the variability of job skill requirements.

[0105] The job matching module is used to combine the constructed job requirement skill knowledge graph and the skill possession knowledge graph with the course assessment data uploaded by the teacher, the change analysis results of the job requirement skills and the student's learning ability assessment results to perform job matching scoring. The specific operation is as follows: match the student's skill possession knowledge graph with the job requirement skill knowledge graph for the same skills, and identify the successfully matched skills.

[0106] It needs to be explained that the successful matching mentioned above means that the skill knowledge graph is consistent with the skill knowledge graph required for the position at a certain node.

[0107] Based on the successfully matched skills, the assessment scores corresponding to each course of the students in each year of study are extracted from the uploaded course assessment data, and the assessment scores of the courses associated with the successfully matched skills are screened out.

[0108] The successfully matched skills are matched at the edge of the skill knowledge graph and the job requirement skill knowledge graph, and the matching degree is calculated.

[0109] In the above, when analyzing the matching degree of successfully matched skills, it is considered that the edges between the possessed skill knowledge graph and the job requirement skill knowledge graph are degree terms such as possessed degree and required degree. The possessed degree and required degree of the successfully matched skills can be quantified first, so that the matching degree between personal skills and job requirements can be calculated more accurately. The specific quantification process is: First, it is necessary to define and standardize the quantitative values ​​of degree terms. Common degree terms may include "basic", "skilled", "proficient", etc. A numerical range can be assigned to each degree term based on natural language processing. For example: basic: 0.3, skilled: 0.6, proficient: 0.9, and then the quantified possessed degree and required degree are subtracted, and the expression is used. , where Indicates the matching degree, Indicates the degree of quantification. Indicates the quantified degree of demand, Represents a natural constant.

[0110] Statistics on the number of successful matching skills , and combine it with the assessment scores and matching degree of each successfully matched skill-related course to calculate the skill matching coefficient The specific statistical formula is , Indicates the number of job requirement skills in the job requirement skill knowledge graph, Indicates Matching success skill matching degree, Indicates the number of the successfully matched skill. , Indicates The assessment scores of the associated courses that match the successful skills, Indicates the full score.

[0111] The learning ability coefficient and skill matching coefficient are weighted according to the variability of job requirements.

[0112] Specifically, the job skill variability can be compared with a pre-configured critical variability range. For example, the critical variability range is If the job requirement skill variability is greater than the upper limit of the critical variability range, it indicates that the job has a high requirement for learning ability. At this time, a higher weight should be given to the learning ability coefficient to reflect the importance of employees' ability to quickly adapt to new skills. The weight values ​​of the learning ability coefficient and the skill matching coefficient can be set to 0.6 and 0.4. If the job requirement skill variability is within the critical variability range, it indicates that the job has a relatively balanced demand for learning ability and the matching of existing skills. The weight values ​​of the learning ability coefficient and the skill matching coefficient can be set to 0.5 and 0.5. If the job requirement skill variability is less than the lower limit of the critical variability range, it indicates that the job has a low requirement for learning ability, and the matching of existing skills is more important. At this time, a higher weight should be given to the skill matching coefficient. The weight values ​​of the learning ability coefficient and the skill matching coefficient can be set to 0.4 and 0.6.

[0113] The job matching score is obtained by combining the skill matching coefficient and the learning ability coefficient with the corresponding weights and performing weighted average calculation.

[0114] The resume screening and marking module is used to screen out suitable resumes based on the job matching score results of the uploaded resumes, and mark the lacking skills of the resumes that do not meet the requirements.

[0115] The specific process of screening out eligible resumes in the above is as follows: compare the job matching score results of the uploaded resumes with the set standard scores, and screen out resumes that meet the standard scores as eligible resumes.

[0116] To be more specific, the process of marking the missing skills of the non-matching resumes is as follows: from all uploaded resumes, resumes other than the matching resumes are selected as non-matching resumes, and then the skills that are not successfully matched in the matching of the skill knowledge graph of the non-matching resumes and the job requirement skill knowledge graph are obtained as missing skills, and the missing skills are marked in the non-matching resumes.

[0117] Example 2

[0118] Reference Figure 4 As shown, the present invention proposes an intelligent matching method for industry-education integration positions based on natural language processing, comprising the following steps: S1, providing a position upload interface, a resume upload interface and an assessment upload interface to upload job descriptions, student resumes and student course assessment data respectively.

[0119] S2. Use natural language to split the uploaded job description into skill requirements, thereby constructing a job requirement skill knowledge graph.

[0120] S3. Use natural language processing to split the uploaded resume into skills and learning abilities, thereby constructing a skills knowledge graph and a learning ability knowledge graph, and evaluate the student's learning ability based on the learning ability knowledge graph.

[0121] S4. Retrieve the job descriptions uploaded each year within the specified period and conduct a change analysis on the job requirement skills in the job descriptions.

[0122] S5. Combine the constructed job requirement skill knowledge graph with the possessed skill knowledge graph, course assessment data, job requirement skill change analysis results and students’ learning ability assessment results to perform job matching scoring.

[0123] S6. Filter out the matching resumes based on the job matching score results of the uploaded resumes, and mark the missing skills of the resumes that do not match the job.

[0124] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the present invention, they should all fall within the protection scope of the present invention.

Claims

1. An intelligent matching system for industry-education integration positions based on natural language processing, characterized by: Includes the following modules: The data upload module is used to provide job upload interface, resume upload interface and assessment upload interface. Enterprises cooperating with the school's target majors upload job descriptions on the job upload interface, and graduates of the school's target majors upload their resumes on the resume upload interface. At the same time, teachers of the school's target majors upload students' course assessment data on the assessment upload interface, specifically the assessment scores of students for each course in each year of study; The job requirement splitting module is used to split the uploaded job description into skill requirements using natural language processing, thereby constructing a job requirement skill knowledge graph; The resume splitting module is used to split the uploaded resume into skills and learning ability using natural language processing, thereby constructing a skills knowledge graph and a learning ability knowledge graph, and evaluating the student's learning ability based on the learning ability knowledge graph; Skill change analysis module, which is used to retrieve job descriptions uploaded every year within a specified period of time and conduct change analysis on job requirement skills in job descriptions; The job matching module is used to combine the constructed job requirement skill knowledge graph with the skill knowledge graph, the course assessment data uploaded by the teacher, the change analysis results of job requirement skills and the student's learning ability assessment results to perform job matching scoring; The resume screening and marking module is used to screen out suitable resumes based on the job matching score results of the uploaded resumes, and mark the missing skills of the resumes that do not match the job.

2. The intelligent matching system for industry-education integration positions based on natural language processing as claimed in claim 1, characterized in that: The process of constructing the job requirement skill knowledge graph is as follows: Clean the text content of the job description; Perform skill requirement entity recognition on the cleaned job description and extract the skill requirement text; Segment the intercepted skill requirement text into independent sentences, and perform word segmentation and part-of-speech tagging on each sentence; Based on the part-of-speech tagging results, nouns are captured from each sentence to identify skill names, thereby extracting the skill names required for the position and recording the sentences in which they are located; Locate the surrounding participles of the skill name in the sentence where each skill name is located, and capture the adjective-like participles as the demand degree of the skill name; The identified skill name is taken as a node, and the demand degree of the skill name is taken as the edge of the node, thereby constructing a knowledge graph of job requirement skills.

3. The intelligent matching system for industry-education integration positions based on natural language processing as claimed in claim 2, characterized in that: The process of constructing a skill knowledge graph is as follows: Clean the text content of the resume; Perform skill entity recognition on the cleaned resume and extract the skill text; Segment the intercepted skill text into independent sentences, and perform word segmentation and part-of-speech tagging on each sentence; Based on the part-of-speech tagging results, nouns are captured from each sentence to identify the skill names, thereby extracting the skill names and recording the sentences in which they are located; Locate the surrounding participles of the skill name in each sentence, and capture the adjective participles as the mastery degree of the skill; The identified skills are taken as a node, and the mastery level of the skills is taken as the edge of the node, thereby constructing a knowledge graph of skills.

4. The intelligent matching system for industry-education integration positions based on natural language processing as claimed in claim 3, characterized in that: The learning ability knowledge graph is constructed as follows: The cleaned resume content is matched with the predefined learning ability keyword list, and the sentences with the successfully matched learning ability keywords are located and marked as learning ability attribution sentences; Capture learning achievements from learning ability attribution statements and extract achievement descriptions; The learning ability keywords matched in the learning ability attribution sentence are taken as nodes, and the extracted learning achievement descriptions are taken as the edges of the nodes, thereby constructing a learning ability knowledge graph.

5. The intelligent matching system for industry-education integration positions based on natural language processing as claimed in claim 4, characterized in that: The learning ability assessment is described in the following process: Define learning ability assessment indicators, including the number of learning ability keywords, the strength of the relationship between learning ability and achievement, and the quality of learning ability achievement; The number of nodes in the learning ability knowledge graph and the number of edges associated with each node are counted as the number of learning ability keywords and the strength of the relationship between learning ability and achievement; The word count of the achievement description of each edge is counted, and the ratio of the word count to the total word count of the text in the resume is calculated to get the quality of each edge. Then the average quality of each edge associated with the node is taken as the learning ability achievement quality. Substitute learning ability assessment indicators into the assessment formula Get the learning achievement coefficient of your resume , where Indicates the number of learning ability keywords, Indicates the number of learning ability keywords in the learning ability keyword list. Indicates The learning ability keyword corresponds to the strength of the relationship between learning ability and achievement. Indicates the number of the learning ability keyword in the learning ability knowledge graph, , Indicates The quality of learning ability achievement corresponding to the learning ability keywords; Extract the assessment scores of each course in each year of study from the uploaded course assessment data, and screen out the assessment scores of the corresponding courses of non-professional skill courses, and compare them with the full score to obtain the non-professional learning ability assessment coefficient; The student's learning ability coefficient is obtained by taking the weighted average of the learning ability achievement coefficient of the personal resume and the non-professional learning ability assessment coefficient.

6. The intelligent matching system for industry-education integration positions based on natural language processing as claimed in claim 2, characterized in that: The process of analyzing the changes in the job requirements in the job description is as follows: Extract job requirement skills from the job descriptions uploaded each year and form a job requirement skill list. At the same time, count the total number of job requirement skills in the job requirement skill list for each year. Compare the job requirements and skills uploaded each year one by one, count the number of skills that have changed, and obtain the number of years in which the skills required for each job have continued to change; By formula Get the skill change rate of job requirements , where Indicates The number of years of continuous change in skills required for changing positions, The number indicating the skills required for the changed position. , Indicates the number of years within the specified period. Represents the total number of skills required for a job.

7. The intelligent matching system for industry-education integration positions based on natural language processing as claimed in claim 5, characterized in that: The job matching score is as follows: Match the students' skill knowledge graph with the job requirement skill knowledge graph to identify the successfully matched skills; Based on the matched successful skills, the assessment scores of the students for each course in each year of study are extracted from the uploaded course assessment data, and the assessment scores of the courses associated with the matched successful skills are screened out; Match the successfully matched skills at the edge of the skill knowledge graph and the job requirement skill knowledge graph, and calculate the matching degree; Statistics on the number of successful matching skills , and combine it with the assessment scores and matching degree of each successfully matched skill-related course to calculate the skill matching coefficient The specific statistical formula is , Indicates the number of job requirement skills in the job requirement skill knowledge graph, Indicates Matching success skill matching degree, Indicates the number of the successfully matched skill. , Indicates The assessment scores of the associated courses that match the successful skills, Indicates the full score; Use the job requirement skill change degree to assign weights to the learning ability coefficient and skill matching coefficient; The skill matching coefficient and learning ability coefficient are combined with corresponding weights to perform job matching scoring.

8. The intelligent matching system for industry-education integration positions based on natural language processing as claimed in claim 7, characterized in that: The process of screening out resumes that meet the requirements is as follows: The job matching score of the uploaded resume is compared with the set standard score, and the resumes that meet the standard score are selected as qualified resumes.

9. The intelligent matching system for industry-education integration positions based on natural language processing as claimed in claim 8, characterized in that: The implementation of the marked lack of skills that do not match the resume is as follows: From all uploaded resumes, select resumes other than the qualified resumes as non-qualified resumes, and then obtain the skills that are not successfully matched in the matching of the skill knowledge graph of the non-qualified resumes and the job requirement skill knowledge graph as missing skills, and mark the missing skills in the non-qualified resumes.

10. An intelligent matching method for industry-education integration positions based on natural language processing, characterized in that: The following steps are involved: S1. Provide job upload interface, resume upload interface and assessment upload interface to upload job description, student resume and student course assessment data respectively; S2. Use natural language processing to split the uploaded job description into skill requirements, thereby constructing a job requirement skill knowledge graph; S3. Use natural language processing to split the uploaded resume into skills and learning ability, thereby constructing a skills knowledge graph and a learning ability knowledge graph, and evaluate the student's learning ability based on the learning ability knowledge graph; S4. Retrieve the job descriptions uploaded each year within the specified period and conduct a change analysis on the job requirement skills in the job descriptions; S5. Combine the constructed job requirement skill knowledge graph and the possessed skill knowledge graph with the course assessment data, the change analysis results of job requirement skills and the students’ learning ability assessment results to perform job matching scoring; S6. Filter out the matching resumes based on the job matching score results of the uploaded resumes, and mark the missing skills of the resumes that do not match the job.

Images