A student employment ability evaluation method and system based on professional skill tags

Abstract

The application relates to a student employment ability evaluation method and system based on professional skill labels, and the method comprises the following steps: step S1, collecting professional text data D t and student employment data D W respectively; step S2, extracting professional skill labels W of students from the professional text data D t by using an unsupervised word mining algorithm; and step S3, using the employment data D W as a supervision signal, further mapping the extracted professional skill labels W by using a neural network algorithm, and obtaining professional skill features representing the employment ability of students after training. Compared with the prior art, the application obtains higher granularity employment ability dimension features, and improves the accuracy of student employment evaluation.

Description

Technical Field

[0001] This invention relates to the field of data matching technology, and in particular to a method and system for assessing student employability based on professional skills tags. Background Technology

[0002] The increasingly severe employment situation for college graduates has led to greater emphasis on career guidance and planning for university students. With the application of artificial intelligence and machine learning in college job matching and career planning, the college career guidance service industry is becoming increasingly refined and personalized.

[0003] Student competency dimensions, especially professional skills, are a crucial feature. However, many current employment competency matching methods lack the construction of a professional skills competency dimension. Some methods directly use students' comprehensive quality evaluation scores or professional course grades to assess their professional skills dimension and directly use these outputs as features for employment matching and career recommendations. This approach has two drawbacks. First, it provides a coarse-grained description, failing to finely depict students' professional and technical abilities. Second, because competency dimension features do not establish a direct link with employment fields and job positions—that is, employment data is not used as supervisory information during feature construction—the expressive power of competency dimensions is lacking. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the existing technology by providing a method and system for assessing student employability based on professional skills tags, thereby improving the accuracy of student employment assessment through more granular employability dimension features.

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] According to a first aspect of the present invention, a method for assessing student employability based on professional skills tags is provided, the method comprising the following steps:

[0007] Step S1: Collect professional text data D separately t And graduate employment data D W ;

[0008] Step S2: Employ an unsupervised vocabulary mining algorithm to extract vocabulary from professional text data D. t Extract the student's professional skills tag W;

[0009] Step S3: Transfer employment data D W As a supervisory signal, a neural network algorithm is used to further map the extracted professional skill labels W, and after training, professional skill features that characterize students' employability are obtained.

[0010] Step S4: Evaluate students' employability based on professional skills characteristics and weight settings.

[0011] Preferably, the employment data in step S1 includes job positions and employment sectors.

[0012] Preferably, the professional text data in step S1 includes internal data and external data;

[0013] The internal data includes text data related to major introductions, curriculum descriptions, and elective courses.

[0014] The external data includes professionally interpreted text data from job websites.

[0015] Preferably, step S2 includes the following sub-steps:

[0016] Step S21: Use a word segmentation tool to process the professional text data Dt for stop words and word segmentation, and construct a trie based on the segmentation results;

[0017] Step S22: Calculate the left and right entropy and mutual information according to the constructed trie, and add the left and right entropy and mutual information to get the score. After sorting, select the top N words with the highest scores as candidate words.

[0018] Step S23: Perform semantic screening on candidate words, segment the corpus using candidate words, train the word2vec model, and obtain word vectors of candidate words;

[0019] Step S24: Select manually annotated words from the candidate words as the center words, and use word vectors to calculate the distance between other candidate words and the center words to filter the candidate words;

[0020] Step S25: Further filter the candidate words obtained earlier according to rules or manually to obtain professional skill vocabulary;

[0021] Step S26: Obtain the corresponding professional skill tag vocabulary W for each student's internal data.

[0022] Preferably, the word segmentation tool in step S21 is the jieba word segmentation tool.

[0023] Preferably, step S3 includes the following sub-steps:

[0024] Step S31: Based on the graduate employment data D W Constructing a classification learning dataset D c = {W, y}, where W represents the professional skills tags of the graduates and y represents the employment field of the graduates;

[0025] Step S32: Input vocabulary list Wi After passing through the Bert encoder network, the professional competence feature x1 is obtained;

[0026] Step S33: Input the professional ability feature x1 into the decoder network, and output q(x) after passing through the softmax activation function. Use a multi-classification loss function for training.

[0027] Step S34: Using the trained network, the professional ability feature x1 is used as the feature vector of the professional skills dimension in the student's employability dimension.

[0028] Preferably, the loss function expression in step S33 is:

[0029]

[0030] In the formula, p(x) i ) is the true label of category i; n is the number of categories.

[0031] According to a second aspect of the present invention, a student employability assessment system based on professional skills tags is provided, employing any one of the methods described above, the system comprising:

[0032] The data acquisition module is used to collect professional text data and employment data;

[0033] The professional skills tag extraction module is used to extract professional skills tags from professional text data using an unsupervised vocabulary mining algorithm;

[0034] The professional skills feature solving module is used to map professional skills labels to employment data in order to extract professional skills features that represent students' employability.

[0035] According to a third aspect of the present invention, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the program to implement any of the methods described above.

[0036] According to a fourth aspect of the invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements any of the methods described herein.

[0037] Compared with the prior art, the present invention has the following advantages:

[0038] The method of this invention extracts professional skill tags through unsupervised vocabulary mining and uses student employment data as supervisory information to train a neural network, which can obtain more accurate, fine-grained, and employment-strongly correlated student employability dimension features, thereby improving the accuracy of student employability assessment. Attached Figure Description

[0039] Figure 1 This is a flowchart of the method of the present invention;

[0040] Figure 2 This is a schematic diagram of the professional skill tag extraction process of the present invention;

[0041] Figure 3 This is a schematic diagram illustrating the process of constructing the professional capability dimension of this invention. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0043] Example

[0044] like Figure 1 As shown in the figure, this embodiment provides a method for assessing students' employability based on professional skills tags. The method includes the following steps:

[0045] Step S1: Collect professional text data D separately t and employment data D W = {S, y}; where S represents the graduates and y represents the fields of employment for the graduates;

[0046] The professional text data includes internal and external data; the internal data includes text data related to major introductions, training programs, and elective courses, while the external data includes professional interpretation texts from employment websites; the employment data includes graduates' job positions and employment fields.

[0047] Step S2, based on professional text data D t An unsupervised vocabulary mining algorithm is used to extract students' professional skill tags, such as... Figure 2 As shown, specifically;

[0048] Step S21: Use the jieba word segmentation tool to segment the professional text data D. t Perform stop word processing and word segmentation, and construct a trie using the segmentation results;

[0049] Step S22: Calculate the left and right entropy and mutual information based on the constructed trie, and add the left and right entropy and mutual information together as the score. After sorting according to the score, select the N words with the highest scores as candidate words.

[0050] Step S23: Perform semantic filtering on candidate words. Specifically, this involves segmenting the corpus using candidate words, training a word2vec model, and obtaining word vectors for the candidate words.

[0051] Step S24: Manually annotate some words from the candidate words as the center words, and use the previously obtained word vectors to calculate the distance between other candidate words and the center words to filter the candidate words;

[0052] Step S25: Further filter the candidate words obtained earlier according to rules or manually to obtain professional skill vocabulary;

[0053] Step S26: Obtain the corresponding professional skill tag vocabulary W for each student's internal data.

[0054] Step S3: Construct professional skill characteristics, specifically as follows:

[0055] Based on professional skills tag vocabulary W, and using employment data D W As a supervisory signal, the extracted labels are further mapped to obtain professional skills characteristics that can characterize students' employability:

[0056] X = q(D) W W)

[0057] In the formula, X represents the professional skill characteristic; g represents the calculation process of the professional skill characteristic, such as... Figure 3 As shown, specifically:

[0058] Step S31: Based on the graduate employment data D W Constructing a classification learning dataset D c = {W, y}, where W represents the professional skills tags of the graduates and y represents the employment field of the graduates;

[0059] Step S32: Input vocabulary list W i After passing through the Bert encoder network, the professional competence feature x1 is obtained;

[0060] Step S33: Input the professional ability feature x1 into the decoder network, and output q(x) after passing through the softmax activation function. Train the network using a multi-classification loss function; the expression for the loss function is:

[0061]

[0062] In the formula, p(x) i ) is the true label of category i; n is the number of categories;

[0063] Step S34: Using the trained network, the professional ability feature x1 is used as the feature vector of the professional skills dimension in the student's employability dimension.

[0064] Next, a system embodiment of the present invention is given, a student employability assessment system based on professional skill tags, employing the method described above, the system comprising:

[0065] The data acquisition module is used to collect professional text data and employment data;

[0066] The professional skills tag extraction module is used to extract professional skills tags from professional text data using an unsupervised vocabulary mining algorithm;

[0067] The professional skills feature solving module is used to map professional skills labels to employment data in order to extract professional skills features that represent students' employability.

[0068] The electronic device of this invention includes a central processing unit (CPU), which can perform various appropriate actions and processes according to computer program instructions stored in read-only memory (ROM) or loaded from a storage unit into random access memory (RAM). The RAM may also store various programs and data required for device operation. The CPU, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0069] Multiple components in the device are connected to the I / O interface, including: input units such as keyboards and mice; output units such as various types of displays and speakers; storage units such as disks and optical discs; and communication units such as network interface cards (NICs), modems, and wireless transceivers. The communication unit allows the device to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0070] The processing unit executes the various methods and processes described above, such as methods S1 to S4. For example, in some embodiments, methods S1 to S4 may be implemented as computer software programs tangibly contained in a machine-readable medium, such as a storage unit. In some embodiments, part or all of the computer program may be loaded and / or installed on the device via ROM and / or a communication unit. When the computer program is loaded into RAM and executed by the CPU, one or more steps of methods S1 to S4 described above may be performed. Alternatively, in other embodiments, the CPU may be configured to execute methods S1 to S4 by any other suitable means (e.g., by means of firmware).

[0071] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: field programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload programmable logic devices (CPLDs), and so on.

[0072] The program code used to implement the methods of the present invention can be written in any combination of one or more programming languages. This program code can be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code can be executed entirely on the machine, partially on the machine, as a standalone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0073] In the context of this invention, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0074] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for assessing student employability based on professional skills tags, characterized in that, The method includes the following steps: Step S1: Collect professional text data separately. and graduate employment data ; Step S2: Employ an unsupervised vocabulary mining algorithm to extract vocabulary from professional text data. Extract students' professional skills tag vocabulary ; Step S3: Data on employment As a supervisory signal, a neural network algorithm is used to analyze the extracted professional skill tag vocabulary. Further mapping and training yield professional skill characteristics that represent students' employability. Step S4: Evaluate students' employability based on professional skill characteristics and weight settings; The professional text data in step S1 Includes internal and external data; The internal data includes text data related to major introductions, curriculum descriptions, and elective courses. The external data includes professionally interpreted text data from job websites; Step S2 includes the following sub-steps: Step S21: Use a word segmentation tool to process professional text data. Stop word processing and word segmentation are performed, and a trie is constructed based on the segmentation results; Step S22: Calculate the left and right entropies and mutual information based on the constructed trie, and add the left and right entropies and mutual information together as the score. After sorting, select the top scorers. One word was selected as a candidate word; Step S23: Perform semantic screening on candidate words, segment the corpus using candidate words, train the word2vec model, and obtain word vectors of candidate words; Step S24: Select manually annotated words from the candidate words as the center words, and use word vectors to calculate the distance between other candidate words and the center words to filter the candidate words; Step S25: Further filter the candidate words obtained earlier according to rules or manually to obtain professional skill vocabulary; Step S26: Obtain the corresponding professional skill tag vocabulary for each student based on their internal data. ; Step S3 includes the following sub-steps: Step S31: Based on graduate employment data Constructing a classification learning dataset ,in, Vocabulary indicating the professional skills of graduates Indicates the fields of employment for graduates; Step S32: Input vocabulary list Professional competency characteristics are obtained after passing through the Bert encoder network. ; Step S33: Determine professional competence characteristics The input is fed into the BERT decoder network, and after passing through the softmax activation function, the output is... The multi-class loss function is used for training. Step S34: Using the trained Bert encoder network, the professional ability features are... This serves as the feature vector for the professional skills dimension within the student's employability dimension.

2. The student employability assessment method based on professional skills tags according to claim 1, characterized in that, The employment data in step S1 This includes job positions and employment fields.

3. The student employability assessment method based on professional skills tags according to claim 1, characterized in that, The word segmentation tool in step S21 is the jieba word segmentation tool.

4. The student employability assessment method based on professional skills tags according to claim 1, characterized in that, The expression for the multi-class loss function in step S33 is: In the formula, It is a category The true label; The number of categories.

5. A student employability assessment system based on professional skills tags, characterized in that, The system comprising the method according to any one of claims 1 to 4, wherein the system includes: The data acquisition module is used to collect professional text data and employment data; The professional skills tag extraction module is used to extract professional skills tags from professional text data using an unsupervised vocabulary mining algorithm; The professional skills feature solving module is used to map professional skills labels to employment data in order to extract professional skills features that represent students' employability.

6. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1 to 4.

Images