Intelligent talent information matching method and system
By collecting and analyzing job characteristics information, establishing matching evaluation coefficients and fuzzy logic screening, the problem of low matching efficiency of newly released jobs in traditional information matching methods is solved, and fast and accurate talent matching and resource conservation are achieved.
Patent Information
- Application Number
- CN202411803810.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-10
- Publication Date
- 2025-01-10
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Traditional information matching methods cannot quickly and accurately match the most relevant talents of the newly released positions, resulting in waste of resources and inefficient matching.
By collecting job characteristic information and historical release information, establishing a data analysis model, generating matching evaluation coefficients, determining conventional or special matching patterns, and using fuzzy logic to screen talents based on job preference similarity and emerging levels.
It achieves the rapid matching of positions to talents related to them, improves matching efficiency and accuracy, and reduces the waste of resources and time in finding talents.
Smart Images
Figure CN119295036A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of information retrieval, and more specifically, to a talent information intelligent matching method and system. Background Art
[0002] In modern society, with the rapid development of science and technology, the demand and supply of talent resources have undergone profound changes. Employers such as enterprises and institutions need to be able to quickly and accurately find talents that meet job requirements. Traditional information matching methods use technologies such as big data analysis to quickly screen out highly matched talents by collecting massive data, thus achieving the purpose of rapid information matching.
[0003] The prior art has the following deficiencies: At present, the traditional information matching method cannot fully understand the characteristics of some newly released positions. It only finds positions similar to the position and recommends them to users. However, this will result in the position not being able to quickly match the most relevant talents. How to quickly match the newly released positions with relevant talents without wasting matching resources is an urgent problem to be solved. Therefore, a talent information intelligent matching method and system are proposed.
[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention
[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a talent information intelligent matching method and system, which solves the problems raised in the above-mentioned background technology by using different product inspection methods.
[0006] To achieve the above object, the present invention provides the following technical solution, a talent information intelligent matching method, comprising: S1: Collect job characteristic information and historical release information, and process them into data to obtain job requirement similarity, job comparison retrieval rate and historical release matching rate; S2: Obtain job characteristic information and historical release information, establish a data analysis model, and generate a matching evaluation coefficient; S3: Obtain the matching evaluation coefficient and compare it with the matching threshold to obtain the general matching attributes and special matching attributes. Determine the matching mode of the newly released position based on the ratio of the number of general matching attributes to the number of special matching attributes. For the special matching mode, sort the positions from large to small based on the similarity values of the position requirements, and evaluate the number of talents for the most similar matching positions until the primary matching is met. S4: Obtain primary matching information, determine the similarity of job preferences and collect the degree of new job emergence; S5: Use fuzzy logic to determine new job matching results based on the similarity of job preferences and the degree of new job emergence.
[0007] In a preferred embodiment, the job characteristic information includes the job requirement similarity and the comparison job retrieval rate; the historical release information includes the historical release matching rate; The skill requirements are represented by vectorization and are calculated using the cosine similarity formula to obtain the skill requirement similarity. The keywords in the job description are analyzed, and the Jaccard similarity is used to measure the job responsibilities and experience similarity. The similarity of the job requirements is calculated using the weighted formula. ; Where i is the i-th comparison, and the number of compared positions is i; Through the system's built-in search log, check the user's search behavior on the platform, count the number of times the matching position appears in the search results and calculate the ratio with the current total search number to obtain the matching position search rate ; The historical release matching rate is calculated by calculating the ratio of the number of successfully matched talents during the historical release period to the total number of recommended talents during the historical release period. .
[0008] In a preferred embodiment, obtaining the similarity of job requirements , compare job search rate And historical release match rate , and calculate the matching evaluation coefficient , based on the formula: In the formula, is the matching evaluation coefficient, , as well as are the preset proportional coefficients of job requirement similarity, comparison job retrieval rate and historical posting matching rate, respectively, and , as well as Both are greater than 0.
[0009] In a preferred embodiment, after obtaining the matching evaluation coefficient, the matching evaluation coefficient is compared and analyzed with the continuously iterated matching threshold; If the matching evaluation coefficient is greater than or equal to the matching threshold, the newly posted position is marked as a regular matching attribute and a regular signal is generated; If the matching evaluation coefficient is less than the matching threshold, the newly released position is marked as a special matching attribute and a special signal is generated.
[0010] In a preferred embodiment, all matching evaluation coefficients are collected, the number of conventional matching attributes and the number of special matching attributes are found, and the ratio of the number of conventional matching attributes and the number of special matching attributes is calculated. If the ratio is greater than or equal to one, the conventional matching mode is performed for the newly released position; if the ratio is less than one, the special matching mode is performed for the newly released position; For special matching modes, the similarity values of job requirements are sorted from large to small, and the number of talents corresponding to the most similar matching positions is defined as the number of talents for the most similar matching positions and compared with the talent threshold. If it is greater than the talent threshold, the primary match is satisfied. Otherwise, the above sorting is followed and the number of talents for similar matching positions is added until it is greater than the talent threshold.
[0011] In a preferred embodiment, the primary matching information is used to quantify the job characteristics of the talent's intended positions corresponding to the positions, and then the Euclidean distance between the two job characteristic vectors is calculated to obtain the job preference similarity of the compared positions. ; Where g is the comparison position of the g-th primary match; By comparing and analyzing the differences in key characteristics between newly released positions and primary assigned positions, the degree of new positions’ emergence is obtained. .
[0012] In a preferred embodiment, the similarity of compared job preferences and the degree of emergence of newly released jobs are defined as input variables, and they are divided into different fuzzy sets respectively; The new job matching results are defined as output variables and divided into fuzzy sets; Formulate fuzzy rules to describe the impact of job preference similarity and the new job emergence degree on new job matching results; Perform fuzzy reasoning based on fuzzy rules to determine the new job matching results.
[0013] A talent information intelligent matching system, comprising a data collection module, a data processing module, a primary matching module and a matching screening module; The data collection module is used to collect job characteristic information and historical release information, and process them into data to obtain job requirement similarity, comparison job retrieval rate and historical release matching rate, and send them to the data processing module; The data processing module is used to obtain job characteristic information and historical release information, establish a data analysis model, generate matching evaluation coefficients and send them to the primary matching module; The primary matching module is used to receive the matching evaluation coefficient and compare it with the matching threshold to obtain the general matching attributes and the special matching attributes, determine the matching mode of the newly released position according to the ratio of the number of general matching attributes to the number of special matching attributes, and for the special matching mode, sort the positions from large to small according to the similarity values of the position requirements, evaluate the number of talents for the most similar matching positions, until the primary matching is satisfied, and send the primary matching information to the matching screening module; The matching and screening module is used for primary matching information, determining the similarity of job preferences and collecting the new job postings’ degree of novelty, and using fuzzy logic to determine the matching results of new jobs.
[0014] Technical effects and advantages of the present invention: 1. The present invention obtains the job requirement similarity, the comparison job retrieval rate and the historical release matching rate by collecting job characteristic information and historical release information, and establishes a data analysis model to generate a matching evaluation coefficient, which is compared with the matching threshold to obtain conventional matching attributes and special matching attributes. The matching mode of the newly released job is determined based on the quantitative ratio of conventional matching attributes to special matching attributes. For the special matching mode, the job requirements are sorted in descending order according to the similarity values, and the number of talents for the most similar matching jobs is evaluated until the primary match is met, so that the job can be quickly matched to the talents related to it.
[0015] 2. The present invention obtains and compares the similarity of job preferences and the degree of new job emergence, formulates a set of fuzzy rules for fuzzy reasoning, and determines the new job matching plan. While ensuring rapid matching of talents, it further screens out some talents who may not be related to the field, thereby improving matching efficiency and accuracy, reducing the waste of matching resources, and saving time in seeking talents. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] Figure 1 The present invention is a method flow chart of a talent information intelligent matching method.
[0017] Figure 2 The figure is a module diagram of a talent information intelligent matching system of the present invention. DETAILED DESCRIPTION
[0018] 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.
[0019] The present invention is based on a collaborative filtering model and uses a similarity algorithm to make controllable and precise recommendations for newly released positions; Example 1 See also Figure 1 , a talent information intelligent matching method, the specific operation process is as follows: S1: Collect job characteristic information and historical release information, and process them into data to obtain job requirement similarity, job comparison retrieval rate and historical release matching rate; Among them, the job characteristic information includes the job requirement similarity and the comparison job search rate; the historical release information includes the historical release matching rate; By comparing the similarity of the newly released positions with other positions, we can evaluate which positions have the highest similarity with them. There is no limit on the newly released positions, but it is set by the experimenters according to the specific situation. For example, the positions released within one day are limited to the newly released positions, which will not be repeated here. Specifically, a newly released position may be a position in an emerging field or a position released on the platform for the first time, or a position that has been released on the platform before; among them, positions in emerging fields have hidden characteristics, and users cannot actively search for the position without knowing the position, so the number of times users search for the position or terms related to the position is not considered; Job requirement similarity refers to the similarity between the newly posted job and multiple jobs in terms of key requirements, where key requirements include but are not limited to similarities in skills, job responsibilities, and experience; The acquisition logic is obtained through similarity calculations such as cosine similarity, Jaccard similarity or Euclidean distance. Specifically, the job skill requirements are vectorized, and the cosine angle between the vectors of the newly emerging or newly released jobs and the comparison jobs is calculated to obtain the similarity of skill requirements. Among them, the setting of the comparison job is specifically set by the experimenters according to the actual situation, and it is not limited here. Suppose positions A and B, and skill requirements are represented by vector and Indicates that the specific cosine similarity formula is as follows: In the formula, is the dot product of the two skill vectors, is the magnitude of vector A, is the modulus of vector B; according to the formula, we get the skill requirement similarity ; Where i is the i-th comparison, and the number of compared positions is i; The skill requirement vector needs to be limited, which may include general requirements. For example, a technical position requires strong technical skills, but the education requirement is a college degree. A sales position requires eloquence and sales experience, and the education requirement may also be a college degree. This will lead to the similarity of popular attributes and misleading. Therefore, a penalty for popular attributes is introduced, that is, the score contribution of popular attributes is reduced to accurately calculate the skill similarity. Specifically, we analyze the keywords in the job description, which contains job responsibilities, experience and other requirements. We use Jaccard similarity to measure the similarity between two jobs in terms of responsibilities and experience. The Jaccard similarity is defined as the ratio of the intersection and union of the job responsibilities and experience descriptions of the two jobs. The specific Jaccard similarity formula is as follows: In the formula, is the number of common keywords in the job responsibilities and experience descriptions of the two positions (i.e., the intersection), is the total number of all keywords (i.e., the union); It is the similarity of job responsibilities and experience; Calculate the similarity of job requirements through the similarity of the above skill requirements and the similarity of job responsibilities and experience : in, is the similarity of job requirements, and is the preset ratio coefficient of the similarity of skill requirements and the similarity of job responsibilities and experience, and and All are greater than 0; The logic for obtaining the matching job search rate is to check the user's search behavior on the platform through the system's built-in search log, count the number of times the matching job appears in the search results, and calculate the ratio with the current total search number to obtain the matching job search rate. ; The current total search count is set based on the time period of the system's built-in search log, such as daily, weekly, monthly, etc. Usually, the time period for the comparison positions of the newly released positions is generally set to the total search count of the comparison positions yesterday, so as to evaluate the popularity of the comparison positions in the short term; The historical release matching rate refers to the newly released position that has been released on the platform before and has a historical record. Since the position has a historical release, the historical release matching rate of the position can be used to match the current position based on the matching information released at that time; its acquisition logic is that the system searches whether there is a historical release record of the same position on the platform for the newly released position, and calculates the historical release matching rate based on the ratio of the number of successfully matched talents during the historical release period to the total number of recommended talents during the historical release period. On the contrary, if the system does not detect the historical record of the newly released position on this platform, the historical release matching rate of the newly released position will be set to zero by default.
[0020] S2: Obtain job characteristic information and historical release information, establish a data analysis model, and generate a matching evaluation coefficient; Get job requirements similarity , compare job search rate And historical release match rate , and calculate the matching evaluation coefficient , based on the formula: In the formula, is the matching evaluation coefficient, , as well as are the preset proportional coefficients of job requirement similarity, comparison job retrieval rate and historical posting matching rate, respectively, and , as well as All are greater than 0; Among them, the similarity of job requirements, the comparison job retrieval rate and the historical release matching rate are all data-based expressions that directly express how the current newly released jobs match.
[0021] From the above formula, it can be seen that when the historical release matching rate is higher, it means that the newly released position belongs to the position that has been released before and recruited the corresponding talent. The higher the matching evaluation coefficient is, the less it is necessary to change the matching mode. On the contrary, if the position requirement similarity and the comparison position retrieval rate are higher, it means that the new position is more similar to the comparison position, and the comparison position is more popular, the matching evaluation coefficient is lower, and the matching mode needs to be changed more. Among them, for matching modes, the system usually has at least one matching mode built in. Specifically, the conventional matching mode is established through talent attention or talent information. This mode can cope with most application scenarios well and is highly efficient. However, for positions that have never been posted on the platform, the system cannot accurately find the corresponding matching talents through talent information. Even through calculations such as job similarity, it is impossible to accurately match from a large number of talents. Therefore, the matching mode needs to be adjusted. S3: Obtain the matching evaluation coefficient and compare it with the matching threshold to obtain the general matching attributes and special matching attributes. Determine the matching mode of the newly released position based on the ratio of the number of general matching attributes to the number of special matching attributes. For the special matching mode, sort the positions from large to small based on the similarity values of the position requirements, and evaluate the number of talents for the most similar matching positions until the primary matching is met. The logic for obtaining the matching threshold is to collect a distribution sample set of the historical job matching information database, divide the data set into a training set and a test set, set the evaluation index and clustering algorithm, and in each round of cross-validation, train the model on the training set and evaluate the model performance on the test set. Then, adjust the matching threshold based on the performance of the validation set. Therefore, the matching threshold is constantly updated. In the present invention, clustering algorithm is a kind of unsupervised learning algorithm, which is used to divide the data points in the data set into groups or clusters with similarity; the common one is K-means clustering, which divides the data points in the data set into K clusters, so that the distance between the data point of each curve and the center point (centroid) of the cluster to which it belongs is minimized, and finally the effect of the adjusted matching threshold is measured by the Euclidean distance, so as to set the matching threshold; After obtaining the matching evaluation coefficient, the matching evaluation coefficient is compared and analyzed with the continuously iterated matching threshold; If the matching evaluation coefficient is greater than or equal to the matching threshold, the newly posted position is marked as a regular matching attribute and a regular signal is generated; If the matching evaluation coefficient is less than the matching threshold, the newly released position is marked as a special matching attribute and a special signal is generated; Collect all matching evaluation coefficients, find the number of regular matching attributes and the number of special matching attributes, and calculate the ratio of the number of regular matching attributes to the number of special matching attributes. If the ratio is greater than or equal to one, the regular matching mode is used for the newly released position. If the ratio is less than one, the special matching mode is used for the newly released position. Specifically, the special matching mode refers to finding the most similar comparison position to the newly released position and finding the corresponding number of talents. When the number of talents reaches the specified requirement, the newly released position is matched to the corresponding talents. For special matching modes, the job requirements similarity values are sorted from large to small, and the number of talents corresponding to the most similar matching positions is defined as the number of talents for the most similar matching positions and compared with the talent threshold. If it is greater than the talent threshold, the primary match is satisfied. Otherwise, the above sorting is followed and the number of talents for similar matching positions is added until it is greater than the talent threshold. The specific talent threshold is set by the publisher or company HR. For example, when publishing, the system pops up a dialog box to ask the publisher how many people they expect to match, etc. The primary match refers to the number of people who meet the publisher's expectations and are matched; For example, the publisher or company HR sets the expected number of matches to be 300, and the number of talents for the most similar matching positions is 130. Then, according to the second position in the descending order of the similarity values of the job requirements, the number of talents for similar matching positions is 120. If it is still less than the talent threshold after the comparison, then according to the third position in the descending order of the similarity values of the job requirements, the number is greater than the talent threshold of 300. The present invention obtains the job requirement similarity, the comparison job retrieval rate and the historical release matching rate by collecting job characteristic information and historical release information, and establishes a data analysis model to generate a matching evaluation coefficient, which is compared with a matching threshold to obtain conventional matching attributes and special matching attributes. The matching mode of the newly released job is determined according to the quantitative ratio of the conventional matching attributes to the special matching attributes. For the special matching mode, the job requirement similarity values are sorted in descending order, and the number of talents for the most similar matching jobs is evaluated until the primary matching is met, so that the job can be quickly matched to the talents related to it.
[0022] Example 2 In Example 1 of the present invention, examples are given to collect job characteristic information and historical release information, obtain job requirement similarity, compare job retrieval rate and historical release matching rate, establish a data analysis model, generate a matching evaluation coefficient, compare with the matching threshold, obtain conventional matching attributes and special matching attributes, determine the matching mode of the newly released job based on the ratio of the number of conventional matching attributes to the number of special matching attributes, and for the special matching mode, sort the job requirements from large to small in order according to the similarity values of the job requirements, evaluate the number of talents in the most similar matching jobs, until the primary matching operation strategy is met; however, in Example 1, talents are allocated only from the primary allocation. Obviously, although this method can quickly match the corresponding talents, it does not take into account the popularity of certain occupations, so that most users still search for the occupation although they are not technical talents in the field, so that the value of the primary allocated talents is not high. In other words, the talents in the primary allocation may not be related personnel in this field, thereby reducing the talent matching efficiency; in view of the above problems, Example 2 of the present invention is further refined; S4: Obtain primary matching information, determine the similarity of job preferences and collect the degree of new job emergence; Among them, the comparison positions in the comparison position preference similarity are the primary matching comparison positions screened out by the talent threshold in the above embodiment 1, and will not be described in detail here; It is understandable that, for example, there is a network virtual anchor position among the similar comparison positions to the newly released position. It is precisely because the network virtual anchor position is less professional than other occupations and the salary is relatively high, so there is a large base of talent groups, but not all talents are suitable for network virtual anchor positions. It is more likely that talents follow the trend or seek opportunities in multiple directions. Therefore, some talents are screened out to match talents more accurately. Specifically, the larger the base of the talent group of the comparison positions similar to the newly released position, the more they can be screened out, and vice versa, the more they need to be retained; Therefore, the preference similarity of the compared positions is limited. For example, if the compared position is less than 35% of the talent threshold, there is no need to collect the preference similarity of the compared position; Specifically, the limiting conditions for the above parameters can be adjusted according to specific circumstances. Those skilled in the art can set the parameters according to the setting of talent thresholds and the employment rate of the current platform, and are not limited here. The similarity of the compared job preference refers to the sum of the similarities between the other preferred jobs of the talent in the current period and the compared job. The acquisition logic is to quantify the characteristics of the talent's preferred job and then calculate the Euclidean distance between the two job feature vectors. , the specific formula is: In the formula, and They represent the feature vectors of the comparison position and other intended positions of the talent, respectively. m is the number of feature dimensions, q is the qth feature dimension, and then the similarity between the comparison position and all intended positions is accumulated to obtain the preference similarity of the comparison position. ; Where g is the comparison position of the g-th primary match; The current time period corresponds to the time period of the system built-in retrieval log described in the above embodiment 1; Specifically, when the similarity of the compared job preferences is higher, it means that the talent correlation between the compared job and the newly released job is closer, and it is less necessary to screen out talents; that is, it means that the talent correlation corresponding to the compared job is high; The degree of new job emergence refers to the similarity and difference with the primary assigned comparison positions. It can be understood that the degree of new job emergence is obtained by comparing and analyzing the differences in key features between the new job and the primary assigned comparison positions. The logic of obtaining it is to obtain the difference in key features between the new job and the primary assigned comparison positions based on the similarity of the corresponding job requirements minus the percentage similarity, that is, the degree of new job emergence ; Specifically, the corresponding job requirement similarity refers to the job requirement similarity of the primary assigned comparison job, which parameter has been described in detail in Example 1 and will not be repeated here; S5: Determine the new job matching result by using fuzzy logic based on the similarity of job preferences and the newness of the newly released jobs; For example, "High", "Low", and "Medium" are used to judge the similarity of job preferences, and "Novel", "Refreshing", and "Unoriginal" are used to judge the novelty of newly released jobs. Formulate a set of fuzzy rules to describe the impact of different input variables on output variables. The definition of rules can be based on professional knowledge or obtained through data analysis and experiments. For example: Mark the job preference similarity as X, the new job emergence as U, and the new job matching results as C_results; Then we can define: Rule 1: IF (X is High) AND (Uis Unoriginal) THEN (C_results is Low) Rule 2: IF (U is Low) AND (U is Novel) THEN (C_results is High) Perform fuzzy reasoning based on fuzzy rules to determine the new job matching plan; It should be noted that the division of fuzzy sets can be adjusted according to actual conditions. For example, although this embodiment takes three fuzzy sets as an example, it can actually be divided into more than three sets to facilitate better accurate adjustment according to different similarities. Furthermore, for the judgment of the similarity of the compared job preferences and the high, medium and low degree of the newly released job, thresholds can be set according to the actual situation for judgment. For example, when the similarity of the compared job preferences exceeds 62%, it is marked as "High", and when the degree of the newly released job (i.e., the difference in key features between the newly released job and the compared job of the primary allocation) is higher than 70%, it is marked as "Novel", etc., which will not be elaborated here; Specifically, this embodiment takes as an example that there is only one primary assigned comparison position, and the actual situation may include multiple primary assigned comparison positions; the situation where there are multiple primary assigned comparison positions can be implemented according to specific plans, for example, when there are two or more new position matching plans, the talents corresponding to more than two primary assigned comparison positions are preferentially screened out, etc., and the specific implementation plan is not limited; The present invention obtains and compares the similarity of job preferences and the degree of emergence of newly released jobs, formulates a set of fuzzy rules for fuzzy reasoning, and determines a new job matching plan. While ensuring rapid matching of talents, it further screens out a part of talents who may not be related to the field, thereby improving matching efficiency and accuracy, reducing the waste of matching resources, and saving time in seeking talents.
[0023] Example 3 See also Figure 2 , a talent information intelligent matching system, including a data collection module, a data processing module, a primary matching module and a matching screening module; The data collection module is used to collect job characteristic information and historical release information, and process them into data to obtain job requirement similarity, comparison job retrieval rate and historical release matching rate, and send them to the data processing module; The data processing module is used to obtain job characteristic information and historical release information, establish a data analysis model, generate matching evaluation coefficients and send them to the primary matching module; The primary matching module is used to receive the matching evaluation coefficient and compare it with the matching threshold to obtain the general matching attributes and the special matching attributes, determine the matching mode of the newly released position according to the ratio of the number of general matching attributes to the number of special matching attributes, and for the special matching mode, sort the positions from large to small according to the similarity values of the position requirements, evaluate the number of talents for the most similar matching positions, until the primary matching is satisfied, and send the primary matching information to the matching screening module; The matching and screening module is used for primary matching information, determining the similarity of job preferences and collecting the new job postings’ degree of novelty, and using fuzzy logic to determine the matching results of new jobs.
[0024] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.
[0025] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.
[0026] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.
[0027] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0028] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0029] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0030] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0031] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
[0032] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.
[0033] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. A talent information intelligent matching method, characterized by: include: S1: Collect job characteristic information and historical release information, and process them into data to obtain job requirement similarity, job comparison retrieval rate and historical release matching rate; S2: Obtain job characteristic information and historical release information, establish a data analysis model, and generate a matching evaluation coefficient; S3: Obtain the matching evaluation coefficient and compare it with the matching threshold to obtain the general matching attributes and special matching attributes. Determine the matching mode of the newly released position based on the ratio of the number of general matching attributes to the number of special matching attributes. For the special matching mode, sort the positions from large to small based on the similarity values of the position requirements, and evaluate the number of talents for the most similar matching positions until the primary matching is met. S4: Obtain primary matching information, determine the similarity of job preferences and collect the degree of new job emergence; S5: Use fuzzy logic to determine new job matching results based on the similarity of job preferences and the degree of new job emergence.
2. A talent information intelligent matching method according to claim 1, characterized in that: Job characteristic information includes job requirement similarity and matching job retrieval rate; historical release information includes historical release matching rate; The skill requirements are represented by vectorization and are calculated using the cosine similarity formula to obtain the skill requirement similarity. The keywords in the job description are analyzed, and the Jaccard similarity is used to measure the job responsibilities and experience similarity. The similarity of the job requirements is calculated using the weighted formula. ; Where i is the i-th comparison, and the number of compared positions is i; Through the system's built-in search log, check the user's search behavior on the platform, count the number of times the matching position appears in the search results, and calculate the ratio with the current total search number to obtain the matching position search rate ; The historical release matching rate is calculated by calculating the ratio of the number of successfully matched talents during the historical release period to the total number of recommended talents during the historical release period. .
3. A talent information intelligent matching method according to claim 2, characterized in that: Get job requirements similarity , compare job search rate And historical release match rate , and calculate the matching evaluation coefficient , based on the formula: In the formula, is the matching evaluation coefficient, , as well as are the preset proportional coefficients of job requirement similarity, comparison job retrieval rate and historical posting matching rate, respectively, and , as well as Both are greater than 0.
4. A talent information intelligent matching method according to claim 3, characterized in that: After obtaining the matching evaluation coefficient, the matching evaluation coefficient is compared and analyzed with the continuously iterated matching threshold; If the matching evaluation coefficient is greater than or equal to the matching threshold, the newly posted position is marked as a regular matching attribute and a regular signal is generated; If the matching evaluation coefficient is less than the matching threshold, the newly released position is marked as a special matching attribute and a special signal is generated.
5. A talent information intelligent matching method according to claim 4, characterized in that: Collect all matching evaluation coefficients, find the number of regular matching attributes and the number of special matching attributes, and calculate the ratio of the number of regular matching attributes to the number of special matching attributes. If the ratio is greater than or equal to one, the regular matching mode is used for the newly released position. If the ratio is less than one, the special matching mode is used for the newly released position. For special matching modes, the similarity values of job requirements are sorted from large to small, and the number of talents corresponding to the most similar matching positions is defined as the number of talents for the most similar matching positions and compared with the talent threshold. If it is greater than the talent threshold, the primary match is satisfied. Otherwise, the above sorting is followed and the number of talents for similar matching positions is added until it is greater than the talent threshold.
6. A talent information intelligent matching method according to claim 5, characterized in that: Through the primary matching information, the job characteristics of the talent's intention corresponding to the position are quantified, and then the Euclidean distance between the two job feature vectors is calculated to obtain the similarity of the job preference. ; Where g is the comparison position of the g-th primary match; By comparing and analyzing the differences in key characteristics between newly released positions and primary assigned positions, the degree of new positions’ emergence is obtained. .
7. A talent information intelligent matching method according to claim 6, characterized in that: The similarity of compared job preferences and the degree of new job emergence are defined as input variables, and they are divided into different fuzzy sets respectively; The new job matching results are defined as output variables and divided into fuzzy sets; Formulate fuzzy rules to describe the impact of job preference similarity and the new job emergence degree on new job matching results; Perform fuzzy reasoning based on fuzzy rules to determine the new job matching results.
8. A talent information intelligent matching system, used to implement a talent information intelligent matching method according to any one of claims 1 to 7, characterized in that: It includes a data collection module, a data processing module, a primary matching module and a matching screening module; The data collection module is used to collect job characteristic information and historical release information, and process them into data to obtain job requirement similarity, comparison job retrieval rate and historical release matching rate, and send them to the data processing module; The data processing module is used to obtain job characteristic information and historical release information, establish a data analysis model, generate matching evaluation coefficients and send them to the primary matching module; The primary matching module is used to receive the matching evaluation coefficient and compare it with the matching threshold to obtain the general matching attributes and the special matching attributes, determine the matching mode of the newly released position according to the ratio of the number of general matching attributes to the number of special matching attributes, and for the special matching mode, sort the positions from large to small according to the similarity values of the position requirements, evaluate the number of talents for the most similar matching positions, until the primary matching is satisfied, and send the primary matching information to the matching screening module; The matching and screening module is used for primary matching information, determining the similarity of job preferences and collecting the new job postings’ degree of novelty, and using fuzzy logic to determine the matching results of new jobs.