Employment recommendation method, device and electronic equipment

Abstract

A method, device and electronic device for job recommendation, relating to the field of data processing. In the method, a first evaluation group of employed students is obtained; keywords are extracted from the first evaluation group to obtain a first keyword group; a second evaluation group of unemployed students is obtained; keywords are extracted from the second evaluation group to obtain a second keyword group; the first keyword group and the second keyword group are matched to determine the proportion of similar keywords; if the proportion is determined to be greater than or equal to a preset proportion threshold, the employment enterprise is added to a recommendation list, and the employment enterprises are sorted from high to low according to the proportion in the recommendation list; the recommendation list is pushed to the unemployed students. The technical solution provided in this application is implemented, and the matching degree of the two types of students is determined through the evaluation content of the teacher on the employed students and the unemployed students, and then the employment enterprises are recommended, so that the employment recommendation is more personalized.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular to an employment recommendation method, device and electronic equipment. Background Art

[0002] With the popularization of higher education and the increasingly fierce employment competition among college students, efficient and accurate employment recommendation methods have become the key to helping students find suitable jobs.

[0003] At present, in the process of recommending students for employment, teachers' evaluation of students often contains a lot of valuable information, which can comprehensively reflect students' professional ability, comprehensive quality and development potential. Compared with standardized information such as resumes, teachers' evaluation will pay more attention to the exploration of students' personalized characteristics. Therefore, it is very necessary to make employment recommendations based on teachers' evaluation of students.

[0004] Therefore, there is an urgent need for a job recommendation method, device and electronic equipment. Summary of the invention

[0005] The present application provides a method, device and electronic device for job recommendation, which determines the matching degree of employed students and unemployed students through the evaluation content of the teacher on the two types of students, and then recommends employment companies, making the job recommendation more personalized.

[0006] In a first aspect of the present application, a method for recommending employment is provided, the method comprising: obtaining employment information of employed students and a first evaluation group of the employed students by teachers based on a first reference basis during the period of study of the employed students in school, the first reference basis being a reference content when the teacher evaluates the employed students, the first reference basis including the academic performance data and in-school behavior records of the employed students, the first evaluation group consisting of a plurality of first evaluations, the first evaluation including four evaluation dimensions of professional skills, communication skills, cooperation skills and learning skills, the employment information including employment companies and employment positions; key evaluation of the first evaluation group Extract words to obtain a first keyword group; obtain a second evaluation group of the unemployed students made by teachers based on a second reference basis during the period of study at school, wherein the second reference basis is the reference content when the teacher evaluates the unemployed students; extract keywords from the second evaluation group to obtain a second keyword group; match the first keyword group and the second keyword group to determine the proportion of similar keywords; if it is determined that the proportion is greater than or equal to a preset proportion threshold, add the employment information to a recommendation list, and sort the employment information in the recommendation list from high to low according to the proportion; push the recommendation list to the unemployed students.

[0007] By adopting the above technical solution, the structured indicator data of four dimensions, namely professional skills, communication skills, cooperation ability and learning ability, are comprehensively considered, which is convenient for teachers to comprehensively evaluate students' comprehensive quality and career potential. At the same time, teachers' evaluation of employed students is used to mine the first keyword reflecting students' characteristics, find employed students similar to unemployed students, and recommend suitable employment companies. Teachers' evaluation can more comprehensively and deeply reflect students' abilities, specialties, personality and other qualities, and contains richer student portrait information. By capturing the first keyword in the first evaluation of employed students and matching it with the second keyword in the second evaluation of unemployed students, the similarity of students can be measured at the semantic level, and then the companies that match the characteristics of unemployed students can be inferred. This method makes full use of evaluation resources, makes employment recommendations more personalized and accurate, and provides targeted employment guidance and selection references for unemployed students.

[0008] Optionally, obtaining the employed companies of the employed students and a first evaluation group of the employed students by teachers based on a first reference basis during their studies in the school, specifically includes: constructing a standardized evaluation template, the standardized evaluation template including evaluation dimensions and evaluation indicators, the evaluation dimensions including professional skills, communication skills, cooperation skills and learning ability, and the evaluation indicators being quantification and refinement of the evaluation dimensions; associating and matching the first reference basis with the evaluation indicators to obtain structured indicator data; obtaining a first evaluation group obtained by multiple teachers evaluating the employed students based on the structured indicator data.

[0009] By constructing a standardized evaluation template, the evaluation dimensions can be refined into quantitative indicators. By associating and matching the first reference basis (such as academic performance, behavior in school, etc.) with the evaluation indicators, the unstructured evaluation text can be converted into structured indicator data, which can ensure that different teachers have a unified reference standard when evaluating students.

[0010] Optionally, the first evaluation group is composed of multiple first evaluations, the first keyword group is composed of multiple first keywords, and the keyword extraction of the first evaluation group to obtain the first keyword group specifically includes: extracting keywords from each of the first evaluations to obtain multiple keywords; removing duplicates from each of the keywords to obtain a target keyword; determining the frequency of each of the target keywords, one target keyword corresponds to one frequency, and the frequency calculation formula is:

[0011] ;

[0012] Among them, T dis the frequency, W1 is the number of times the target keyword appears in the first evaluation group, W2 is the total number of keywords included in the first evaluation group, A is the number of first evaluations included in the first evaluation group, and D1 is the number of first evaluations that include the target keyword; sort the frequencies from high to low to obtain a sorting result; from the sorting result, select the target keyword with the highest frequency and a preset number of keywords as the first keyword.

[0013] By adopting the above technical solution, the importance of a keyword can be evaluated more comprehensively and objectively by comprehensively considering multiple dimensions such as the number of occurrences, proportion, and number of evaluations of the keyword. Keywords with high frequency can more accurately reflect the key characteristics of students and are more representative. By sorting by frequency and selecting some of the most frequent keywords, the core information of the evaluation can be highly refined and condensed, and some non-critical words can be removed, making the extracted keyword group more refined and high-quality. This not only improves the expression efficiency of the keyword group and reduces the amount of calculation for subsequent matching, but also makes the matching results more accurate and reliable.

[0014] Optionally, the second keyword group is composed of multiple second keywords, and the first keyword group and the second keyword group are matched to determine the proportion of similar keywords, specifically including: calculating the comprehensive similarity between the target first keyword and the target second keyword, the target first keyword is any first keyword among the multiple first keywords, and the target second keyword is any second keyword among the multiple second keywords; generating a keyword similarity matrix based on the comprehensive similarity; applying a bipartite graph matching algorithm to the keyword similarity matrix to obtain similar keywords; and calculating the proportion of the similar keywords in the first keyword group.

[0015] By adopting the above technical solution, by calculating the similarity of two keywords, the similarity between employed students and unemployed students in different keyword dimensions is comprehensively measured, and a similarity matrix is ​​formed. Then, the bipartite graph matching algorithm is used to find the global optimal matching solution in the matrix so that the overall similarity of the matched students is the highest. This globally optimized matching method can make full use of the information provided by all keywords to find a group of employed students with the closest overall characteristics as a reference for subsequent corporate recommendations. At the same time, by setting the proportion threshold, the granularity and scale of the match can be flexibly controlled, and a balance can be struck between the matching accuracy and the recommendation range to meet the personalized needs of different students.

[0016] Optionally, the calculation of the comprehensive similarity between the target first keyword and the target second keyword specifically includes: calculating the literal similarity between the target first keyword and the target second keyword; calculating the semantic similarity between the target first keyword and the target second keyword; determining whether the target first keyword and the target second keyword are synonyms; if it is determined that the target first keyword and the target second keyword are synonyms, determining that the synonym judgment result is 1; if it is determined that the target first keyword and the target second keyword are not synonyms, determining that the synonym judgment result is 0; according to a preset weight database, weightedly summing the literal similarity, the semantic similarity and the synonym judgment results to obtain the comprehensive similarity.

[0017] By adopting the above technical solution, the literal similarity, semantic similarity and synonyms are comprehensively considered. The literal form of the two keywords is compared, and their semantic proximity is compared. The dictionary is also consulted to determine whether they belong to the same concept. This multi-dimensional similarity calculation can characterize the similarity of words from different angles, which is more comprehensive and accurate than single character matching. Through weighted summation, the importance of different similarities can be flexibly adjusted to meet different application needs.

[0018] Optionally, if it is determined that the proportion is greater than or equal to a preset proportion threshold, then before sending the employment enterprise to the unemployed students, the method also includes: obtaining the satisfaction score of the employed students for the employment enterprise; if it is determined that the satisfaction score is greater than a preset satisfaction score, determining the preset proportion threshold as a preset first threshold; if it is determined that the satisfaction score is less than or equal to the preset satisfaction score, determining the preset proportion threshold as a preset second threshold, and the preset first threshold is less than the preset second threshold.

[0019] By adopting the above technical solution, by collecting the satisfaction scores of employed students on the companies they work for, and dynamically adjusting the recommendation threshold, it is possible to improve itself according to the actual results. When the satisfaction score of an employment company is high, the recommendation threshold is appropriately lowered to recommend it to more students; when the company's satisfaction is low, the recommendation threshold is raised to avoid leading the unemployed students to an option that may not meet their expectations.

[0020] Optionally, before pushing the recommendation list to the employed students, the method also includes: obtaining the intended positions of the employed students; screening out relevant positions from the recommendation list according to the intended positions, the relevant positions being employment positions having a similarity with the intended positions greater than a preset similarity; placing the relevant positions at the front of the recommendation list, and sorting the relevant positions from high to low according to the proportion; treating other positions after the relevant positions are screened out as remaining positions, and sorting the remaining positions from high to low according to the proportion; and using the employment information corresponding to the sorted related positions and the employment information corresponding to the remaining positions as the recommendation list.

[0021] By adopting the above technical solution, by obtaining the intended positions of employed students and screening out relevant positions with high similarity from the recommendation list based on the intended positions, more personalized and accurate employment recommendations can be achieved. This dual optimization based on students' subjective wishes and objective matching can maximize the satisfaction of students' employment expectations and improve the success rate and conversion rate of recommendations.

[0022] In a second aspect of the present application, a job recommendation device is provided, which includes: an acquisition module and a processing module, wherein the acquisition module is used to acquire the employment company of an employed student and a first evaluation group of the employed student made by the teacher during the period of study at school; the processing module is used to perform keyword extraction on the first evaluation group to obtain a first keyword group; the acquisition module is also used to acquire a second evaluation group of the unemployed student made by the teacher during the period of study at school; the processing module is used to perform keyword extraction on the second evaluation group to obtain a second keyword group; the processing module is also used to match the first keyword group and the second keyword group to determine the proportion of similar keywords; the processing module is also used to add the employment company to a recommendation list if it is determined that the proportion is greater than or equal to a preset proportion threshold, and sort the employment company in the recommendation list from high to low according to the proportion; the processing module is also used to push the recommendation list to the unemployed student.

[0023] In the third aspect of the present application, an electronic device is provided, including a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes any one of the methods described above.

[0024] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores instructions, and when the instructions are executed, any of the methods described above is executed.

[0025] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0026] 1. Comprehensively consider the structured indicator data of four dimensions: professional skills, communication skills, cooperation ability, and learning ability, so that teachers can comprehensively evaluate students' comprehensive qualities and career potential. At the same time, teachers use their evaluation of employed students to mine the first keyword that reflects students' characteristics, find employed students who are similar to unemployed students, and recommend suitable employment companies. Teachers' evaluations can more comprehensively and deeply reflect students' abilities, specialties, personalities and other qualities, and contain richer student portrait information. By capturing the first keyword in the first evaluation of employed students and matching it with the second keyword in the second evaluation of unemployed students, the degree of similarity between students can be measured at the semantic level, and then the companies that match the characteristics of unemployed students can be inferred. This method makes full use of evaluation resources, makes employment recommendations more personalized and accurate, and provides targeted employment guidance and selection references for unemployed students.

[0027] 2. Taking into account the number of occurrences, proportion, number of evaluations, and other dimensions of the keyword, the importance of a keyword can be evaluated more comprehensively and objectively. Keywords with high frequency can more accurately reflect the key characteristics of students and are more representative. By sorting by frequency and selecting the most frequent keywords, the core information of the evaluation can be highly refined and condensed, and some non-critical words can be removed, making the extracted keyword group more refined and high-quality. This not only improves the expression efficiency of the keyword group and reduces the calculation amount of subsequent matching, but also makes the matching results more accurate and reliable.

[0028] 3. By calculating the similarity of two keywords, the similarity between employed students and unemployed students in different keyword dimensions is comprehensively measured, and a similarity matrix is ​​formed. Then, the bipartite graph matching algorithm is used to find the global optimal matching scheme in the matrix so that the overall similarity of the matched students is the highest. This globally optimized matching method can make full use of the information provided by all keywords to find a group of employed students with the closest overall characteristics as a reference for subsequent corporate recommendations. At the same time, by setting the proportion threshold, the granularity and scale of the match can be flexibly controlled, and a balance can be struck between the matching accuracy and the recommendation range to meet the personalized needs of different students. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a flowchart of an employment recommendation method disclosed in an embodiment of the present application;

[0030] Figure 2 is an exemplary schematic diagram of an employment recommendation method disclosed in an embodiment of the present application;

[0031] Figure 3 It is a module schematic diagram of an employment recommendation device disclosed in an embodiment of the present application;

[0032] Figure 4 It is a structural schematic diagram of an electronic device disclosed in an embodiment of the present application.

[0033] Explanation of the reference numerals: 301, acquisition module; 302, processing module; 400, electronic device; 401, processor; 402, communication bus; 403, user interface; 404, network interface; 405, memory. DETAILED DESCRIPTION

[0034] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0035] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

[0036] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0037] This application provides an employment recommendation method, referring to Figure 1 , Figure 1 1 is a flow chart of an employment recommendation method provided in an embodiment of the present application. The method is applied to a server, which is the core server of the employment recommendation system and is used to provide employment recommendation services for schools, students and employers. The server can be a single server, a server cluster consisting of multiple servers, or a cloud computing service center. The method includes steps S101 to S106, which are as follows:

[0038] Step S101: Obtain the employed companies of employed students and the first evaluation group of employed students by teachers based on the first reference basis during their study in school. The first reference basis is the reference content when teachers evaluate employed students. The first reference basis includes the academic performance data and in-school behavior records of employed students. The first evaluation group is composed of multiple first evaluations, and the first evaluation includes four dimensions: professional skills, communication skills, cooperation skills and learning ability.

[0039] In step S101, a standardized evaluation template is constructed, which includes evaluation dimensions and evaluation indicators. The evaluation dimensions include professional skills, communication skills, cooperation skills and learning ability; the first reference basis is associated and matched with the evaluation indicators to obtain structured indicator data; and a first evaluation group is obtained by obtaining a plurality of teachers' evaluations of employed students based on the structured indicator data.

[0040] Specifically, the school has deployed a teaching management platform, which can record students' daily learning data in real time, including the academic performance data and school behavior records of employed students. Students' school records include academic performance data (such as test scores, course design scores, and practical session scores of each course) and school behavior records (such as attendance, participation in activities, and awards). At the same time, the server needs to obtain the employment company information of employed students from the school's employment management system. The employment company information includes the company name, industry, and job requirements; the server constructs a standardized evaluation template based on preset rules. The standardized evaluation template includes two levels: evaluation dimensions and evaluation indicators. The evaluation dimension is a summary of the comprehensive quality of students, including professional skills, communication skills, cooperation skills, and learning ability. The evaluation indicators are further refinements and quantifications of each evaluation dimension. For each evaluation dimension, several measurable evaluation indicators are set. For the professional skills evaluation dimension, the evaluation indicators are broken down into professional course grades, practical project experience, and professional certificate acquisition; for the communication ability evaluation dimension, the evaluation indicators are broken down into foreign language ability, speech and oral expression ability, and written expression ability; for the cooperation ability evaluation dimension, the evaluation indicators are broken down into team project experience, student work performance, and participation in club activities; the server stores the definition of the evaluation indicators in the database to form a standardized evaluation template.

[0041] The server matches the first reference with the corresponding evaluation index according to the preset rules. The preset rules are the mapping rules between the reference and the evaluation index set in advance by the user. The preset rules define how to match and associate the students' various academic performance data and school behavior records with the evaluation indicators. It stipulates which specific evaluation index each reference corresponds to, as well as the specific matching methods and conditions. Through the preset rules, the scattered and heterogeneous student data can be automatically mapped to a unified evaluation index system to form structured and standardized evaluation data. The preset rules are designed and defined by users (such as teachers, academic administrators, etc.) according to the specific needs and standards of the evaluation indicators. In the preset rules, the server can use pattern matching algorithms such as regular expressions to map students' academic performance data and school behavior records to various evaluation indicators. For example, the course containing "communication" in the course name is matched to the "professional course performance" evaluation indicator, and the project containing "national level" in the project name is matched to the "practical project experience" evaluation indicator. Match the skill certificates, qualification certificates, etc. related to the major obtained to the "Professional Certificate Acquisition" evaluation indicator; match the participation in English speech competitions, foreign language debates, etc. to the "Speech and Oral Expression" evaluation indicator; match the experience of serving as a student leader to the "Student Work Performance" evaluation indicator.

[0042] For example, suppose there is a student named Zhang San, whose first reference (student performance data and school behavior records) includes the following: course grades: Communication Principles (90 points), Data Structure (85 points), College English (92 points); project experience: participated in a national college student innovation and entrepreneurship project; obtained certificate: passed the second-level computer exam and obtained a certificate; English proficiency: participated in the school-level English speech competition and won the second prize; student work: served as class monitor and performed well; the server associates and matches the above first reference with the evaluation indicators to form the following structured indicator data:

[0043] The course "Communication Principles" and "Data Structure" in the course grades are matched to the "Professional Course Grades" indicator. The course "College English" is matched to the "Foreign Language Ability" evaluation indicator. The national-level innovation and entrepreneurship project in the project experience is matched to the "Practical Project Experience" evaluation indicator because the project name contains the keyword "national level". This project reflects the student's innovation and practical ability. The second-level computer certificate that will be obtained is matched to the "Professional Certificate Acquisition" evaluation indicator because it is related to the computer major. This certificate reflects the student's professional skill level. The school-level English speech competition that was participated in is matched to the "Speech and Oral Expression" evaluation indicator because the competition name contains the keyword "English Speech". The second prize won reflects the student's strong English oral expression ability. The experience of serving as a class monitor is matched to the "Student Work Performance" evaluation indicator because it is a student cadre job. The excellent performance reflects the student's outstanding organizational management and communication and coordination abilities. Through the above matching process, Zhang San's first reference basis was mapped to multiple evaluation indicators, covering multiple evaluation dimensions such as professional skills, communication skills, and cooperation ability, and structured evaluation data was obtained.

[0044] The server sends an evaluation invitation to the relevant teachers, asking them to evaluate the employed students according to the structured indicator data. The server generates a standardized evaluation form, and teachers can complete the evaluation according to the structured indicator data. The evaluation form is presented in the form of an online questionnaire, which is convenient for teachers to fill in anytime and anywhere. The server collects the evaluation results of multiple teachers (such as class tutors, subject teachers, internship instructors, etc.) for each student, and summarizes them to form a complete first evaluation group, which includes multiple first evaluations of employed students by multiple teachers. The first evaluation group is stored in a data structure such as a list or array, and each element corresponds to an evaluation record, that is, a first evaluation.

[0045] For example, the server obtains the information of student Zhang San from the database: Zhang San works in software company B in city A as a Java development engineer. During his time at school, Zhang San took professional courses such as Java programming, data structure, and software engineering, and his scores in each subject were all above 85 points; he participated in school-enterprise cooperation projects and received praise from the company's instructors; he participated in programming competitions many times and won the provincial second prize; he served as a student union cadre and organized many large-scale activities. The server classifies and quantifies this information according to preset rules to generate structured indicator data. Zhang San's class tutor, subject teacher, project instructor, etc. evaluated him respectively. The class tutor focused on evaluating Zhang San's learning ability and gave evaluations such as "hardworking and diligent in study" and "solid professional foundation"; the subject teacher mainly evaluated Zhang San's professional skills and believed that he had "strong programming ability" and "rich project development experience"; the project instructor mainly evaluated Zhang San's communication and cooperation abilities and believed that he was "good at communication and unity and cooperation". The server aggregated the evaluations from all parties, removed the duplicates, and formed the first evaluation group for Zhang San.

[0046] In this application, the process of the server collecting student data is based on the authorization of the school and the students. As the operator of the teaching management platform, the school has ownership and use rights of student learning data. Through a data sharing agreement with the server, the school authorizes the server to access and use this data for employment recommendation services. In the process of data collection and processing, the server strictly abides by the principle of minimum sufficiency and adheres to the standards of necessity and relevance. The server only collects student learning data that is truly needed for employment recommendation services, such as teacher evaluations, and does not excessively collect other sensitive information of students. In addition, the server will also perform necessary desensitization processing on the collected student data through means such as data anonymization and encryption.

[0047] Step S102: extract keywords from the first evaluation group to obtain a first keyword group.

[0048] In step S102, the first evaluation group is composed of multiple first evaluations, and the first keyword group is composed of multiple first keywords. Keyword extraction is performed on the first evaluation group to obtain the first keyword group, which specifically includes: keyword extraction is performed on each first evaluation to obtain multiple keywords; each keyword is deduplicated to obtain a target keyword; the frequency of each target keyword is determined, and one target keyword corresponds to one frequency. The calculation formula of the frequency is:

[0049] ;

[0050] Among them, T dis the frequency, W1 is the number of times the target keyword appears in the first evaluation group, W2 is the total number of keywords included in the first evaluation group, A is the number of first evaluations included in the first evaluation group, and D1 is the number of first evaluations that include the target keyword; sort the frequencies from high to low to get the sorting results; from the sorting results, select the target keyword with the highest frequency and the preset number of keywords as the first keyword.

[0051] Specifically, the server obtains a first evaluation group consisting of multiple first evaluations. The server performs word segmentation and keyword extraction on each first evaluation. In an embodiment of the present application, the server uses the TF-IDF algorithm to extract keywords. After the keyword extraction step, the server obtains the keywords corresponding to each first evaluation in the first evaluation group. The server performs deduplication processing on all keywords to obtain a unique set of target keywords. After deduplication, the server calculates the frequency of each target keyword, that is, the frequency and importance of the target keyword in the entire first evaluation group. In the frequency calculation formula, W1 represents the number of times the target keyword appears in all evaluations (first evaluation group). The server needs to traverse all evaluations and count the number of occurrences of the target keyword. The more times, the more the keyword is repeatedly mentioned in the evaluation, and the higher the importance. W2 represents the total number of keywords contained in the entire first evaluation group. The total number of keywords reflects the scale and richness of the evaluation data. The more the total number of keywords, the lower the importance of a single target keyword. A represents the number of first evaluations contained in the first evaluation group. The more first evaluations there are, the larger the amount of data, and the more reliable the statistical results of the keywords. D1 represents the number of evaluations containing the target keyword. For a keyword, the more reviews it appears in, the more widely it is distributed and the more important it is. The logarithmic function log plays a smoothing role to avoid large differences in frequency values. It is used to smooth the frequency calculation results and reduce the impact of extreme values. When the target keyword appears frequently in the first evaluation group, the value of this item is close to 1; when the frequency is low, the value of this item is close to 0. This can prevent the frequency calculation results from being affected by extreme values. The power function is introduced to adjust the frequency distribution, making the frequency differentiation more obvious. When the target keyword appears more times in different first reviews, the value of this item will be larger; when it appears less times, the value of this item will be smaller. This can widen the gap between frequencies and make the differentiation more obvious.

[0052] The calculation formula for frequency has been given. The server only needs to count the number of occurrences of each target keyword and other indicators according to the formula, and then substitute them into the formula for calculation. After calculating the frequency of all target keywords, the server sorts them from high to low according to frequency to obtain the sorting result. From the sorting result, the server selects the top N target keywords with the highest frequency as the first keyword, where N is the preset number of keywords. The value of N can be set according to actual needs. Selecting the top 20 keywords can summarize the main points of the evaluation data. This application does not limit this.

[0053] For example, the server extracts 100 first evaluations of a student from the teaching management platform, and after word segmentation and keyword extraction, 500 keywords are obtained. After deduplication, 200 target keywords are obtained. The server counts the number of occurrences and the number of evaluations of each target keyword, substitutes them into the frequency calculation formula, and obtains the frequency score of each target keyword. Then, the server sorts these 200 target keywords from high to low according to frequency, and selects the 10 keywords with the highest target frequency as the first keywords of the student, such as "hardworking", "good at cooperation", "responsible", etc. These keywords provide important clues for subsequent employment recommendations.

[0054] Step S103: obtaining a second evaluation group of the unemployed students by the teacher based on a second reference basis during the unemployed students' study in school, wherein the second reference basis is a reference content when the teacher evaluates the unemployed students.

[0055] In step S103, the server obtains a second evaluation group of the unemployed students from the teaching management platform according to the operating steps in step S101, and the teacher makes a second evaluation group of the unemployed students according to the second reference basis during their study in school. The method of obtaining the second evaluation group of the unemployed students is the same as that of the employed students, and this application does not elaborate on this.

[0056] Step S104: extract keywords from the second evaluation group to obtain a second keyword group.

[0057] In step S104, the server extracts keywords from the second evaluation group, and the step of obtaining the corresponding second keyword group is consistent with the step of obtaining the first keyword group in step S102, which is not described in detail in this application.

[0058] Step S105: Match the first keyword group and the second keyword group to determine the percentage of similar keywords.

[0059] In step S105, the second keyword group is composed of multiple second keywords, and the first keyword group and the second keyword group are matched to determine the proportion of similar keywords, specifically including: calculating the comprehensive similarity of the target first keyword and the target second keyword, the target first keyword is any first keyword among the multiple first keywords, and the target second keyword is any second keyword among the multiple second keywords; generating a keyword similarity matrix based on the comprehensive similarity; applying a bipartite graph matching algorithm to the keyword similarity matrix to obtain similar keywords; and calculating the proportion of similar keywords in the first keyword group.

[0060] Specifically, the server calculates the comprehensive similarity between each first keyword and each second keyword. After calculating the comprehensive similarity of all first keywords and second keywords, the server needs to generate a keyword similarity matrix. The keyword similarity matrix is ​​a two-dimensional array, the row represents each first keyword in the first keyword group, the column represents each second keyword in the second keyword group, and each element represents the comprehensive similarity of the corresponding row and column keywords. In the embodiment of the present application, the first keyword group contains N keywords, the second keyword group contains N keywords, and the shape of the similarity matrix is ​​N*N. The server takes the similarity matrix as input and applies the bipartite graph matching algorithm to obtain the keyword similarity matrix. Bipartite graph matching is a classic graph algorithm used to find a one-to-one matching relationship between two sets to maximize the total weight of the matching. In the embodiment of the present application, the first keyword group and the second keyword group are regarded as two vertex sets of the bipartite graph, and the comprehensive similarity is regarded as the weight of the edge. The goal is to find a set of matches, and the keywords in the matching group are similar keywords. At the same time, the server calculates the proportion of similar keywords in the first keyword group. The proportion refers to the ratio of the number of keywords in the first keyword group that successfully match the second keyword group to the total number of keywords in the first keyword group.

[0061] For example, suppose that after processing, the following two groups of keywords are obtained: the first keyword group: {"learning", "hard work", "effort", "communication", "leadership"} the second keyword group: {"learning", "diligence", "communication", "teamwork", "leadership"}; the server calculates the comprehensive similarity between each keyword in the first keyword group and each keyword in the second keyword group. For the keywords "learning" (first keyword group) and "learning" (second keyword group): the literal similarity is 1 (exactly the same), the semantic similarity is 1 (exactly the same), and the synonym judgment result is 1 (exactly the same); assuming that the weights of the literal similarity, semantic similarity, and synonym judgment results are 0.2, 0.3, and 0.5 respectively, the comprehensive similarity is: 0.2*1+0.3*1+0.5*1=1. By calculating the comprehensive similarity between each pair of keywords, a keyword similarity matrix is ​​obtained, such as Figure 2 As shown. After applying the bipartite graph matching algorithm, we may get the following similar keyword pairs: ("learning", "learning"); ("hard work", "diligence"); ("effort", "diligence"); ("communication", "communication"); ("leadership", "leadership").

[0062] In a possible implementation, the comprehensive similarity between the target first keyword and the target second keyword is calculated, specifically including: calculating the literal similarity between the target first keyword and the target second keyword; calculating the semantic similarity between the target first keyword and the target second keyword; determining whether the target first keyword and the target second keyword are synonyms; if it is determined that the target first keyword and the target second keyword are synonyms, determining the synonym judgment result to be 1; if it is determined that the target first keyword and the target second keyword are not synonyms, determining the synonym judgment result to be 0; according to a preset weight database, weighted summing the literal similarity, semantic similarity and synonym judgment results is performed to obtain a comprehensive similarity.

[0063] Among them, calculating the literal similarity between the target first keyword and the target second keyword specifically includes: performing a character string comparison on the target first keyword and the target second keyword to obtain the longest common substring between the two; calculating the common substring length of the longest common substring, and dividing the common substring length by the average length of the target first keyword and the target second keyword to obtain the literal similarity. Calculating the semantic similarity between the target first keyword and the target second keyword specifically includes: using a preset word vector model to obtain a first word vector of the target first keyword and a second word vector of the target second keyword; calculating the cosine similarity between the first word vector and the second word vector, and using the cosine similarity as the semantic similarity.

[0064] Specifically, the server calculates the comprehensive similarity between the target first keyword and the target second keyword. The comprehensive similarity consists of three parts: literal similarity, semantic similarity, and synonym judgment result. The server calculates these three indicators separately, and then performs weighted summation according to preset weights to obtain the final comprehensive similarity.

[0065] First, the server calculates the literal similarity between the target first keyword and the target second keyword. The literal similarity is used to measure the degree of overlap in the character composition of two keywords. The server first finds the longest common substring of the two keywords, and then divides the length of the longest common substring by the average length of the two keywords.

[0066] Specifically, the server regards the target first keyword and the target second keyword as two strings, and then uses a dynamic programming algorithm to solve their longest common substring. The dynamic programming algorithm is an algorithm that decomposes a complex problem into overlapping subproblems, and starts from the simplest subproblem and gradually derives the optimal solution to the original problem. The server defines a two-dimensional array dp, where dp[i][j] represents the length of the longest common substring of the first i characters of the first keyword and the first j characters of the second keyword. Through the recursive formula dp[i][j]=dp[i-1][j-1]+1; if str1[i]==str2[j]; else0, the server can fill the entire dp array within the time complexity of O(mn), where m and n are the lengths of the two keywords respectively. The maximum value in the dp array is the length of the longest common substring.

[0067] Next, the server calculates the semantic similarity between the target first keyword and the target second keyword. Semantic similarity measures the proximity of two keywords at the semantic level, that is, how similar the meanings they express are. The server uses a preset word vector model, such as Word2Vec, to map the keywords into real-valued vectors of fixed dimensions, and then calculates the cosine similarity of the two vectors.

[0068] Specifically, the server first loads the preset word vector model, which is usually stored on the server's disk in the form of a binary file or a text file. The server uses the target first keyword as an index and queries the first word vector corresponding to the target first keyword from the word vector model. The server will obtain a real-valued vector of a fixed dimension (such as 100 dimensions or 300 dimensions). The query process for the target second keyword is the same as above. After the server obtains the first word vector and the second word vector, the cosine similarity algorithm is used to calculate the cosine similarity between the first word vector and the second word vector.

[0069] Finally, the server determines whether the target first keyword and the target second keyword are synonyms. The server uses a synonym dictionary, such as WordNet. The synonym dictionary stores a large number of synonym sets (Synset), each of which contains a group of words with similar meanings. The server can load the synonym dictionary into memory in advance, or make a remote call in the form of an API. The server queries the synonym dictionary for the synonym sets to which the target first keyword and the target second keyword belong. If the target first keyword and the target second keyword appear in the same synonym set, then they are synonyms, and the server sets the synonym judgment result to 1; otherwise, the server sets the synonym judgment result to 0.

[0070] Combining the above three indicators, the server performs weighted summation on the literal similarity, semantic similarity and synonym judgment results according to the weight values ​​stored in the preset weight database to obtain the final comprehensive similarity. The preset weight database is a configuration file or database table on the server, which is used to store the weight values ​​of each indicator. These weight values ​​can be set based on experience or experimental results, and this application does not limit this.

[0071] Step S106: If it is determined that the proportion is greater than or equal to the preset proportion threshold, the employment information is added to the recommendation list, and the employment information is sorted from high to low in the recommendation list according to the proportion.

[0072] Before step S106, the method also includes: if it is determined that the proportion is greater than or equal to the preset proportion threshold, before sending the employment enterprise to the unemployed students, the method also includes: obtaining the satisfaction score of the employed students for the employment enterprise; if it is determined that the satisfaction score is greater than the preset satisfaction score, determining the preset proportion threshold as the preset first threshold; if it is determined that the satisfaction score is less than or equal to the preset satisfaction score, determining the preset proportion threshold as the preset second threshold.

[0073] Specifically, the server obtains the satisfaction scores of employed students for their employed companies. The satisfaction score is a quantitative indicator that reflects the students' subjective evaluation of various aspects of the company, such as working environment, salary and benefits, development prospects, interpersonal relationships, etc. The satisfaction score can be collected in a variety of ways, such as setting a scoring function on the alumni platform, regularly distributing questionnaires, or collecting feedback through manual interviews and telephone follow-up. The satisfaction score is set to a 5-point or 10-point scale. The server compares the satisfaction score with the preset satisfaction score to decide which preset percentage threshold to use. The preset satisfaction score is a standard pre-set by the user, which means that when the satisfaction reaches this score, the company is considered to be of sufficient quality and the recommendation conditions can be appropriately relaxed. This application does not limit the specific value of the preset satisfaction score.

[0074] For example, if the preset satisfaction score is 4 points, it means that only when the overall satisfaction of the enterprise reaches 4 points or above, it is considered worthy of being strongly recommended. At this time, the server will select a relatively low preset proportion threshold (preset first threshold), such as 60%, which means that even if the keyword matching degree (proportion rate) between unemployed students and employed students is not very high, as long as it reaches 60%, the enterprise can be included in the recommendation list. The purpose of this is to provide more exposure opportunities for high-quality enterprises, so that more unemployed students can understand and apply.

[0075] On the contrary, if the satisfaction score of the enterprise is lower than the preset satisfaction score, the server will select a relatively high preset percentage threshold (preset second threshold), such as 80%. This means that only when the keyword matching degree (percentage) of the unemployed students and the employed students is very high, will the employment enterprise be considered for recommendation to them. The purpose of this is to set a higher threshold for less ideal enterprises to avoid leading unemployed students to an option that may not meet their expectations.

[0076] In step S106, the server compares the proportion with the preset proportion threshold. The proportion refers to the ratio of the number of similar keywords in the first keyword group and the second keyword group to the total number of the second keyword group, reflecting the similarity between employed students and unemployed students in teacher evaluations. When the proportion is greater than or equal to the preset proportion threshold, the server will add the corresponding employment information to the recommendation list. Since there are multiple employed students, the corresponding employment companies that meet the conditions include one or more. The recommendation list is a dynamically generated ordered set, which contains all the employment information that meets the conditions, arranged from high to low according to the proportion.

[0077] Step S107: Push the recommendation list to the job-seeking students.

[0078] Before step S107, the method also includes: obtaining the intended positions of the employed students; screening out relevant positions from the recommendation list according to the intended positions, the relevant positions being employment positions whose similarity with the intended positions is greater than a preset similarity; placing the relevant positions at the front of the recommendation list, and sorting the relevant positions from high to low according to the proportion; treating other positions after the relevant positions are screened out as remaining positions, and sorting the remaining positions from high to low according to the proportion; and using the employment information corresponding to the sorted related positions and the employment information corresponding to the remaining positions as a recommendation list.

[0079] Specifically, the server provides a user interface for the unemployed students to directly input or select the job category corresponding to the intended job they are interested in. After obtaining the intended job of each student, the server further selects jobs with a high degree of relevance to the intended job from the generated recommendation list (based on the matching results of the first keyword group and the second keyword group), which are called "related jobs". The judgment criteria for screening are: calculating the similarity between each job in the recommendation list and the intended job of the student, and if the similarity is greater than the preset similarity threshold, it is considered a related job. The calculation of similarity can use natural language processing technology to map the text descriptions of the intended job and the recommended job into a semantic space, and calculate their distance or angle in the space as the similarity. The selected related jobs will be prioritized in the front of the recommendation list, and the related jobs will still be sorted from high to low according to the original proportion (i.e., the matching proportion of the first keyword group and the second keyword group). The higher the proportion, the higher the matching degree between the job and the student's ability and characteristics, and the higher the ranking. Other jobs that have not been selected as related jobs are called "remaining jobs" and are ranked after the related jobs. The remaining jobs are also sorted from high to low according to the proportion. The relevant positions and remaining positions that have been screened and sorted as above, as well as their corresponding employment information (such as employment companies and employment positions), together constitute the recommendation list that is finally pushed to students.

[0080] In step S107, the server may push the recommendation list to the unemployed students in the form of a pop-up window through a message center in an application used by the unemployed students.

[0081] Reference Figure 3The present application also provides an employment recommendation device, which is a server, and the server includes an acquisition module 301 and a processing module 302; the acquisition module 301 is used to obtain the employment information of employed students and the first evaluation group of employed students made by teachers based on a first reference basis during their study at school. The first reference basis is the reference content when teachers evaluate employed students. The first reference basis includes the academic performance data and school behavior records of employed students. The first evaluation group is composed of multiple first evaluations. The first evaluation includes four evaluation dimensions: professional skills, communication skills, cooperation skills, and learning skills. The employment information includes employment companies and employment positions; the processing module 302 is used to extract keywords from the first evaluation group to obtain the first keyword word group; the acquisition module 301 is also used to obtain a second evaluation group of the unemployed students made by the teacher based on the second reference basis during the period of study at the school, wherein the second reference basis is the reference content when the teacher evaluates the unemployed students; the processing module 302 is also used to extract keywords from the second evaluation group to obtain a second keyword group; the processing module 302 is also used to match the first keyword group and the second keyword group to determine the proportion of similar keywords; the processing module 302 is also used to add the employment information to the recommendation list if it is determined that the proportion is greater than or equal to the preset proportion threshold, and sort the employment information in the recommendation list from high to low according to the proportion; the processing module 302 is also used to push the recommendation list to the unemployed students.

[0082] In a possible implementation, the acquisition module 301 acquires the employed companies of employed students and a first evaluation group of employed students by teachers based on a first reference basis during their studies at school, specifically including: the processing module 302 constructs a standardized evaluation template, the standardized evaluation template includes evaluation dimensions and evaluation indicators, the evaluation dimensions include professional skills, communication skills, cooperation skills and learning ability, and the evaluation indicators are quantification and refinement of the evaluation dimensions; the processing module 302 associates and matches the first reference basis with the evaluation indicators to obtain structured indicator data; the acquisition module 301 acquires a first evaluation group obtained by multiple teachers evaluating employed students based on the structured indicator data.

[0083] In a possible implementation, the first evaluation group is composed of multiple first evaluations, the first keyword group is composed of multiple first keywords, and the processing module 302 extracts keywords from the first evaluation group to obtain the first keyword group, specifically including: the processing module 302 extracts keywords from each first evaluation to obtain multiple keywords; the processing module 302 removes duplicates from each keyword to obtain a target keyword; the processing module 302 determines the frequency of each target keyword, one target keyword corresponds to one frequency, and the frequency calculation formula is:

[0084] ;

[0085] Among them, T d is the frequency, W1 is the number of times the target keyword appears in the first evaluation group, W2 is the total number of keywords included in the first evaluation group, A is the number of first evaluations included in the first evaluation group, and D1 is the number of first evaluations that include the target keyword; sort the frequencies from high to low to get the sorting results; from the sorting results, select the target keyword with the highest frequency and the preset number of keywords as the first keyword.

[0086] In a possible implementation, the second keyword group is composed of multiple second keywords, and the processing module 302 matches the first keyword group and the second keyword group to determine the proportion of similar keywords, specifically including: the processing module 302 calculates the comprehensive similarity between the target first keyword and the target second keyword, the target first keyword is any first keyword among the multiple first keywords, and the target second keyword is any second keyword among the multiple second keywords; the processing module 302 generates a keyword similarity matrix based on the comprehensive similarity; the processing module 302 applies a bipartite graph matching algorithm to the keyword similarity matrix to obtain similar keywords; the processing module 302 calculates the proportion of similar keywords in the first keyword group.

[0087] In a possible implementation, the processing module 302 calculates the comprehensive similarity between the target first keyword and the target second keyword, specifically including: the processing module 302 calculates the literal similarity between the target first keyword and the target second keyword; the processing module 302 calculates the semantic similarity between the target first keyword and the target second keyword; the processing module 302 determines whether the target first keyword and the target second keyword are synonyms; if the processing module 302 determines that the target first keyword and the target second keyword are synonyms, the synonym judgment result is determined to be 1; if the processing module 302 determines that the target first keyword and the target second keyword are not synonyms, the synonym judgment result is determined to be 0; the processing module 302 performs weighted summation on the literal similarity, semantic similarity and synonym judgment results according to a preset weight database to obtain a comprehensive similarity.

[0088] In a possible implementation, if the processing module 302 determines that the proportion is greater than or equal to the preset proportion threshold, before sending the employment enterprise to the unemployed students, the method also includes: the acquisition module 301 obtains the satisfaction score of the employed students for the employment enterprise; if the processing module 302 determines that the satisfaction score is greater than the preset satisfaction score, the preset proportion threshold is determined to be the preset first threshold; if the processing module 302 determines that the satisfaction score is less than or equal to the preset satisfaction score, the preset proportion threshold is determined to be the preset second threshold, and the preset first threshold is less than the preset second threshold.

[0089] In a possible implementation, before the processing module 302 pushes the recommendation list to the employed students, the method also includes: the acquisition module 301 acquires the intended positions of the employed students; the processing module 302 selects relevant positions from the recommendation list according to the intended positions, and the relevant positions are employment positions whose similarity with the intended positions is greater than a preset similarity; the processing module 302 puts the relevant positions at the front of the recommendation list, and sorts the relevant positions from high to low according to the proportion; the processing module 302 regards other positions after the relevant positions are screened out as remaining positions, and sorts the remaining positions from high to low according to the proportion; the processing module 302 uses the employment information corresponding to the sorted related positions and the employment information corresponding to the remaining positions as the recommendation list.

[0090] It should be noted that: when the device provided in the above embodiment realizes its function, only the division of the above functional modules is used as an example. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0091] The present application also provides an electronic device. Figure 4 , Figure 4 The electronic device 400 may include: at least one processor 401 , at least one network interface 404 , a user interface 403 , a memory 405 , and at least one communication bus 402 .

[0092] The communication bus 402 is used to realize the connection and communication between these components.

[0093] The user interface 403 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 403 may also include a standard wired interface and a wireless interface.

[0094] The network interface 404 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0095] Among them, the processor 401 may include one or more processing cores. The processor 401 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 405, and calling data stored in the memory 405. Optionally, the processor 401 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 401 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 401, and it can be implemented separately through a chip.

[0096] Among them, the memory 405 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 405 includes a non-transitory computer-readable storage medium. The memory 405 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 405 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 405 may optionally also be at least one storage device located away from the aforementioned processor 401. Refer to Figure 4 , the memory 405 as a computer storage medium may include an operating system, a network communication module, a user interface module and an application program of an employment recommendation method.

[0097] exist Figure 4In the electronic device 400 shown, the user interface 403 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 401 can be used to call an application program storing an employment recommendation method in the memory 405, and when executed by one or more processors 401, the electronic device 400 executes one or more of the methods described in the above embodiments. It should be noted that for the aforementioned method embodiments, for the sake of simple description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0098] The present application also provides a computer-readable storage medium, which stores instructions. When executed by one or more processors 401, the electronic device 400 executes one or more of the methods described in the above embodiments.

[0099] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0100] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which 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 service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0101] 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.

[0102] 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. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0103] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: various media that can store program codes, such as USB flash drives, mobile hard drives, magnetic disks or optical disks.

[0104] The above is only an exemplary embodiment of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification and the truth of practice, those skilled in the art will easily think of other embodiments of the present disclosure.

[0105] This application is intended to cover any variation, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the art not described in the present disclosure. The description and examples are to be regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A method for recommending employment, characterized in that: The method comprises: Obtaining employment information of employed students and a first evaluation group of the employed students by teachers based on a first reference basis during the period of study of the employed students in school, wherein the first reference basis is a reference content when the teacher evaluates the employed students, the first reference basis includes the academic performance data and school behavior records of the employed students, the first evaluation group is composed of a plurality of first evaluations, the first evaluation includes four evaluation dimensions of professional skills, communication skills, cooperation skills and learning ability, and the employment information includes the employed company and the job position; Extracting keywords from the first evaluation group to obtain a first keyword group; Obtaining a second evaluation group of the unemployed student by the teacher based on the second reference basis during the unemployed student's study at the school, wherein the second reference basis is the reference content when the teacher evaluates the unemployed student; Extracting keywords from the second evaluation group to obtain a second keyword group; Matching the first keyword group and the second keyword group to determine a ratio of similar keywords; If it is determined that the proportion is greater than or equal to the preset proportion threshold, the employment information is added to a recommendation list, and the employment information is sorted from high to low in the recommendation list according to the proportion; Pushing the recommendation list to the employed students; The first keyword group is composed of a plurality of first keywords, and the keyword extraction of the first evaluation group to obtain the first keyword group specifically includes: Extracting keywords from each of the first comments to obtain a plurality of keywords; De-duplication of each of the keywords to obtain a target keyword; Determine the frequency of each target keyword, one target keyword corresponds to one frequency, and the calculation formula of the frequency is: ； Among them, T d is the frequency, W1 is the number of times the target keyword appears in the first evaluation group, W2 is the total number of keywords included in the first evaluation group, A is the number of first evaluations included in the first evaluation group, and D1 is the number of first evaluations including the target keyword; Sort the frequencies from high to low to obtain a sorting result; From the ranking results, selecting the target keyword with the highest frequency and a preset number of keywords as the first keyword; The matching of the first keyword group and the second keyword group to determine the proportion of similar keywords specifically includes: Calculating a comprehensive similarity between a target first keyword and a target second keyword, wherein the target first keyword is any first keyword among the plurality of the first keywords, and the target second keyword is any second keyword among the plurality of the second keywords; Generating a keyword similarity matrix according to the comprehensive similarity; Applying a bipartite graph matching algorithm to the keyword similarity matrix to obtain similar keywords; The proportion of the similar keywords in the first keyword group is calculated.

2. The method according to claim 1, characterized in that: The obtaining of the employed enterprises of the employed students and the first evaluation group of the employed students by the teachers according to the first reference basis during the period of study of the employed students in the school specifically include: Constructing a standardized evaluation template, wherein the standardized evaluation template includes evaluation dimensions and evaluation indicators, wherein the evaluation dimensions include professional skills, communication skills, cooperation skills, and learning skills, and the evaluation indicators are quantified and refined for the evaluation dimensions; Associating and matching the first reference basis with the evaluation index to obtain structured index data; A first evaluation group is obtained by obtaining a plurality of teachers evaluating the employed students according to the structured indicator data.

3. The method according to claim 1, characterized in that The calculating of the comprehensive similarity between the target first keyword and the target second keyword specifically includes: Calculating the literal similarity between the target first keyword and the target second keyword; Calculating the semantic similarity between the target first keyword and the target second keyword; Determining whether the target first keyword and the target second keyword are synonyms; If it is determined that the target first keyword and the target second keyword are synonyms, then the synonym judgment result is determined to be 1; If it is determined that the target first keyword and the target second keyword are not synonyms, then determining the synonym judgment result to be 0; According to a preset weight database, the literal similarity, the semantic similarity and the synonym judgment result are weighted and summed to obtain the comprehensive similarity.

4. The method according to claim 1, characterized in that: If it is determined that the proportion is greater than or equal to the preset proportion threshold, before sending the employment information to the employed students, the method further includes: Obtaining the satisfaction rating of the employed students towards the employed enterprises; If it is determined that the satisfaction score is greater than the preset satisfaction score, then determining the preset proportion threshold as the preset first threshold; If it is determined that the satisfaction score is less than or equal to the preset satisfaction score, the preset proportion threshold is determined to be the preset second threshold, and the preset first threshold is less than the preset second threshold.

5. The method according to claim 1, characterized in that Before pushing the recommendation list to the job-seeking students, the method further includes: Obtain the intended positions of the employed students; Filtering out relevant positions from the recommendation list according to the intended position, wherein the relevant positions are employment positions whose similarity with the intended position is greater than a preset similarity; Arrange the related positions at the front of the recommendation list, and sort the related positions from high to low according to the proportion; The other positions after the relevant positions are screened out are regarded as the remaining positions, and the remaining positions are sorted from high to low according to the proportion; The employment information corresponding to the sorted related positions and the employment information corresponding to the remaining positions are used as the recommendation list.

6. An employment recommendation device, characterized in that: The device is used to execute the method according to any one of claims 1 to 5, and the device comprises an acquisition module (301) and a processing module (302), wherein: the acquisition module (301) and the processing module (302); The acquisition module (301) is used to acquire the employment information of employed students and a first evaluation group of the employed students by teachers based on a first reference basis during the period of study of the employed students in school, wherein the first reference basis is the reference content when the teacher evaluates the employed students, the first reference basis includes the academic performance data and school behavior records of the employed students, the first evaluation group is composed of a plurality of first evaluations, the first evaluation includes four evaluation dimensions of professional skills, communication skills, cooperation skills and learning ability, and the employment information includes the employed company and the job position; The processing module (302) is used to extract keywords from the first evaluation group to obtain a first keyword group; The acquisition module (301) is further used to acquire a second evaluation group of the unemployed student by the teacher based on a second reference basis during the unemployed student's study at school, wherein the second reference basis is a reference content when the teacher evaluates the unemployed student; The processing module (302) is further used to extract keywords from the second evaluation group to obtain a second keyword group; The processing module (302) is further used to match the first keyword group and the second keyword group to determine the proportion of similar keywords; The processing module (302) is further configured to add the employment information to a recommendation list if it is determined that the proportion is greater than or equal to a preset proportion threshold, and sort the employment information in the recommendation list from high to low according to the proportion; The processing module (302) is also used to push the recommendation list to the job-seeking student.

7. An electronic device, characterized in that: The electronic device (400) comprises a processor (401), a memory (405), a user interface (403) and a network interface (404), wherein the memory (405) is used to store instructions, the user interface (403) and the network interface (404) are used to communicate with other devices, and the processor (401) is used to execute the instructions stored in the memory (405) so that the electronic device (400) executes the method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1 to 5 is performed.

Images