CN119919252A

CN119919252A - A training management system, device and storage medium based on intelligent recommendation strategy

Info

Publication number: CN119919252A
Application number: CN202411821718.7A
Authority: CN
Inventors: 戴训华; 齐中国; 陶俊; 胡志华; 尧德鹏
Original assignee: Jiangxi Telecom Information Industry Co ltd
Current assignee: Jiangxi Telecom Information Industry Co ltd
Priority date: 2024-12-11
Filing date: 2024-12-11
Publication date: 2025-05-02

Abstract

The present invention discloses a training management system based on intelligent recommendation strategy, including: a data processing module for obtaining employee information and course information, and performing preprocessing operations; a course recommendation module for converting employee information and course information into feature vectors for clustering, and substituting the employee clusters and course clusters obtained after clustering into a deep neural network model for deep learning, so as to obtain recommended courses for employee clusters and employees suitable for course clusters; an examination supervision module for taking examinations after employees have learned the recommended courses, and using anti-cut screen technology for automatically determining cheating and automatically handing in the examination paper when the examinee switches to other programs; a system update module for updating the system in real time according to changes in data in the system, regularly evaluating and adjusting the recommendation algorithm, and establishing a user feedback mechanism. The present invention can make personalized recommendations based on information such as employees' job positions and years of service.

Description

Training management system, device and storage medium based on intelligent recommendation strategy

Technical Field

The invention relates to the technical field of intelligent recommendation, in particular to a training management system, training management equipment and a storage medium based on an intelligent recommendation strategy.

Background

In the current rapid-development digital age, a training management system is an important tool for improving the capacity of staff or students for enterprises, education institutions and the like, the system generally integrates functions of course management, learning progress tracking, online testing and evaluation and the like, aims to provide an efficient and flexible learning environment, is constructed based on a cloud computing architecture in technical level, utilizes big data analysis to personally recommend learning content, in addition, with the popularization of the mobile Internet, supporting cross-platform access (including but not limited to PC end and mobile phone APP) is also a great feature of the system, so that training forms are greatly enriched, training effects are remarkably improved, and the training management system can be widely applied in daily use and is suitable for training requirements in different fields of online learning platforms, mental health management and the like.

For example, an online learning platform course intelligent recommendation method with a publication number of CN115222564A in the prior art carries out hot rating processing on different lesson-generation teachers, and can recommend different teachers and different teaching materials according to different learning conditions of different students, so that the whole learning effect is improved, different students can obtain the learning improvement effect, or a psychological health course personalized intelligent recommendation system and method with a publication number of CN114969545A in the prior art can provide similar content recommendation service for users through keywords.

However, in the practical use process, the technology adopted in the prior art has weak pertinence and scientificity, or has low recommendation accuracy and low user experience, and the requirements of different users on training resources are quite different and cannot meet personalized learning requirements.

Therefore, the application provides a training management system based on an intelligent recommendation strategy to solve the technical problems.

Disclosure of Invention

The invention mainly aims to provide a training management system based on intelligent recommendation strategies, which improves training efficiency, ensures training quality and improves satisfaction of students through informatization means, realizes whole-course tracking management of learning conditions of the students and comprehensive grasp of learning and training requirements of staff through online participated course training, examination competition, test question practice, questionnaires, training communication and other conditions, and finally creates a one-stop service platform for whole-flow online training and examination so as to solve the technical problems in the background technology.

The invention adopts the following technical scheme to solve the technical problems:

The training management system based on the intelligent recommendation strategy comprises a data processing module, a course recommendation module, an examination supervision module and a system update module, wherein:

the data processing module is used for accessing an external manpower system, acquiring employee information and course information from the manpower system and preprocessing the employee information and the course information;

the course recommendation module is used for converting the employee information and the course information which are subjected to basic pretreatment into feature vectors for clustering, and substituting the employee clusters and the course clusters which are obtained after clustering into the deep neural network model for deep learning so as to obtain recommended courses of the employee clusters and employees suitable for the course clusters;

The examination supervision module is used for establishing an examination system by adopting a VUE technology, realizing user interaction through a custom service based on JSON, and carrying out examination of related courses after the recommended courses are learned by staff, and in the running process of the examination system, the examination supervision module is used for logging in by using a face recognition technology, and the examination supervision module is internally provided with an anti-screen-cutting technology and is used for automatically judging cheating and automatically delivering papers when an examinee switches to other programs;

and the system updating module is used for updating the system in real time according to the change of the data in the system, monitoring the state and the running state of the system in real time, regularly backing up and maintaining the data, regularly updating software and hardware, regularly evaluating and adjusting a recommendation algorithm and establishing a user feedback mechanism.

Preferably, the data processing module obtains information and preprocessing operation internally includes:

staff information, namely 4A account number, name, department, post name, working year, learning progress, learning duration and examination result;

course information, namely course names, course duration and course scores;

And preprocessing, namely executing data cleaning operation and deleting employee information which is already away from the job and unsuitable course information.

Preferably, the specific processing steps of employee information and course information in the course recommendation module include:

S1, converting employee information into feature vectors to obtain employee information feature vectors, calculating the similarity of the employee information feature vectors by using cosine similarity, and clustering the similarity of the employee information feature vectors by using a DBSCAN clustering algorithm to obtain employee clusters, wherein the concrete calculation formula of the cosine similarity calculation is as follows:

Wherein x represents an employee information feature vector of the first employee, y represents an employee information feature vector of the second employee, and θ represents an included angle with the coordinate axis;

S2, converting course information into feature vectors to obtain course information feature vectors, calculating course information feature vector similarity by using cosine similarity, and clustering the course information feature vector similarity by using a DBSCAN clustering algorithm to obtain course class clusters, wherein a concrete calculation formula of the cosine similarity calculation is as follows:

Wherein w represents an employee information feature vector of the first employee, z represents an employee information feature vector of the second employee, and μ represents an included angle with the coordinate axis;

s3, constructing a deep convolutional neural network model, and inputting the employee cluster and the course information feature vector into the deep neural network model to obtain a recommended course of the employee cluster;

s4, inputting the course class cluster and the employee information feature vector into a deep neural network model to obtain an employee with a proper course class cluster, and recommending the course class cluster to the employee.

Preferably, the specific implementation step of clustering by the DBSCAN clustering algorithm includes:

l1, establishing a data set for initialization, selecting data from the data set, calculating three-dimensional characteristics of user information, and presetting a user quantity threshold value a which accords with user conditions;

l2, starting a cycle, and presetting cluster parameters e and Mi npts of the cycle, wherein e is expressed as a cluster field radius, mi npts is expressed as the minimum field point number of which a given point becomes a core object in the cluster field;

And L3, randomly extracting the user from the data, if the user is not visited, calculating whether at least Mi npts data points exist in the neighborhood of the user, if the user has more than Mi npts data points, establishing a new cluster to put the user in, repeating the calculation process for other data points in the e-neighborhood of the user to expand the cluster, and ending the expansion of the cluster when the situation that at least Mi npts data points can be contained in the neighborhood without the data points is detected.

Step 4, repeating the step 3 until all users in the data set are accessed, selecting the cluster which most accords with the user characteristics from the obtained clusters, setting the number of times of selecting the clusters as K, and selecting the generations to perform clustering operation;

and L5, when K > a, indicating that the selected user meets the requirement, and ending the iteration.

Preferably, the specific operation steps for performing the clustering operation in the step L4 include:

In the first clustering of the clustering, selecting an abnormal user area and a normal user area, wherein the normal user area is marked as a low risk 1 area and is represented as a cluster which is most in line with the behavior characteristics of the normal user, and in the subsequent clustering process, the abnormal user cluster is not selected any more and only the cluster which is in line with the behavior characteristics of the normal user is selected;

In the data set range reduction process, a cluster conforming to malicious user behaviors is defined as a medium risk user area, the medium risk user area is not marked at this time, clusters marked as normal users in the cluster and the subsequent clusters are divided into a low risk 2 area and a low risk 3 area until a low risk k area, the numerical value of the three-dimensional index of the area user is different from that of the low risk 1 area user, and the area user is used as a user of a normal operation virtual machine, and finally the area user enters the low risk 1 area with probability.

Preferably, the specific operation steps of constructing the deep convolutional neural network model in the step S3 include sequentially constructing an embedded layer, a multi-head attention layer, a convolutional layer, a pooling layer and an output layer of the convolutional neural network, wherein:

the embedded layer is used for converting the employee information feature vector into a digital matrix according to the length of the feature vector and outputting the digital matrix to the next layer;

and P2, a multi-head attention layer, which is used for selecting important features and acquiring potential factors from staff information feature clusters by introducing a multi-head attention mechanism, wherein the multi-head attention mechanism specifically comprises the steps of firstly carrying out linear transformation on vector information, inputting the vector information into a zooming dot product attention, carrying out total h times, calculating a head each time, wherein parameters among heads are not shared, splicing zooming dot product attention results of the h times, and finally carrying out linear transformation on values as multi-head attention results, namely:

head_i＝Attention(QWQ_i,KWK_i,VWV_i)

QWQ _i is represented as a result of a linear transformation of the query vector Q _i, KWK _i is represented as a result of a linear transformation of the key vector K _i, and VWV _i is represented as a result of a linear transformation of the value vector V _i;

P3. a convolution layer, which is used for extracting adjacent feature vectors by adopting a one-dimensional convolution check feature vector matrix with window sizes of 1, 3, 5 and 7 respectively, so as to extract context features;

p4. pooling layer, extracting the previous k large value from each pooling block of convolution layer as representative feature by k-max-pooling technology, and processing variable length document by merging operation of constructing fixed length feature vector, wherein extracting representative feature is:

wherein: Represented by the known parameters W, X and variances Where V is observed, N (-) is expressed as a positive too much distribution, V _j is the j-th observation, MACNN _W(X_j) refers to the convolutional neural network through which the parameter W passes,Refers to covariance matrix asA noise model of the multiple identity matrix I;

p5. output layer, K-dimensional space of potential factors of user projected for recommended task, and finally potential vector of document is generated by using conventional nonlinear projection, so that high-level function obtained from upper layer can be used for specific task by conversion.

Preferably, the examination system inside the examination supervision module uses a video technology of combining SRS with WebSocket protocol, wherein:

Establishing connection on the SRS server by using the HTTP protocol, and upgrading to the WebSocket protocol;

When establishing connection, a client sends an HTTP request to a server, wherein the request header contains an Upgrade field, and the value of the Upgrade field is WebSocket;

After receiving the request, if the server supports the WebSocket protocol, returning an HTTP response, wherein the response header contains an Upgrade field with the value of WebSocket, and also contains a Sec-WebSocket-Accept field with the value obtained by encrypting a Sec.WebSocket-Key field and a fixed GUID character string sent by the client;

After the client receives the response, if the response header contains an Upgrade field, the value is WebSocket, and the value of the Sec-WebSocket-Accept field is the same as the Sec-WebSocket-Key field encryption result sent by the client, the connection is successfully established, and the WebSocket communication can be started.

Preferably, the screen-cutting prevention technology in the examination supervision module is used for screen-cutting prevention detection in the webpage program, and specifically comprises the following steps:

defining a page currently being used by a user as a focus page, wherein the focus page covers a common page;

and detecting whether the focus page state is maintained in the examination process, and if the focus page state is lost in the examination process, considering that the examinee executes the screen cutting operation to trigger the screen cutting cheating rule.

In yet another aspect, the invention also discloses a computer readable storage medium storing a computer program which, when executed by a processor, causes the processor to perform the internal algorithm steps of the above system.

In yet another aspect, the invention also discloses a computer device comprising a memory and a processor, the memory storing a computer program which, when executed by the processor, causes the processor to perform the internal algorithm steps of the system as above.

According to the technical scheme, the training management system based on the intelligent recommendation strategy is provided. Compared with the prior art, the invention has the following advantages:

1. According to the invention, the DBSCAN clustering algorithm and the deep convolutional neural network model are arranged in the course recommendation module, so that the functions of accurately identifying staff and course characteristics and effectively clustering can be realized, personalized recommendation can be realized according to information such as working positions and working years of the staff, proper courses can be recommended to similar staff clusters, and the course clusters can be recommended to proper staff, therefore, the accuracy of recommendation can be improved by comprehensive double recommendation, and the pertinence and satisfaction of staff learning can be improved.

2. According to the method, the multi-head attention mechanism is arranged in the course recommendation module, so that the function of acquiring potential factors from the employee information feature clusters can be achieved, the effect of improving accuracy of recommendation algorithms is achieved, and finally learning requirements of different employees can be better met.

3. The invention adopts the video technology combining the Vue technology, the JSON interaction and the SRS and the WebSocket protocol in the examination supervision module, and can play a role in improving the response speed of the system and the user interaction experience, thereby achieving the real-time and efficient examination supervision effect and finally ensuring the fairness and the safety of the examination process.

4. According to the invention, the face recognition login and the screen-cut prevention detection technology are arranged in the examination supervision module, so that the function of preventing cheating is achieved, the effect of maintaining examination fairness is achieved, and finally the authenticity and credibility of examination results can be ensured.

It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the invention or to delineate the scope of the invention. Other features of the present invention will become apparent from the description that follows. Of course, it is not necessary for any one product to practice the invention to achieve all of the advantages set forth above at the same time.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application. In the drawings:

FIG. 1 is a schematic diagram of the overall steps of the system of the present invention;

FIG. 2 is a schematic diagram of a system module frame of the present invention;

FIG. 3 is a schematic diagram illustrating an overall operation procedure of an internal system of the course recommendation module according to the present invention.

Detailed Description

The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the application. Embodiments of the application and features of the embodiments may be combined with each other without conflict. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

In the examples, see fig. 1 to 3 in detail.

As shown in fig. 1 and fig. 2, the training management system based on the intelligent recommendation policy provided by the embodiment of the invention includes a data processing module, a course recommendation module, an examination supervision module, and a system update module, wherein:

(1) The data processing module is used for accessing an external manpower system, acquiring employee information and course information from the manpower system and preprocessing the employee information and the course information;

the employee information comprises a 4A account number, a name, a department, a post name, a working year, a learning progress, a learning duration and an examination result;

The course information comprises a course name, a course duration and a course score;

The preprocessing operation comprises performing data cleaning operation, such as deleting employee information and unsuitable course information which are already away from the job;

in conclusion, the data processing module performs data cleaning by setting a preprocessing function, so that the information of the workers who leave and the information of the courses which are not suitable for the workers can be deleted, the data quality and accuracy are improved, and the data base of the follow-up recommending and supervising module can be ensured to be more reliable;

(2) The course recommendation module is used for converting the employee information and the course information which are subjected to basic pretreatment into feature vectors for clustering, and substituting the employee clusters and the course clusters which are obtained after clustering into the deep neural network model for deep learning so as to obtain recommended courses of the employee clusters and employees suitable for the course clusters;

the specific processing steps of employee information and course information in the course recommendation module are shown in fig. 3, including:

S1, converting employee information into feature vectors to obtain employee information feature vectors, calculating the similarity of the employee information feature vectors by using cosine similarity, and clustering the similarity of the employee information feature vectors by using a DBSCAN clustering algorithm to obtain employee clusters, wherein a concrete calculation formula for calculating the cosine similarity is as follows:

S2, converting the course information into feature vectors to obtain course information feature vectors, calculating the similarity of the course information feature vectors by using cosine similarity, and clustering the similarity of the course information feature vectors by using a DBSCAN clustering algorithm to obtain course class clusters, wherein a concrete calculation formula of the cosine similarity calculation is as follows:

The specific implementation steps of clustering by the DBSCAN clustering algorithm in the step S1 and the step S2 include:

L1, establishing a data set for initialization, selecting data from the data set, calculating three-dimensional characteristics of each piece of user information, determining the number of users approximately conforming to user conditions by observing the data, and presetting a user number threshold value a conforming to the user conditions;

In particular, the specific operation steps of selecting generation to perform clustering operation at this time include:

In the first clustering of the clustering, the abnormal user area and the normal user area are selected, the normal user area at the moment is marked as a low risk 1 area and is expressed as a cluster which is most in line with the behavior characteristics of the normal user, and in the subsequent clustering process, the abnormal user cluster is not selected any more and only the cluster which is in line with the behavior characteristics of the normal user is selected, because the cluster which is in line with the malicious user behavior at the moment cannot be directly defined as the abnormal user due to the reduction of the range of the data set, the cluster which is in line with the malicious user behavior is defined as the medium risk user area, the medium risk user area is not marked, the cluster which is marked as the normal user in the present clustering and the subsequent clustering is divided into a low risk 2 area and a low risk 3 area until the low risk K area (K is the selected number of the clustering), and the users still have the behavior characteristics of the normal user, but the numerical value of the three-dimensional index of the user is possibly different from the user in the low risk 1 area, and the users still are users of the normal running virtual machine and can finally enter the low risk 1 area;

And L5, when K is larger than a, indicating that the selected user meets the requirement, and ending the iteration;

s3, constructing a deep convolutional neural network model, and inputting employee clusters and course information feature vectors into the deep neural network model to obtain recommended courses of the employee clusters;

It may be further additionally described that the specific operation steps of constructing the deep convolutional neural network model include sequentially constructing an embedded layer, a multi-head attention layer, a convolutional layer, a pooling layer and an output layer of the convolutional neural network, wherein:

And P1, an embedding layer, which is used for converting the employee information feature vector into a digital matrix according to the length of the feature vector and outputting the digital matrix to the next layer, for example, if one feature vector exists, the embedded vector in the employee information feature is represented by the matrix. Then, further training the feature vectors through an optimization process;

P2. a multi-head attention layer for selecting important features and obtaining potential factors from staff information feature clusters by introducing a multi-head attention mechanism between an embedding layer and a convolution layer of a convolution neural network, wherein the multi-head attention mechanism specifically comprises the steps of firstly carrying out linear transformation on vector information, inputting the vector information into scaled dot product attention, carrying out total calculation for h times, calculating parameters between heads each time without sharing, then splicing scaled dot product attention results of the h times, and finally carrying out linear transformation to obtain values as multi-head attention results, wherein the multi-head attention is different in that h times of calculation are carried out instead of only one time, namely:

head_i＝Attention(QWQ_i,KWK_i,VWV_i)

p5. output layer, which projects K-dimensional space of potential factors of user for recommended task, and finally generates potential vector of document by using conventional nonlinear projection, so that high-level function obtained from upper layer can be used for specific task by conversion;

therefore, by setting a multi-head attention mechanism in the course recommendation module, the function of acquiring potential factors from the employee information feature clusters can be achieved, the accuracy of a recommendation algorithm is improved, and the learning requirements of different employees can be better met;

S4, inputting the course class clusters and the employee information feature vectors into a deep neural network model, obtaining the employees suitable for the course class clusters, and recommending the course class clusters to the employees;

In conclusion, the course recommendation module can accurately identify staff and course characteristics and effectively cluster the staff by setting feature vector conversion, cosine similarity calculation, DBSCAN clustering algorithm and deep convolutional neural network model, can conduct personalized recommendation according to information such as working positions and working years of staff, can recommend proper courses for similar staff clusters, and can recommend course clusters to proper staff, so that accuracy of recommendation can be improved by comprehensive double recommendation, and pertinence and satisfaction of staff learning are improved;

(3) The examination supervision module is used for establishing an examination system by adopting a VUE technology, interacting based on JSON, realizing user interaction through rich service and custom service, and performing examination of related courses after staff learn recommended courses, and logging in by using a face recognition technology in the operation process of the examination system, wherein the examination supervision module is internally provided with an anti-screen-cutting technology and is used for automatically judging cheating and automatically delivering papers when an examinee switches to other programs;

Specifically, an examination system in the examination supervision module uses a video technology combining SRS and WebSocket protocols, wherein the WebSocket on an SRS server establishes connection through an HTTP protocol and then upgrades to the WebSocket protocol; when establishing connection, a client sends an HTTP request to a server, wherein the request header contains an Upgrade field with the value of WebSocket, after the server receives the request, if the request supports the WebSocket protocol, an HTTP response is returned, the response header contains the Upgrade field with the value of WebSocket, meanwhile, the response header also contains a Sec-WebSocket-Accept field with the value of Sec-WebSocket-Key field sent by the client and a fixed GU ID character string are encrypted, and after the client receives the response, if the response header contains the Upgrade field with the value of WebSocket, and the value of Sec-WebSocket-Key field is the same as the encryption result of Sec-WebSocket-Key field sent by the client, the connection can be successfully established, and the WebSocket communication can be started;

Therefore, the examination supervision module can improve the response speed of the system and the user interaction experience by adopting a video technology combining a Vue technology, JSON interaction and SRS and WebSocket protocol, thereby achieving the real-time and efficient examination supervision effect and ensuring the fairness and the safety of the examination process;

The screen-cutting prevention technology in the examination supervision module is used for screen-cutting prevention detection in a webpage program and specifically comprises the steps of defining a page currently being used by a user as a focus page, enabling the focus page to cover a common page, enabling screen-cutting prevention detection to be network examination software to check whether the state of the focus page is maintained or not, detecting whether the state of the focus page is maintained or not in the examination process, and enabling an examinee to execute screen-cutting operation if the state of the focus page is lost at a certain moment by the network examination software, so that screen-cutting cheating rules are triggered, and the screen-cutting prevention function principle is adopted in online examination. For example, in the wi ndows system, if the user opens multiple folders, the top color of the folder being operated will be darkened and topped, while the top of the other unoperated folders will appear light, which is the difference between the focus page and the normal page. The same holds true for programs and programs. The focus page usually covers the common page, so that the use of a user is more convenient, and the network examination software is only one webpage program;

therefore, the face recognition login and the screen-cut prevention detection technology are arranged in the examination supervision module, so that the function of preventing cheating can be achieved, examination fairness is maintained, and the authenticity and the credibility of examination results are ensured;

(4) The system updating module is used for updating the system in real time according to the change of the data in the system, monitoring the state and the running state of the system in real time, regularly backing up and maintaining the data, regularly updating software and hardware, regularly evaluating and adjusting a recommendation algorithm and establishing a user feedback mechanism;

at this time, the system updating module can play a role in keeping the stable operation and continuous optimization of the system by setting real-time data updating, periodical backup and maintenance, software and hardware updating, system state monitoring and user feedback mechanisms, so that the overall performance and user experience of the system are improved, and the long-term reliability and user satisfaction of the system can be ensured.

In summary, the system can bear a training course system constructed by enterprises based on job position and job position requirements in the actual use process, and records various internal and external training course data (including accessory document videos and the like), wherein the training courses and the data thereof can be disclosed to appointed staff on the internet as learning data, and meanwhile, the system comprises a training lecturer and management of a training channel related to the training courses.

In yet another aspect, the invention also discloses a computer readable storage medium storing a computer program which, when executed by a processor, causes the processor to perform the steps of the method as described above.

In yet another aspect, the invention also discloses a computer device comprising a memory and a processor, the memory storing a computer program which, when executed by the processor, causes the processor to perform the steps of the method as above.

In yet another embodiment of the present application, there is also provided a computer program product containing instructions that, when run on a computer, cause the computer to perform the internal module algorithm steps of any of the intelligent recommendation policy based training management systems of the above embodiments.

It may be understood that the system provided by the embodiment of the present invention corresponds to the method provided by the embodiment of the present invention, and explanation, examples and beneficial effects of the related content may refer to corresponding parts in the above method.

The embodiment of the application also provides an electronic device, which comprises a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory are communicated with each other through the communication bus,

A memory for storing a computer program;

And the processor is used for realizing the algorithm steps of the internal module of the training management system based on the intelligent recommendation strategy when executing the programs stored in the memory.

The communication bus mentioned above for the electronic device may be a peripheral component interconnect standard (PC I) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus may be classified as an address bus, a data bus, a control bus, or the like.

The communication interface is used for communication between the electronic device and other devices.

The memory may include Random Access Memory (RAM) or may include non-volatile memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one memory device located remotely from the aforementioned processor.

The processor may be a general-purpose processor including a Central Processing Unit (CPU), a Network Processor (NP), etc., or may be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device, or discrete hardware components.

It should be noted that the electronic device further includes a terminal device, which may also be referred to as a terminal, a user equipment, a mobile station, a mobile terminal, or the like. The terminal device may be a mobile phone, a smart television, a wearable device, a tablet computer, a computer with a wireless transceiving function, a virtual reality terminal device, an augmented reality terminal device, a wireless terminal in industrial control, a wireless terminal in unmanned operation, a wireless terminal in teleoperation, a wireless terminal in smart grid, a wireless terminal in transportation security, a wireless terminal in smart city, a wireless terminal in smart home, or the like. The embodiment of the application does not limit the specific technology and the specific equipment form adopted by the terminal equipment.

In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, produces a flow or function in accordance with embodiments of the present application, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by a wired (e.g., coaxial cable, fiber optic, digital subscriber line), or wireless (e.g., infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains an integration of one or more available media. The usable medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium, or a semiconductor medium, etc.

The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.

In addition, if a directional indication (such as up, down, left, right, front, and rear) is referred to in the embodiment of the present invention, the directional indication is merely used to explain a relative positional relationship between the components, a movement condition, and the like in a specific posture, and if the specific posture is changed, the directional indication is changed accordingly.

In addition, if there is a description of "first", "second", etc. in the embodiments of the present invention, the description of "first", "second", etc. is for descriptive purposes only and is not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In addition, the meaning of "and/or" as it appears throughout includes three parallel schemes, for example "A and/or B", including the A scheme, or the B scheme, or the scheme where A and B are satisfied simultaneously. In addition, in the embodiment of the present invention, "a plurality of" means two or more. In addition, the technical solutions of the embodiments may be combined with each other, but it is necessary to base that the technical solutions can be realized by those skilled in the art, and when the technical solutions are contradictory or cannot be realized, the combination of the technical solutions should be considered to be absent and not within the scope of protection claimed in the present invention.

Claims

1. A training management system based on intelligent recommendation strategy, characterized by comprising a data processing module, a course recommendation module, an examination supervision module, and a system update module, wherein:

The data processing module is used to access the external human resources system and obtain employee information and course information from the human resources system, and perform pre-processing operations on the employee information and course information;

The course recommendation module is used to convert the employee information and course information after basic preprocessing into feature vectors for clustering, and substitute the employee clusters and course clusters obtained after clustering into the deep neural network model for deep learning to obtain the recommended courses for the employee clusters and the employees suitable for the course clusters;

The examination supervision module uses VUE technology to establish an examination system and implements user interaction through custom services based on JSON. It is used to take examinations for related courses after employees have completed the recommended courses. During the operation of the examination system, the examination supervision module is used to log in using face recognition technology. The examination supervision module has built-in anti-screen switching technology, which is used to automatically determine cheating and automatically hand in the examination paper when the examinee switches to other programs;

The system update module is used to update the system in real time according to the changes in the data within the system, monitor the system status and operation status in real time, regularly back up and maintain data, regularly update software and hardware, regularly evaluate and adjust the recommendation algorithm, and establish a user feedback mechanism.

2. The training management system based on intelligent recommendation strategy according to claim 1, characterized in that the information acquisition and preprocessing operations within the data processing module include:

Employee information: 4A account, name, department, job title, years of work experience, learning progress, learning time, and test results;

Course information: course name, course duration, course score;

Preprocessing operation: Perform data cleaning operations to delete information about employees who have left and inappropriate courses.

3. The training management system based on intelligent recommendation strategy according to claim 1, characterized in that the specific steps of processing employee information and course information within the course recommendation module include:

S1. Convert employee information into feature vectors to obtain employee information feature vectors, use cosine similarity to calculate the similarity of employee information feature vectors, cluster the employee information feature vector similarities using the DBSCAN clustering algorithm to obtain employee clusters, wherein the specific calculation formula for cosine similarity calculation is:

Among them, x represents the employee information feature vector of the first employee, y represents the employee information feature vector of the second employee, and θ represents the angle with the coordinate axis;

S2. Convert the course information into a feature vector to obtain a course information feature vector, use cosine similarity to calculate the similarity of the course information feature vector, cluster the course information feature vector similarity using the DBSCAN clustering algorithm to obtain a course cluster, wherein the specific calculation formula for cosine similarity calculation is:

Where w represents the employee information feature vector of the first employee, z represents the employee information feature vector of the second employee, and μ represents the angle with the coordinate axis;

S3. construct a deep convolutional neural network model, input the employee cluster and course information feature vector into the deep neural network model, and obtain recommended courses for the employee cluster;

S4. Input the course cluster and employee information feature vector into a deep neural network model to obtain an employee suitable for the course cluster, and recommend the course cluster to the employee.

4. The training management system based on intelligent recommendation strategy according to claim 3, characterized in that the specific execution steps of clustering by the DBSCAN clustering algorithm include:

L1. Establish a data set for initialization, select data from the data set, calculate the three-dimensional features of user information, and preset a threshold a of the number of users who meet the user conditions;

L2. Start the loop and preset the clustering parameters e and Minpts for this loop, where e represents the radius of the clustering domain and Minpts represents the minimum number of domain points for a given point to become a core object in the clustering domain;

L3. Randomly extract a user from the data. If the user has not been visited, calculate whether there are at least Minpts data points in the neighborhood of the user. If there are more than Minpts data points, create a new cluster to put the user in. Repeat the calculation process for other data points in the e-neighborhood of the user to expand the cluster. When it is detected that the neighborhood without data points can contain at least Minpts data points, the expansion of the cluster ends.

L4. Repeat step L3 until all users in the data set have been visited, select the cluster that best matches the user characteristics from the obtained clusters, set the number of clustering generations to K, and perform clustering operations for generations;

L5. When K>a, it means that the selected users have met the needs, and the iteration ends.

5. The training management system based on intelligent recommendation strategy according to claim 4, characterized in that the specific operation steps of selecting and performing clustering operation in the L4 step include:

In the first clustering, the abnormal user area and the normal user area are selected. At this time, the normal user area will be marked as low-risk area 1, indicating that it is the cluster that best meets the normal user behavior characteristics. In the subsequent clustering process, the abnormal user cluster will no longer be selected, but only the cluster that meets the normal user behavior characteristics will be selected;

In the process of narrowing the data set, the cluster that meets the malicious user behavior is defined as the medium-risk user area. At this time, the medium-risk user area is not marked. The clusters marked as normal users in this clustering and subsequent clustering will be divided into low-risk area 2, low-risk area 3, and finally low-risk area k. The numerical values of the three-dimensional indicators of users in the above areas are different from those of users in low-risk area 1, and as users of normally running virtual machines, they have the probability of entering low-risk area 1 in the end.

6. The training management system based on intelligent recommendation strategy according to claim 3, characterized in that the specific operation steps of constructing the deep convolutional neural network model in the step S3 include: sequentially constructing an embedding layer, a multi-head attention layer, a convolution layer, a pooling layer and an output layer of the convolutional neural network, wherein:

P1. Embedding layer, which is used to convert the employee information feature vector into a digital matrix according to the length of the feature vector, and then output it to the next layer;

P2. The multi-head attention layer is used to select important features and obtain potential factors from the employee information feature cluster by introducing a multi-head attention mechanism. The multi-head attention mechanism specifically includes: first, the vector information undergoes a linear transformation, and then it is input into the scaled dot product attention, a total of h times, each time is counted as a head, and the parameters between the heads are not shared. Then, the h times of scaled dot product attention results are spliced, and finally a linear transformation is performed to obtain the value as the result of the multi-head attention, that is:

head _i =Attention(QWQ _i ,KWK _i ,VWV _i )

Where: QWQ _i represents the result of a linear transformation of the query vector _Qi , KWK _i represents the result of a linear transformation of the key vector _Ki , and VWV _i represents the result of a linear transformation of the value vector _Vi ;

P3. Convolution layer, used to extract adjacent feature vectors from the feature vector matrix using one-dimensional convolution kernels with window sizes of 1, 3, 5, and 7, thereby extracting context features;

P4. Pooling layer uses k-max-pooling technology to extract the top k values from each pooling block of the convolutional layer as representative features, and processes documents of variable length by constructing a merging operation of fixed-length feature vectors, where the representative features extracted are:

in: Indicates that when the parameters W, X and variance are known is the probability of observing V under the condition of , N(·) is represented by the normal distribution, V _j is the jth observation value, MACNN _W (X _j ) refers to the convolutional neural network with parameter W, The covariance matrix is The noise model of the times unit matrix I;

P5. The output layer projects the K-dimensional space of the user’s latent factors for the recommendation task and finally generates the latent vector of the document by using conventional nonlinear projection, thereby transforming the high-level features obtained from the previous layer for specific tasks.

7. The training management system based on intelligent recommendation strategy as claimed in claim 1, characterized in that the examination system inside the examination supervision module uses video technology combining SRS and WebSocket protocol, wherein:

On the SRS server, WebSocket establishes a connection through the HTTP protocol and then upgrades to the WebSocket protocol;

When establishing a connection, the client sends an HTTP request to the server, and the request header contains the Upgrade field, whose value is WebSocket;

After receiving the request, if the server supports the WebSocket protocol, it returns an HTTP response. The response header contains the Upgrade field, whose value is WebSocket. The response header also contains the Sec-WebSocket-Accept field, whose value is obtained by encrypting the Sec.WebSocket-Key field sent by the client and a fixed GUID string.

After the client receives the response, if the response header contains the Upgrade field, the value is WebSocket, and the value of the Sec-WebSocket-Accept field is the same as the encrypted result of the Sec-WebSocket-Key field sent by the client, it means that the connection has been successfully established and WebSocket communication can begin.

8. The training management system based on intelligent recommendation strategy according to claim 1, characterized in that the anti-screen cutting technology inside the test supervision module is used for anti-screen cutting detection in the web program, specifically including:

The page currently being used by the user is defined as a focus page, where the focus page covers the common page;

Check whether the focus page status is maintained during the exam. If the focus page status is lost during the process, it is regarded as the candidate has performed a screen switching operation, triggering the screen switching cheating rules.

9. A computer-readable storage medium, characterized in that a computer program is stored therein, and when the computer program is executed by a processor, the processor executes the internal algorithm steps of the system according to any one of claims 1 to 8.

10. A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the internal algorithm steps of the system according to any one of claims 1 to 8.