Two-dimensional code-based resume and interview content matching method and system

Abstract

The invention relates to the field of resume matching, in particular to a resume and interview content matching method and system based on a two-dimensional code. According to the method, the plurality of resumes corresponding to the recruitment post are obtained, the candidate resumes are determined based on the similarity merging resumes, whether the text description of the candidate resumes meets the preset demand association condition or not is further judged, and then the labels are set for the candidate resumes. And further determining historical stable characteristics of the historical invitation resumes and the demand keywords, and performing periodic coverage verification to determine a stable demand keyword set. And generating a stable priority sequence for sequence recommendation based on the stable demand keyword set and the stable association value described by the candidate text, or judging whether the stable priority sequence has an association relationship with the demand keyword or not, and determining whether a preset matching condition is met or not so as to judge whether sequence recommendation is performed or not. Finally, the electronic resume with the label and the sequence corresponding to the label can be presented by scanning the two-dimensional code, so that the resume screening efficiency and accuracy are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of resume matching, in particular to a method and system for matching job-hunting resumes and interview content based on a two-dimensional code. BACKGROUND

[0002] With the rapid development of big data, artificial intelligence and natural language processing technology, modern talent matching systems are not only closely related to the intelligent processing capability of massive unstructured data, but also closely connected with high-concurrency, low-latency real-time search architecture. Efficient and accurate resume intelligent screening has become an indispensable part of the digital transformation of enterprise human resources.

[0003] Chinese Patent Publication No. CN116976646A provides an information recognition method, device, electronic equipment and storage medium. The method includes: obtaining a connection behavior table corresponding to the interaction of the target position by the job seeker and the recruitment party in the target dimension, the target dimension at least including the conversation dimension, the resume dimension and the interview dimension; determining the conversation intention table based on the conversation content of the job seeker and the recruitment party; generating an interaction behavior sequence according to the connection behavior table and the conversation intention table, the interaction behavior sequence including a plurality of interaction degree labels corresponding to the interaction of the job seeker and the recruitment party; determining whether the job seeker and the recruitment party reach a mutual matching intention for the target position based on the plurality of interaction degree labels. This application can introduce more and deeper connection behaviors while considering the hysteresis between behaviors, reasonably analyze the interaction behavior sequence of the job seeker and the recruitment party, measure the effective connection of the job seeker and the recruitment party in multiple dimensions based on the recruitment platform, and ensure relatively accurate identification of the degree of mutual achievement.

[0004] However, the prior art has the following problems, In a modern talent recruitment and resume screening system, a large number of resumes may accumulate instantaneously for a popular position, and human resource personnel need to invest a lot of time in repetitive manual browsing and preliminary screening, which may result in insufficient screening accuracy and prolonged recruitment cycle, thereby reducing the overall operational efficiency of the recruitment process. SUMMARY

[0005] Therefore, the present application provides a method and system for matching job-hunting resumes and interview content based on a two-dimensional code to overcome the problem in the prior art that in a modern talent recruitment and resume screening system, a large number of resumes may accumulate instantaneously for a popular position, and human resource personnel need to invest a lot of time in repetitive manual browsing and preliminary screening, which may result in insufficient screening accuracy and prolonged recruitment cycle, thereby reducing the overall operational efficiency of the recruitment process.

[0006] To achieve the above-mentioned purpose, the present application provides a method and system for matching job-hunting resumes and interview content based on a two-dimensional code, which includes, Obtaining a plurality of resumes corresponding to the recruitment position, extracting text descriptions of the resumes, determining the similarity between the text descriptions, merging the resumes based on the similarity, and determining a plurality of candidate resumes; Obtaining a demand text description of the recruitment position and extracting demand keywords, obtaining a candidate text description corresponding to the candidate resume, determining whether the candidate text description meets a preset demand association condition, and setting a label for the candidate resume, including an obviously matching tendency resume and a fuzzy matching tendency resume; For the obviously matching tendency resume, a historical stable feature of the historical invitation resume and the demand keywords is determined to determine whether to perform periodic coverage verification on the demand keywords, a stable demand keyword set is selected based on the verification result, a stable association value between the stable demand keyword set and the candidate text description is calculated, and a stable priority sequence is generated according to the stable association value, and the obviously matching tendency resume is recommended in sequence; For the fuzzy matching tendency resume, it is determined whether the fuzzy matching tendency resume is associated with the demand keywords to determine whether to select a paragraph in which the demand keywords are located, the paragraph is observed for matching, it is determined whether a preset matching condition is met to determine whether to recommend the fuzzy matching tendency resume in sequence; A corresponding two-dimensional code is generated for the candidate resume that is recommended for recommendation, the two-dimensional code is connected to a visual dynamic interface, and the electronic resume with the label and the sequence corresponding to the label is presented by scanning the corresponding two-dimensional code.

[0007] Further, the process of determining a plurality of candidate resumes includes, The semantic similarity between the text descriptions of the resumes is determined as the similarity; If the similarity is greater than or equal to a preset similarity threshold, the resumes are merged, and the merged resumes are determined as candidate resumes.

[0008] Further, the process of extracting demand keywords includes, Determining the nouns and noun phrases in the text description corresponding to the recruitment position; The nouns and noun phrases are determined as demand keywords.

[0009] Further, the process of determining whether the candidate text description meets a preset demand association condition includes, Determining the text semantic association degree between the demand text description and the candidate text description; If a preset proportion of demand keywords falls into the candidate text description and the text semantic association degree is greater than or equal to a preset text semantic association threshold, it is determined that the candidate text description meets the preset demand association condition.

[0010] Further, the process of setting a label for the candidate resume includes, If the candidate text description meets the preset requirement association condition, the candidate resume is determined as an obviously matching tendency resume; If the candidate text description does not meet the preset requirement association condition, the candidate resume is determined as a fuzzy matching tendency resume.

[0011] Further, the process of determining the historical stability feature of the historical invitation resume and the requirement keyword to determine whether to perform periodic coverage verification on the requirement keyword includes, Obtaining historical invitation resumes corresponding to the recruitment post in a predetermined period; Determining the ratio of the number of historical invitation resumes containing the requirement keyword to the total number of historical invitation resumes as the historical stability feature; If the historical stability feature is greater than or equal to a preset historical stability threshold, performing periodic coverage verification on the requirement keyword; The periodic coverage verification includes dividing the predetermined period into a plurality of recruitment time periods and verifying whether the requirement keyword appears in a predetermined proportion of the recruitment time periods.

[0012] Further, the process of selecting a stable requirement keyword set based on the verification result and calculating a stable association value of the stable requirement keyword set and the candidate text description includes, If the periodic coverage verification is passed, the requirement keyword is selected as a stable requirement keyword and is included in the stable requirement keyword set; Determining each stable requirement keyword in the stable requirement keyword set; Determining the number of each stable requirement keyword in each candidate text description; Determining the number as the stable association value.

[0013] Further, the process of determining whether to select the paragraph where the requirement keyword is located, performing matching observation on the paragraph, and determining whether the preset matching condition is met includes, If the fuzzy matching tendency resume contains any requirement keyword, it is determined that the fuzzy matching tendency resume has an association relationship with the requirement keyword, and the paragraph where the requirement keyword is located is selected; Determining the candidate text description corresponding to the fuzzy matching tendency resume; Determining the semantic association degree of the paragraph and the requirement text description; If the semantic association degree is greater than or equal to a preset semantic association threshold, the paragraph is marked; determining a ratio of a total length of texts of all marked paragraphs to a total length of candidate text descriptions; The matching condition is that the ratio needs to be greater than or equal to a predetermined text length proportion threshold.

[0014] Further, the process of determining whether to perform sequence recommendation on the fuzzy matching tendency resume includes, If the matching condition is met, the fuzzy matching tendency resume is performed sequence recommendation.

[0015] In another aspect, the application provides a two-dimensional code-based matching system for resumes and interview content, comprising: The acquisition module is used to obtain a plurality of resumes corresponding to a recruitment position, extract text descriptions of the resumes, determine the similarity between the text descriptions, merge the resumes based on the similarity, and determine a plurality of candidate resumes; The label setting module is connected with the acquisition module and is used to obtain a demand text description of a recruitment position and extract demand keywords, obtain candidate text descriptions corresponding to the candidate resumes, determine whether the candidate text descriptions meet a preset demand association condition, and set labels for the candidate resumes, including obviously matching tendency resumes and fuzzy matching tendency resumes; The obvious matching analysis module is connected with the label setting module and is used to determine historical stable features of a historical invitation resume and demand keywords for the obviously matching tendency resumes, determine whether to perform periodic coverage verification on the demand keywords, select a stable demand keyword set based on the verification result, calculate a stable association value of the stable demand keyword set and the candidate text descriptions, generate a stable priority sequence based on the stable association value, and perform sequence recommendation on the candidate resumes; The fuzzy matching analysis module is connected with the obvious matching analysis module and is used to determine whether the fuzzy matching tendency resumes are associated with the demand keywords for the fuzzy matching tendency resumes, determine whether to select a paragraph where the demand keywords are located, perform matching observation on the paragraph, determine whether a preset matching condition is met, and determine whether to perform sequence recommendation on the candidate resumes; The visualization module is connected with the acquisition module, the label setting module, the obvious matching analysis module, and the fuzzy matching analysis module, respectively, and is used to generate a corresponding two-dimensional code for the recommended candidate resumes, connect the two-dimensional code to a visual dynamic interface, and scan the corresponding two-dimensional code to present an electronic resume with a label and a sequence corresponding to the label.

[0016] Compared with the prior art, the present application obtains a plurality of job resumes corresponding to the recruitment position, determines the candidate job resume based on the similarity of the resumes, further determines whether the candidate resume text description meets the preset demand correlation condition, and then sets a label for the candidate resume. Further determine the historical stability characteristics of the historical invitation resume and the demand keywords, perform periodic coverage verification to determine the stable demand keyword set. Based on the stable demand keyword set and the stable correlation value of the candidate text description, generate a stable priority sequence for sequence recommendation, or determine whether it is associated with the demand keywords to determine whether it meets the preset matching condition to determine whether to perform sequence recommendation. Finally, scanning the two-dimensional code can present an electronic resume with a label and a sequence corresponding to the label, thereby improving the efficiency and accuracy of resume screening.

[0017] Especially, the present application obtains the candidate text description corresponding to the candidate job resume, determines whether the candidate text description meets the preset demand correlation condition, and sets a label for the candidate job resume. In the actual resume screening scene, since the demand keywords of the recruitment position are the direct embodiment of the core requirements of the position, and the text semantic correlation degree can measure the semantic fit degree of the candidate resume and the job demand, using a single determination standard may lead to one-sided screening results, for example, it may lead to candidate resumes with piled-up screening keywords but inconsistent actual experience, or it may miss high-quality resumes that precisely have core skills but have large differences in expression from the recruitment text. Therefore, if the preset proportion of demand keywords falls into the candidate text description and the text semantic correlation degree is greater than or equal to the preset text semantic correlation threshold, it is determined that the candidate text description meets the preset demand correlation condition. Through this double determination mechanism, resumes that are highly matched with the position can be more accurately screened, and they are divided into obviously matching tendency job resumes and fuzzy matching tendency job resumes, providing a reliable basis for subsequent differentiated analysis strategies for candidate job resumes, thereby improving the efficiency and accuracy of resume screening.

[0018] In particular, this invention targets resumes with a clear matching tendency, determines the historical stable features of historically invited resumes and required keywords, and determines whether to perform periodic coverage verification of the required keywords. Based on the verification results, a stable set of required keywords is selected, and the stable association value between the stable set of required keywords and candidate text descriptions is calculated. A stable priority sequence is generated based on the stable association value, and the resumes with a clear matching tendency are recommended sequentially. In actual resume screening scenarios, since the keyword co-occurrence patterns reflected in historically invited resumes can accurately reflect the core elements of job requirements, historical stable features are determined based on the ratio of the number of historically invited resumes containing specific required keywords to the total number of invited resumes within a predetermined period. This is used to determine whether the keyword has general representativeness. However, since statistical results within a single time span may be affected by temporary or periodic factors such as specific recruitment batches or market hotspots, some temporary and non-core requirements may be misjudged as stable requirements. For example, for the "event execution" position, which is recruited in large numbers in a short period to cope with a temporary technology summit, the keyword "exhibition planning" will have an abnormally high historical stable feature in the current period, but it is not necessarily a stable requirement. The core competencies for this position are determined over a long period. Therefore, when the historical stability characteristics exceed the preset historical stability threshold, the required keywords are verified through periodic coverage. This involves dividing the predetermined period into multiple consecutive recruitment time periods and checking whether the required keywords appear consistently and stably in multiple time periods exceeding a preset proportion. This allows for the selection of stable required keywords that truly span the recruitment cycle and represent stable job requirements. A set of stable required keywords is then constructed, ensuring that the generated priority sequence is based on long-term stable job requirements rather than short-term fluctuations. Based on the proportion of stable required keywords in resumes with a clear matching tendency, the sequence recommendation prioritizes candidate resumes containing stable required keywords, thereby improving the efficiency and accuracy of resume screening.

[0019] In particular, this invention targets resumes with a tendency towards fuzzy matching. It determines whether the resume is related to the required keywords, thereby determining whether to select the paragraph containing the required keywords. The paragraph is then matched and observed to determine if it meets preset matching conditions, thus determining whether to perform sequence recommendation on the resume. In actual resume screening scenarios, although resumes with a tendency towards fuzzy matching may have insufficient overall semantic relevance to the job requirements or fail to meet a preset proportion of required keywords falling into the candidate text description, some paragraphs may contain in-depth information highly relevant to the core requirements. For example, a candidate's resume might be categorized as such because their main technology stack is Python and they are in the fintech field, which deviates significantly from the overall description of the target Java backend position. Fuzzy matching; however, its detailed description of the key paragraph on "low-latency optimization of high-concurrency transaction systems" in a core project profoundly reflects the core capabilities of "high concurrency" and "performance optimization" that are crucial to the "backend development expert" position. Although the context and some technical keywords are different, it still has a potential matching tendency. Therefore, if the resume with the fuzzy matching tendency contains any requirement keyword, it is determined that the resume with the fuzzy matching tendency is related to the requirement keyword. The paragraph is matched and observed. If the matching conditions are met, it means that the candidate's resume has a certain suitability. Then, sequence recommendation is performed. This refined screening strategy based on local semantic analysis can discover potential suitable talents, avoid misscreening, and thus improve the efficiency and accuracy of resume screening. Attached Figure Description

[0020] Figure 1 A schematic diagram illustrating the steps of a method for matching job application resumes with interview content based on QR codes, as an embodiment of the invention; Figure 2 A logic diagram for setting tags on candidate job application resumes according to an embodiment of the invention; Figure 3 This is a logic diagram for determining whether to perform periodic coverage verification on the required keywords in an embodiment of the invention. Figure 4 This is a logic diagram illustrating whether to perform sequence recommendation on job resumes that are prone to fuzzy matching, as described in an embodiment of the invention. Detailed Implementation

[0021] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0022] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0023] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the term "connection" should be interpreted broadly. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0024] Please see Figure 1 The diagram illustrates the steps of a QR code-based method for matching job resumes with interview content, as described in an embodiment of the invention. The method includes: Obtain several job application resumes corresponding to the job postings, extract the text descriptions from the job application resumes, determine the similarity between the text descriptions, merge the job application resumes based on the similarity, and determine several candidate job application resumes. Obtain the job posting requirements text description and extract the requirements keywords, obtain the candidate text description corresponding to the candidate job application resume, determine whether the candidate text description meets the preset requirements association conditions, and set tags for the candidate job application resume, including job application resumes with obvious matching tendency and job application resumes with fuzzy matching tendency. For resumes with obvious matching tendencies, the historical stable features of historical invitation resumes and demand keywords are determined to determine whether to perform periodic coverage verification of demand keywords. Based on the verification results, a stable demand keyword set is selected, and the stable association value between the stable demand keyword set and the candidate text description is calculated. A stable priority sequence is generated based on the stable association value, and the sequence recommendation is performed on the resumes with obvious matching tendencies. For job resumes with a tendency to match fuzzy matching, determine whether the job resumes with a tendency to match fuzzy matching are related to the required keywords, in order to determine whether to select the paragraph containing the required keywords, perform matching observation on the paragraphs, determine whether to meet the preset matching conditions, and then determine whether to perform sequence recommendation on the job resumes with a tendency to match fuzzy matching. A corresponding QR code is generated for each recommended candidate resume. The QR code is then connected to a visual dynamic interface. Scanning the QR code displays an electronic resume with tags and their corresponding sequences.

[0025] In practice, this can be achieved by accessing the data interface of a third-party recruitment platform or parsing the stored content of the company's own talent database. As long as the complete resume documents submitted by job seekers can be obtained, further details will not be provided.

[0026] Specifically, the process of determining a number of candidate resumes includes, The semantic similarity between the text descriptions of the job application resumes is determined as the similarity score; If the similarity is greater than or equal to a preset similarity threshold, the resumes are merged, and the merged resumes are identified as candidate job application resumes.

[0027] In implementation, the purpose of the similarity threshold is to characterize the minimum standard of text similarity sufficient for effective merging. The preset standard similarity is predetermined. Those skilled in the art can determine the mean semantic similarity between resumes by calculating the semantic similarity of resume texts from historical recruitment cycles, representing the normal situation of resume text similarity. To represent the situation of effective merging in practical applications, the similarity threshold is set as the product of the mean semantic similarity and the similarity error coefficient. Typically, the similarity error coefficient is selected within the range of [1.05, 1.25], and is preferably 1.2 in implementation.

[0028] Specifically, the characteristic feature is that the process of extracting demand keywords includes, Identify the nouns and noun phrases in the text description corresponding to the job posting; The nouns and noun phrases mentioned above are identified as the required keywords.

[0029] Specifically, the process of determining whether the candidate text description meets the preset requirement association conditions includes, Determine the semantic correlation between the requirement text description and the candidate text description; If a preset proportion of the required keywords fall into the candidate text description and the semantic relevance of the text is greater than or equal to the preset semantic relevance threshold, then the candidate text description is determined to meet the preset requirement relevance condition.

[0030] In implementation, the purpose of the text semantic association threshold is to characterize the minimum acceptable degree of matching between the requirement text and the resume text at the semantic level. The text semantic association threshold is predetermined. Those skilled in the art can statistically analyze the resume texts of successfully invited candidates in historical recruitment data and determine the average text semantic association degree between the successfully invited resume texts and the requirement text of the recruitment position to represent the normal text semantic matching situation. To account for the fluctuations in semantic association degree caused by differences in text expression in practical applications, the text semantic association threshold is set as the text error coefficient of the average text semantic association degree. Typically, the text error coefficient is selected within the range [0.7, 1.35], and preferably 0.8 in implementation.

[0031] In implementation, the key requirement for the predetermined ratio is that it is predetermined. The predetermined ratio is usually selected within the range of [60%, 85%], and is preferably 75% in implementation.

[0032] Please see Figure 2 The diagram shown is a logic decision diagram for setting tags on candidate job application resumes according to an embodiment of the invention. Specifically, the process of setting tags on candidate job application resumes includes: If the candidate text description meets the preset demand association conditions, then the candidate job application resume is identified as a job application resume with obvious matching tendency. If the candidate text description does not meet the preset requirement association conditions, the candidate job application resume will be identified as a job application resume with fuzzy matching tendency.

[0033] This invention obtains candidate text descriptions corresponding to candidate job applicant resumes, determines whether the candidate text descriptions meet preset requirement association conditions, and assigns tags to the candidate job applicant resumes. In actual resume screening scenarios, since the required keywords for a job posting directly reflect the core requirements of the position, and the semantic relevance of the text measures the semantic fit between the candidate resume and the job requirements, using a single judgment standard may lead to one-sided screening results. For example, it may lead to screening candidate resumes with keyword stuffing but inconsistent actual experience, or relying solely on semantic relevance may miss high-quality resumes that accurately possess core skills but whose expression differs significantly from the job description. Therefore, if a preset proportion of required keywords fall into the candidate text description and the text semantic relevance is greater than or equal to a preset text semantic relevance threshold, then the candidate text description is determined to meet the preset requirement association conditions. Through this dual judgment mechanism, resumes that are highly matched to the job can be screened more accurately, divided into resumes with obvious matching tendency and resumes with fuzzy matching tendency, providing a reliable basis for subsequent differentiated analysis strategies for candidate job applicant resumes, thereby improving the efficiency and accuracy of resume screening.

[0034] Please seeFigure 3 The diagram shown is a logic diagram for determining whether to perform periodic coverage verification on demand keywords according to an embodiment of the invention. Specifically, the process of determining the historical stability characteristics of historical invitation resumes and demand keywords to determine whether to perform periodic coverage verification on demand keywords includes... Obtain historical resumes for job openings within the designated period; The ratio of the number of historical resumes containing the keyword of demand to the total number of historical resumes is determined as the historical stable characteristic; If the historical stability feature is greater than or equal to the preset historical stability threshold, then periodic coverage verification is performed on the required keywords; The periodic coverage verification includes dividing the predetermined period into several recruitment time periods and verifying whether the required keywords appear in a predetermined proportion of the recruitment time periods.

[0035] In implementation, the purpose of the historical stability threshold is to characterize the minimum standard for judging whether the demand keywords have long-term stability and need to be periodically verified. The historical stability threshold is predetermined and is usually selected within the range [70%, 95%], and is preferably 85% in implementation.

[0036] In practice, the predetermined period is determined in advance. Typically, the predetermined period is selected within the range of [6 months and 24 months], and is preferably 12 months in practice.

[0037] In practice, dividing the recruitment period into several recruitment time periods means dividing the predetermined period into several recruitment time periods. Preferably, if the predetermined period is 12 months, it can be divided into 4 recruitment time periods, each recruitment time period being 3 months.

[0038] In implementation, the required keywords are considered to be within a predetermined proportion of the recruitment time period, with the predetermined proportion preferably being 75%.

[0039] Specifically, the process of selecting a stable demand keyword set based on the verification results and calculating the stable association value between the stable demand keyword set and the candidate text description includes, If the periodic coverage verification is passed, the aforementioned demand keywords will be selected as stable demand keywords and included in the stable demand keyword set. Identify the stable demand keywords in the aforementioned stable demand keyword set; Determine the number of each stable demand keyword in each candidate text description; The quantity is determined as a stable correlation value.

[0040] In practice, a stable priority sequence is generated based on the stable association values, and the resumes with obvious matching tendencies are recommended sequentially. Preferably, the resumes with obvious matching tendencies are arranged from high to low according to their corresponding stable association values ​​to generate a stable priority sequence, and recommendations are made based on this sequence.

[0041] This invention targets resumes with a clear matching tendency. It determines the historical stable features of historically invited resumes and required keywords to decide whether to perform periodic coverage verification of the required keywords. Based on the verification results, a stable set of required keywords is selected, and the stable association value between the stable required keyword set and candidate text descriptions is calculated. A stable priority sequence is generated based on the stable association value, and sequence recommendation is performed on the resumes with a clear matching tendency. In actual resume screening scenarios, the keyword co-occurrence patterns reflected in historically invited resumes can accurately reflect the core elements of job requirements. Therefore, the historical stable features are determined based on the ratio of the number of historically invited resumes containing specific required keywords to the total number of invited resumes within a predetermined period. This helps determine whether the keyword has general representativeness. However, statistical results within a single time span may be affected by temporary or periodic factors such as specific recruitment batches or market trends, which may lead to some temporary, non-core requirements being misjudged as stable requirements. For example, for a "event execution" position recruited in large numbers for a short period to cope with a temporary technology summit, the keyword "exhibition planning" may have an abnormally high historical stable feature in that period, but it is not necessarily a stable requirement. The core competencies for this position are determined over a long period. Therefore, when the historical stability characteristics exceed the preset historical stability threshold, the required keywords are verified through periodic coverage. This involves dividing the predetermined period into multiple consecutive recruitment time periods and checking whether the required keywords appear consistently and stably in multiple time periods exceeding a preset proportion. This allows for the selection of stable required keywords that truly span the recruitment cycle and represent stable job requirements. A set of stable required keywords is then constructed, ensuring that the generated priority sequence is based on long-term stable job requirements rather than short-term fluctuations. Based on the proportion of stable required keywords in resumes with a clear matching tendency, the sequence recommendation prioritizes candidate resumes containing stable required keywords, thereby improving the efficiency and accuracy of resume screening.

[0042] Specifically, the process of determining whether the paragraph containing the desired keyword has been selected, performing matching observation on the paragraph, and determining whether the preset matching conditions are met includes, If the resume to which the fuzzy match is favored contains any of the required keywords, then it is determined that the resume to which the fuzzy match is favored is related to the required keywords, and the paragraph containing the required keywords is selected; Determine the candidate text descriptions that fuzzy match tends to match with job application resumes; Determine the semantic relevance between the paragraph and the requirement text description; If the semantic relevance is greater than or equal to a preset semantic relevance threshold, then the paragraph is marked; Determine the ratio of the total text length of all marked paragraphs to the total length of the candidate text descriptions; The matching condition is that the ratio must be greater than or equal to a predetermined text length percentage threshold.

[0043] In implementation, the semantic association threshold aims to characterize the minimum acceptable degree of semantic matching between the job posting and the resume text. This semantic association threshold is predetermined. Those skilled in the art can statistically analyze the resume texts of successfully invited candidates in historical recruitment data and determine the average semantic association degree between the successfully invited resume texts and the job posting's requirements text to represent the normal semantic matching situation. To account for fluctuations in semantic association degree due to differences in textual expression in practical applications, the semantic association threshold is set as the product of the average value and the semantic error coefficient. Typically, the semantic error coefficient is selected within the range of 0.85 to 1.35, and is preferably 1.25 in implementation.

[0044] In practice, the text length percentage threshold is usually selected within the range of [25%, 65%], and is preferably 40%.

[0045] Please see Figure 4 The diagram shown illustrates the logic for determining whether to perform sequence recommendation on resumes showing a tendency towards fuzzy matching, according to an embodiment of the invention. Specifically, the process of determining whether to perform sequence recommendation on resumes showing a tendency towards fuzzy matching includes... If the matching conditions are met, then a sequence recommendation is performed on the job resumes with the fuzzy matching tendency.

[0046] In practice, the texts are ranked from highest to lowest based on the ratio of the total length of the marked paragraphs to the total length of the candidate text descriptions, and then a sequence recommendation is performed.

[0047] In particular, this invention targets resumes with a tendency towards fuzzy matching. It determines whether the resume is related to the required keywords, thereby determining whether to select the paragraph containing the required keywords. The paragraph is then matched and observed to determine if it meets preset matching conditions, thus determining whether to perform sequence recommendation on the resume. In actual resume screening scenarios, although resumes with a tendency towards fuzzy matching may have insufficient overall semantic relevance to the job requirements or fail to meet a preset proportion of required keywords falling into the candidate text description, some paragraphs may contain in-depth information highly relevant to the core requirements. For example, a candidate's resume might be categorized as such because their main technology stack is Python and they are in the fintech field, which deviates significantly from the overall description of the target Java backend position. Fuzzy matching; however, its detailed description of the key paragraph on "low-latency optimization of high-concurrency transaction systems" in a core project profoundly reflects the core capabilities of "high concurrency" and "performance optimization" that are crucial to the "backend development expert" position. Although the context and some technical keywords are different, it still has a potential matching tendency. Therefore, if the resume with the fuzzy matching tendency contains any requirement keyword, it is determined that the resume with the fuzzy matching tendency is related to the requirement keyword. The paragraph is matched and observed. If the matching conditions are met, it means that the candidate's resume has a certain suitability. Then, sequence recommendation is performed. This refined screening strategy based on local semantic analysis can discover potential suitable talents, avoid misscreening, and thus improve the efficiency and accuracy of resume screening.

[0048] Specifically, embodiments of the present invention also provide a QR code-based matching system for job application resumes and interview content, including: The data collection module is used to acquire several job application resumes corresponding to the job postings, extract the text descriptions of the job application resumes, determine the similarity between the text descriptions, merge the job application resumes based on the similarity, and determine several candidate job application resumes. The tag setting module is connected to the acquisition module. It is used to obtain the text description of the job requirements and extract the keywords of the requirements, obtain the candidate text description corresponding to the candidate job application resume, determine whether the candidate text description meets the preset requirements association conditions, and set tags for the candidate job application resume, including job application resumes with obvious matching tendency and job application resumes with fuzzy matching tendency. The obvious matching analysis module, which is connected to the tag setting module, is used to determine the historical stable features of historical invitation resumes and demand keywords for job resumes with obvious matching tendency, to determine whether to perform periodic coverage verification of demand keywords, select a stable demand keyword set based on the verification results, calculate the stable association value between the stable demand keyword set and the candidate text description, generate a stable priority sequence based on the stable association value, and perform sequence recommendation of the candidate job resumes. The fuzzy matching analysis module, which is connected to the explicit matching analysis module, is used to determine whether the fuzzy matching tendency job resumes are related to the demand keywords, so as to determine whether to select the paragraph containing the demand keywords, perform matching observation on the paragraph, determine whether to meet the preset matching conditions, and determine whether to recommend the candidate job resumes in sequence. The visualization module is connected to the acquisition module, the tag setting module, the explicit matching analysis module, and the fuzzy matching analysis module, respectively. It is used to generate corresponding QR codes for recommended candidate resumes, connect the QR codes to the visualization dynamic interface, and scan the corresponding QR codes to present electronic resumes with tags and tag sequences.

[0049] In implementation, there are no restrictions on the structure of the label setting module, explicit matching analysis module, fuzzy matching analysis module, and visualization module. They can be composed of logical components or combinations of logical components. Logical components include field-programmable processors, computers, or microprocessors in computers, which will not be elaborated further.

[0050] In implementation, there are no restrictions on the structure of the data collection module. It can be an API interface of an online recruitment platform, a query system of an enterprise's internal database, or text data extracted from PDF resumes using OCR technology. The only requirement is to ensure that the collected resume data is complete and accurate. This will not be elaborated further.

[0051] In implementation, there are no restrictions on the specific implementation form of the visualization dynamic interface. It can be an API interface based on REST architecture, a GraphQL query interface, or a WebSocket real-time communication interface. As long as it can dynamically generate and return a resume display page containing tag information and sequence ranking based on the scan request, it will not be elaborated further.

[0052] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A method for matching job application resumes with interview content based on QR codes, characterized in that, include, Obtain several job application resumes corresponding to the job postings, extract the text descriptions from the job application resumes, determine the similarity between the text descriptions, merge the job application resumes based on the similarity, and determine several candidate job application resumes. Obtain the job posting requirements text description and extract the requirements keywords, obtain the candidate text description corresponding to the candidate job application resume, determine whether the candidate text description meets the preset requirements association conditions, and set tags for the candidate job application resume, including job application resumes with obvious matching tendency and job application resumes with fuzzy matching tendency. For resumes with obvious matching tendencies, the historical stable features of historical invitation resumes and demand keywords are determined to determine whether to perform periodic coverage verification of demand keywords. Based on the verification results, a stable demand keyword set is selected, and the stable association value between the stable demand keyword set and the candidate text description is calculated. A stable priority sequence is generated based on the stable association value, and the sequence recommendation is performed on the resumes with obvious matching tendencies. For job resumes with a tendency to match fuzzy matching, determine whether the job resumes with a tendency to match fuzzy matching are related to the required keywords, in order to determine whether to select the paragraph containing the required keywords, perform matching observation on the paragraphs, determine whether to meet the preset matching conditions, and then determine whether to perform sequence recommendation on the job resumes with a tendency to match fuzzy matching. A corresponding QR code is generated for each recommended candidate resume. The QR code is then connected to a visual dynamic interface. Scanning the QR code displays an electronic resume with tags and their corresponding sequences.

2. The method for matching job application resumes and interview content based on QR codes according to claim 1, characterized in that, The process of determining a number of candidate resumes includes: The semantic similarity between the text descriptions of the job application resumes is determined as the similarity score; If the similarity is greater than or equal to a preset similarity threshold, the resumes are merged, and the merged resumes are identified as candidate job application resumes.

3. The method for matching job application resumes and interview content based on QR codes according to claim 1, characterized in that, The process of extracting the required keywords includes, Identify the nouns and noun phrases in the text description corresponding to the job posting; The nouns and noun phrases mentioned above are identified as the required keywords.

4. The method for matching job application resumes and interview content based on QR codes according to claim 1, characterized in that, The process of determining whether the candidate text description meets the preset requirements and association conditions includes: Determine the semantic correlation between the requirement text description and the candidate text description; If a preset proportion of the required keywords fall into the candidate text description and the semantic relevance of the text is greater than or equal to the preset semantic relevance threshold, then the candidate text description is determined to meet the preset requirement relevance condition.

5. The method for matching job application resumes and interview content based on QR codes according to claim 4, characterized in that, The process of tagging candidate resumes includes: If the candidate text description meets the preset demand association conditions, then the candidate job application resume is identified as a job application resume with obvious matching tendency. If the candidate text description does not meet the preset requirement association conditions, the candidate job application resume will be identified as a job application resume with fuzzy matching tendency.

6. The method for matching job application resumes and interview content based on QR codes according to claim 1, characterized in that, The process of determining the historical stability characteristics of historical invitation resumes and demand keywords to decide whether to perform periodic coverage verification of demand keywords includes, Obtain historical resumes for job openings within the designated period; The ratio of the number of historical resumes containing the keyword of demand to the total number of historical resumes is determined as the historical stable characteristic; If the historical stability feature is greater than or equal to the preset historical stability threshold, then periodic coverage verification is performed on the required keywords; The periodic coverage verification includes dividing the predetermined period into several recruitment time periods and verifying whether the required keywords appear in a predetermined proportion of the recruitment time periods.

7. The method for matching job application resumes and interview content based on QR codes according to claim 1, characterized in that, The process of selecting a stable demand keyword set based on the verification results and calculating the stable association value between the stable demand keyword set and the candidate text descriptions includes: If the periodic coverage verification is passed, the aforementioned demand keywords will be selected as stable demand keywords and included in the stable demand keyword set. Identify the stable demand keywords in the aforementioned stable demand keyword set; Determine the number of each stable demand keyword in each candidate text description; The quantity is determined as a stable correlation value.

8. The method for matching job application resumes and interview content based on QR codes according to claim 1, characterized in that, The process of determining whether the paragraph containing the desired keyword has been selected, performing a matching observation on the paragraph, and determining whether the preset matching conditions are met includes, If the resume to which the fuzzy match is favored contains any of the required keywords, then it is determined that the resume to which the fuzzy match is favored is related to the required keywords, and the paragraph containing the required keywords is selected; Determine the candidate text descriptions that fuzzy match tends to match with job application resumes; Determine the semantic relevance between the paragraph and the requirement text description; If the semantic relevance is greater than or equal to a preset semantic relevance threshold, then the paragraph is marked; Determine the ratio of the total text length of all marked paragraphs to the total length of the candidate text descriptions; The matching condition is that the ratio must be greater than or equal to a predetermined text length percentage threshold.

9. The method for matching job application resumes and interview content based on QR codes according to claim 1, characterized in that, The process of determining whether to perform sequence recommendation on the fuzzy-matched job resumes includes... If the matching conditions are met, then a sequence recommendation is performed on the job resumes with the fuzzy matching tendency.

10. A matching system for job application resumes and interview content based on QR codes, as described in any one of claims 1-9, characterized in that, include: The data collection module is used to acquire several job application resumes corresponding to the job postings, extract the text descriptions of the job application resumes, determine the similarity between the text descriptions, merge the job application resumes based on the similarity, and determine several candidate job application resumes. The tag setting module is connected to the acquisition module. It is used to obtain the text description of the job requirements and extract the keywords of the requirements, obtain the candidate text description corresponding to the candidate job application resume, determine whether the candidate text description meets the preset requirements association conditions, and set tags for the candidate job application resume, including job application resumes with obvious matching tendency and job application resumes with fuzzy matching tendency. The obvious matching analysis module, which is connected to the tag setting module, is used to determine the historical stable features of historical invitation resumes and demand keywords for job resumes with obvious matching tendency, to determine whether to perform periodic coverage verification of demand keywords, select a stable demand keyword set based on the verification results, calculate the stable association value between the stable demand keyword set and the candidate text description, generate a stable priority sequence based on the stable association value, and perform sequence recommendation of the candidate job resumes. The fuzzy matching analysis module, which is connected to the explicit matching analysis module, is used to determine whether the fuzzy matching tendency job resumes are related to the demand keywords, so as to determine whether to select the paragraph containing the demand keywords, perform matching observation on the paragraph, determine whether to meet the preset matching conditions, and determine whether to recommend the candidate job resumes in sequence. The visualization module is connected to the acquisition module, the tag setting module, the explicit matching analysis module, and the fuzzy matching analysis module, respectively. It is used to generate corresponding QR codes for recommended candidate resumes, connect the QR codes to the visualization dynamic interface, and scan the corresponding QR codes to present electronic resumes with tags and tag sequences.

Images