Intelligent campus certificate credit data management system based on artificial intelligence

Abstract

The invention discloses an intelligent campus certificate credit data management system based on artificial intelligence, which relates to the technical field of artificial intelligence and comprises a multi-source data acquisition unit, a certificate verification unit, a credit progress monitoring unit and a credit supervision unit. The method comprises the following steps: acquiring a personal certificate data set, verifying the authenticity of an electronic seal and anti-counterfeiting watermark characteristics, marking the verified personal certificate data set as a target data set, extracting an issuing mechanism in a certificate, acquiring time and skill levels, carrying out assignment representation, and integrating into a certificate value data set; the credit conversion coefficient is calculated according to the authority, skill scarcity and industry acceptance degree of a certificate issuing mechanism, the certificate, the course score and the practice record of the same student are associated, and the student credit acquisition progress is predicted through time sequence analysis. And automatically generating a personalized repair scheme and sending the personalized repair scheme to the student terminal and the monitoring party associated with the student terminal.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a smart campus certificate credit data management system based on artificial intelligence. Background Technology

[0002] Based on the intelligent development of smart campuses, a credit bank system is introduced to manage students' certificates and credits. The credit bank simulates and draws on the mechanisms, functions, and characteristics of a bank, using credits as the unit of measurement. Through the classification, grading, and description of learning outcomes, and corresponding standard systems, it certifies and assigns values ​​to various learning achievements obtained by learners, thus realizing an educational management system and learning incentive system for the accumulation and transformation of learning outcomes. It utilizes information technology to expand the coverage of high-quality educational resources and is an effective way to build a "flyover" for lifelong learning and promote educational equity.

[0003] Credits earned by learners (students or continuing education personnel, etc.) through formal educational institutions can be directly deposited into the credit bank; various training certificates, skill certificates, etc., can also be converted into credits and deposited into the credit bank. However, the existing credit bank only plays a management and storage role. In the course of students' learning progress, there are situations where credits cannot meet the learning progress due to personal reasons. For such students, it is necessary to improve the supervision and management of them. To address the aforementioned technical deficiencies, a solution is proposed. Summary of the Invention

[0004] The purpose of this invention is to predict students' credit acquisition progress through time series analysis, and when graduation requirements are not expected to be met, automatically generate personalized remedial courses and send the personalized remedial courses to the student and the guardian associated with the student, thereby realizing the management of students' learning progress and ensuring that students' credits can meet graduation requirements.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: an artificial intelligence-based smart campus certificate and credit data management system, comprising a multi-source data acquisition unit, a certificate verification unit, a credit progress monitoring unit, and a credit supervision unit; wherein: The multi-source data acquisition unit is used to collect structured certificate data and unstructured proof materials in real time from the campus academic affairs system, third-party certification platform and blockchain evidence storage node, and integrate them into a personal certificate dataset and send it to the certificate verification unit. The certificate verification unit includes a certificate parsing module and a credit calculation module. The certificate parsing module is used to build an image recognition model based on deep learning, to obtain personal certificate datasets and verify the authenticity of electronic seals and anti-counterfeiting watermark features, to mark the verified personal certificate datasets as target datasets, to extract the issuing authority, acquisition time and skill level from the certificates, and to integrate them into a certificate value dataset after assigning values ​​and sending it to the credit calculation module. The credit calculation module is used to obtain the certificate value dataset and calculate the credit conversion factor based on the authority of the certificate issuing institution, the scarcity of skills, and industry recognition, and then send it to the credit progress monitoring unit. The credit progress monitoring unit is used to obtain student information corresponding to the individual certificate dataset and to associate the certificates, course grades and practice records of the same student. The student information includes the student's name, associated guardian and semester progress data. The unit predicts the student's credit acquisition progress through time series analysis. When it is expected that the graduation requirements cannot be met, a personalized remedial plan is automatically generated and pushed to the credit supervision unit. The credit supervision unit is used to obtain personalized remedial plans and generate progress supervision cycle signals based on the plans. The personalized remedial plans are sent to the student's end and the guardian associated with the student's end. The guardian includes the tutor's end and the parent's end. The unit issues reminders to the guardian based on the progress supervision cycle signals and obtains supervision feedback from the guardian.

[0006] Furthermore, it also includes a blockchain evidence storage subsystem, which adopts a consortium blockchain architecture to store authentication results, accounting processes, and operation logs on the blockchain and supports the output of verifiable credential formats in the W3C standard.

[0007] Furthermore, the certificate verification unit also includes a conflict verification module. The conflict verification module is based on a conflict detection algorithm. When multiple certificates for the same skill are detected to overlap, a manual review process is triggered. The overlapping certificates are filtered through the manual review process to optimize the target dataset.

[0008] Furthermore, the specific method for obtaining the certificate value dataset is as follows: S101. Obtain the personal certificate dataset, scan them one by one, delineate the areas where the electronic seals and anti-counterfeiting watermarks are located, mark them as anti-counterfeiting verification areas, and establish an image recognition model based on deep learning. S102. Extract the image corresponding to the anti-counterfeiting verification area and input it into the image recognition model. Output the orientation verification result to obtain the target dataset. S103. Extract the certificate data from the target dataset, as follows: For PDF certificates, scan the certificate text in the target dataset one by one, use PyPDF2 to extract the text content, and retain the position information corresponding to the text content; For image certificates, use TesseractOCR for text recognition; For electronic certificates, directly parse the structured fields; S104. Define a keyword library of common issuing authorities, and use regular expressions to match certificate data with relevant keywords in the keyword library to obtain the issuing authority in the certificate. S105. Based on date regular expressions, match common date formats. If the date is scattered in multiple places, combine with keyword positioning to obtain the certificate acquisition time. S106. Define a level dictionary and obtain the skill level of the certificate by matching through similarity calculation; S107. Assign the value Ah to the issuing organization in the certificate according to its importance level in the industry from high to low; assign the value Bh to the acquisition time in the certificate according to the validity period of the certificate acquisition time from long to short; and assign the value Ch to the skill level in the certificate according to the importance level of the skill level in the certificate from high to low.

[0009] Furthermore, the specific method for calculating the credit conversion factor is as follows: S201. Verify whether the institution is registered through the official website of the Ministry of Human Resources and Social Security, giving priority to institutions that are recognized internationally. At the same time, obtain the institution's operating time and the number of times it has been reported by authoritative media, conduct a comprehensive evaluation to obtain the institution's authority assessment result, and assign a value of Dh based on the institution's authority assessment result. S202. Skill scarcity is assessed from two dimensions: exam difficulty and market supply and demand, resulting in the assigned value data Eh for skill scarcity. S203. The industry recognition is assessed by a dual-drive approach of enterprise needs and regional orientation, resulting in the assigned value data Fh for industry recognition. S204. After obtaining the certificate value dataset, calculate the credit conversion factor Ph according to the following formula: Where e1, e2, and e3, as well as α, β, and γ, are preset proportional coefficients. The credit conversion coefficient is used to evaluate the credit value corresponding to the personal certificate dataset. The higher the credit conversion coefficient, the higher the credit value corresponding to the personal certificate dataset, and vice versa.

[0010] Furthermore, the specific method for automatically generating personalized repair plans is as follows: S301. Obtain the student's name, associated guardian, and semester progress data, wherein the semester progress data includes the student ID, semester, earned credits, total credit requirement, course type, and grades; S302. Establish a timeline based on the start time of the semester, and divide the progress verification nodes based on the time granularity. Link the certificate, course grades and practice records of the same student at the progress verification node. S303. Use STL decomposition to split the time axis into trend terms, which are the long-term direction of credit growth over time. S303. Establish a credit trend evaluation model based on a linear trend model, as shown below: , where t is the time corresponding to the progress verification node, λ0 is a preset constant, λ1 is a preset optimization coefficient, εt is the error term, and Pt is the standard credit conversion coefficient of the current progress verification node; S304. Obtain the credit conversion factor. If the credit conversion factor is less than the standard credit conversion factor, the student is not expected to meet the graduation requirements. S305. Calculate the credit gap as follows: Credit gap = Total credits required for graduation - Credits already earned - Credits of courses currently being taken; And sort the courses according to their priority: Uncompleted required courses will be added directly to the make-up course list; Based on the remaining credit requirements, high-credit and low-difficulty courses are selected from the elective course pool and integrated to obtain a remedial course plan. S306. Convert the repair plan into a CSP model and define the variables, domains, and constraints: The variable is specifically the list of courses taken each semester; The domain is specifically a pool of optional courses, which are courses that meet the prerequisites but have not been taken. The specific constraints are as follows: Credit limit: Total credits per semester ≤ maximum student workload; Time conflict: Course times do not overlap; Prerequisites: If you choose a high-priority course, you must have already completed the corresponding foundation courses; Graduation requirements include: meeting the credit requirements for all required and elective courses; S307. Calculate the overall score for each feasible solution according to the following formula: ,in , and The preset weighting coefficients are used, and the Score is the comprehensive score of the feasible solutions. The feasible solution with the highest comprehensive score is selected as the personalized repair solution.

[0011] Furthermore, the specific method for generating the progress monitoring cycle signal is as follows: S401. Obtain a personalized repair plan, which includes the plan content and the plan start time. Establish a plan distribution timeline based on the plan start time, parse the plan content, and divide the plan execution time according to the plan content. S402. Based on the execution time of the plan, define the monitoring and reminder nodes on the time axis of the plan distribution, obtain the density of the monitoring and reminder nodes, obtain the preset balanced density threshold, and if the density of the monitoring and reminder nodes is greater than or equal to the balanced density threshold, generate a progress monitoring cycle signal at the monitoring and reminder node.

[0012] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: This AI-based smart campus certificate credit data management system acquires individual certificate datasets, verifies the authenticity of electronic seals and anti-counterfeiting watermark features, marks verified individual certificate datasets as target datasets, extracts issuing institutions, acquisition dates, and skill levels from certificates, assigns values ​​to them, and integrates them into a certificate value dataset. It calculates credit conversion coefficients based on the authority of the issuing institution, the scarcity of the skill, and industry recognition, links the same student's certificates, course grades, and practical records, and predicts student credit acquisition progress through time series analysis. When graduation requirements are not expected to be met, it automatically generates personalized remedial plans and sends these plans to the student and their associated guardian, thus achieving management of students' learning progress and ensuring that students' credits meet graduation requirements. Attached Figure Description

[0013] Figure 1 A schematic diagram of the overall structure of the present invention is shown. Detailed Implementation

[0014] 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 embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Example:

[0015] like Figure 1 As shown, the AI-based smart campus certificate and credit data management system includes a multi-source data acquisition unit, a certificate verification unit, a credit progress monitoring unit, and a credit supervision unit; among which: The multi-source data acquisition unit is used to collect structured certificate data and unstructured proof materials in real time from the campus academic affairs system, third-party certification platform and blockchain evidence storage node, and integrate them into a personal certificate dataset and send it to the certificate verification unit. The certificate verification unit includes a certificate parsing module and a credit calculation module. The certificate parsing module is used to build an image recognition model based on deep learning, to obtain personal certificate datasets and verify the authenticity of electronic seals and anti-counterfeiting watermark features. The verified personal certificate datasets are marked as target datasets, and the issuing authority, acquisition time and skill level in the certificates are extracted, and after being assigned values, they are integrated into a certificate value dataset and sent to the credit calculation module. The specific method for obtaining the certificate value dataset is as follows: S101. Obtain the personal certificate dataset, scan them one by one, delineate the areas where the electronic seals and anti-counterfeiting watermarks are located, mark them as anti-counterfeiting verification areas, and establish an image recognition model based on deep learning. S102. Extract the image corresponding to the anti-counterfeiting verification area and input it into the image recognition model. Output the orientation verification result to obtain the target dataset. S103. Extract the certificate data from the target dataset, as follows: For PDF certificates, scan the certificate text in the target dataset one by one, use PyPDF2 to extract the text content, and retain the position information corresponding to the text content; For image certificates, use TesseractOCR for text recognition; For electronic certificates, directly parse the structured fields; S104. Define a keyword library of common issuing authorities, and use regular expressions to match certificate data with relevant keywords in the keyword library to obtain the issuing authority in the certificate. S105. Based on date regular expressions, match common date formats. If the date is scattered in multiple places, combine with keyword positioning to obtain the certificate acquisition time. S106. Define a level dictionary and obtain the skill level of the certificate by matching through similarity calculation; S107. Assign the value Ah to the issuing organization in the certificate according to its importance level in the industry from high to low; assign the value Bh to the acquisition time in the certificate according to the validity period of the certificate acquisition time from long to short; and assign the value Ch to the skill level in the certificate according to the importance level of the skill level in the certificate from high to low.

[0016] The credit calculation module is used to obtain the certificate value dataset and calculate the credit conversion factor based on the authority of the certificate issuing institution, the scarcity of skills, and industry recognition, and then send it to the credit progress monitoring unit. The specific method for calculating the credit conversion factor is as follows: S201. Verify whether the institution is registered through the official website of the Ministry of Human Resources and Social Security, giving priority to institutions that are recognized internationally. At the same time, obtain the institution's operating time and the number of times it has been reported by authoritative media, conduct a comprehensive evaluation to obtain the institution's authority assessment result, and assign a value of Dh based on the institution's authority assessment result. S202. Skill scarcity is assessed from two dimensions: exam difficulty and market supply and demand, resulting in the assigned value data Eh for skill scarcity. S203. The industry recognition is assessed by a dual-drive approach of enterprise needs and regional orientation, resulting in the assigned value data Fh for industry recognition. S204. After obtaining the certificate value dataset, calculate the credit conversion factor Ph according to the following formula: Where e1, e2, and e3, as well as α, β, and γ, are preset proportional coefficients. The credit conversion coefficient is used to evaluate the credit value corresponding to the personal certificate dataset. The higher the credit conversion coefficient, the higher the credit value corresponding to the personal certificate dataset, and vice versa.

[0017] The credit progress monitoring unit is used to obtain student information corresponding to the individual certificate dataset and to associate the certificates, course grades and practice records of the same student. The student information includes the student's name, associated guardian and semester progress data. The unit predicts the student's credit acquisition progress through time series analysis. When the graduation requirements are not expected to be met, a personalized remedial plan is automatically generated and pushed to the credit supervision unit. The specific method for automatically generating personalized repair plans is as follows: S301. Obtain student name, associated guardian, and semester progress data. Semester progress data includes student ID, semester, earned credits, total credit requirement, course type, and grades. S302. Establish a timeline based on the start time of the semester, and divide the progress verification nodes based on the time granularity. Link the certificate, course grades and practice records of the same student at the progress verification node. S303. Use STL decomposition to split the time axis into trend terms, which are the long-term direction of credit growth over time. S303. Establish a credit trend evaluation model based on a linear trend model, as shown below: , where t is the time corresponding to the progress verification node, λ0 is a preset constant, λ1 is a preset optimization coefficient, εt is the error term, and Pt is the standard credit conversion coefficient of the current progress verification node; S304. Obtain the credit conversion factor. If the credit conversion factor is less than the standard credit conversion factor, the student is not expected to meet the graduation requirements. S305. Calculate the credit gap as follows: Credit gap = Total credits required for graduation - Credits already earned - Credits of courses currently being taken; And sort the courses according to their priority: Uncompleted required courses will be added directly to the make-up course list; Based on the remaining credit requirements, high-credit and low-difficulty courses are selected from the elective course pool and integrated to obtain a remedial course plan. S306. Convert the repair plan into a CSP model and define the variables, domains, and constraints: The variable is specifically the list of courses taken each semester; The domain is specifically a pool of optional courses, which are courses that meet the prerequisites but have not been taken. The specific constraints are as follows: Credit limit: Total credits per semester ≤ maximum student workload; Time conflict: Course times do not overlap; Prerequisites: If you choose a high-priority course, you must have already completed the corresponding foundation courses; Graduation requirements include: meeting the credit requirements for all required and elective courses; S307. Calculate the overall score for each feasible solution according to the following formula: ,in , and The preset weighting coefficients are used, and the Score is the comprehensive score of the feasible solutions. The feasible solution with the highest comprehensive score is selected as the personalized repair solution.

[0018] The credit supervision unit is used to obtain personalized remedial plans and generate progress supervision cycle signals based on the plans. The personalized remedial plans are sent to the student's end and the guardians associated with the student's end. The guardians include the tutor's end and the parent's end. The unit issues reminders to the guardians based on the progress supervision cycle signals and obtains supervision feedback from the guardians.

[0019] The specific method for generating progress monitoring cycle signals is as follows: S401. Obtain a personalized repair plan. The personalized repair plan includes the plan content and the plan start time. Establish a plan distribution timeline based on the plan start time, and parse the plan content. Divide the plan execution time into segments according to the plan content. S402. Based on the execution time of the plan, define the monitoring and reminder nodes on the time axis of the plan distribution, obtain the density of the monitoring and reminder nodes, obtain the preset balanced density threshold, and if the density of the monitoring and reminder nodes is greater than or equal to the balanced density threshold, generate a progress monitoring cycle signal at the monitoring and reminder node.

[0020] It also includes a blockchain evidence storage subsystem, which adopts a consortium blockchain architecture to store authentication results, accounting processes and operation logs on the blockchain, and supports the output of verifiable credential format in the W3C standard.

[0021] The certificate verification unit also includes a conflict verification module. The conflict verification module is based on a conflict detection algorithm. When multiple overlapping certificates for the same skill are detected, a manual review process is triggered. The overlapping certificates are filtered out through the manual review process to optimize the target dataset.

[0022] This invention acquires a personal certificate dataset, verifies the authenticity of electronic seals and anti-counterfeiting watermark features, marks the verified personal certificate dataset as the target dataset, extracts the issuing institution, acquisition time and skill level from the certificates, assigns values ​​to them and integrates them into a certificate value dataset, calculates credit conversion coefficients based on the authority of the issuing institution, the scarcity of skills and industry recognition, associates the certificates, course grades and practical records of the same student, predicts the student's credit acquisition progress through time series analysis, and automatically generates a personalized remedial plan when graduation requirements are not expected to be met, and sends the personalized remedial plan to the student and the guardian associated with the student, thereby realizing the management of the student's learning progress and ensuring that the student's credits can meet the graduation requirements.

[0023] The size of the interval and threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by those skilled in the art for each set of sample data; as long as it does not affect the ratio between the parameter and the quantized value.

[0024] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation. The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A smart campus certificate and credit data management system based on artificial intelligence, characterized in that: It includes a multi-source data acquisition unit, a certificate verification unit, a credit progress monitoring unit, and a credit supervision unit; among which: The multi-source data acquisition unit is used to collect structured certificate data and unstructured proof materials in real time from the campus academic affairs system, third-party certification platform and blockchain evidence storage node, and integrate them into a personal certificate dataset and send it to the certificate verification unit. The certificate verification unit includes a certificate parsing module and a credit calculation module. The certificate parsing module is used to build an image recognition model based on deep learning, to obtain personal certificate datasets and verify the authenticity of electronic seals and anti-counterfeiting watermark features, to mark the verified personal certificate datasets as target datasets, to extract the issuing authority, acquisition time and skill level from the certificates, and to integrate them into a certificate value dataset after assigning values ​​and sending it to the credit calculation module. The credit calculation module is used to obtain the certificate value dataset and calculate the credit conversion factor based on the authority of the certificate issuing institution, the scarcity of skills, and industry recognition, and then send it to the credit progress monitoring unit. The credit progress monitoring unit is used to obtain student information corresponding to the individual certificate dataset and to associate the certificates, course grades and practice records of the same student. The student information includes the student's name, associated guardian and semester progress data. The unit predicts the student's credit acquisition progress through time series analysis. When it is expected that the graduation requirements cannot be met, a personalized remedial plan is automatically generated and pushed to the credit supervision unit. The credit supervision unit is used to obtain personalized remedial plans and generate progress supervision cycle signals based on the plans. The personalized remedial plans are sent to the student's end and the guardian associated with the student's end. The guardian includes the tutor's end and the parent's end. The unit issues reminders to the guardian based on the progress supervision cycle signals and obtains supervision feedback from the guardian.

2. The AI-based smart campus certificate and credit data management system according to claim 1, characterized in that, It also includes a blockchain evidence storage subsystem, which adopts a consortium blockchain architecture to store authentication results, accounting processes and operation logs on the blockchain, and supports the output of verifiable credential format in the W3C standard.

3. The AI-based smart campus certificate and credit data management system according to claim 1, characterized in that, The certificate verification unit also includes a conflict verification module. The conflict verification module is based on a conflict detection algorithm. When multiple certificates for the same skill are detected to overlap, a manual review process is triggered. The overlapping certificates are filtered out through the manual review process to optimize the target dataset.

4. The AI-based smart campus certificate and credit data management system according to claim 1, characterized in that, The specific method for obtaining the certificate value dataset is as follows: S101. Obtain the personal certificate dataset, scan them one by one, delineate the areas where the electronic seals and anti-counterfeiting watermarks are located, mark them as anti-counterfeiting verification areas, and establish an image recognition model based on deep learning. S102. Extract the image corresponding to the anti-counterfeiting verification area and input it into the image recognition model. Output the orientation verification result to obtain the target dataset. S103. Extract the certificate data from the target dataset, as follows: For PDF certificates, scan the certificate text in the target dataset one by one, use PyPDF2 to extract the text content, and retain the position information corresponding to the text content; For image certificates, use TesseractOCR for text recognition; For electronic certificates, directly parse the structured fields; S104. Define a keyword library of common issuing authorities, and use regular expressions to match certificate data with relevant keywords in the keyword library to obtain the issuing authority in the certificate. S105. Based on date regular expressions, match common date formats. If the date is scattered in multiple places, combine with keyword positioning to obtain the certificate acquisition time. S106. Define a level dictionary and obtain the skill level of the certificate by matching through similarity calculation; S107. Assign the value Ah to the issuing organization in the certificate according to its importance level in the industry from high to low; assign the value Bh to the acquisition time in the certificate according to the validity period of the certificate acquisition time from long to short; and assign the value Ch to the skill level in the certificate according to the importance level of the skill level in the certificate from high to low.

5. The AI-based smart campus certificate and credit data management system according to claim 1, characterized in that, The specific method for calculating the credit conversion factor is as follows: S201. Verify whether the institution is registered through the official website of the Ministry of Human Resources and Social Security, giving priority to institutions that are recognized internationally. At the same time, obtain the institution's operating time and the number of times it has been reported by authoritative media, conduct a comprehensive evaluation to obtain the institution's authority assessment result, and assign a value of Dh based on the institution's authority assessment result. S202. Skill scarcity is assessed from two dimensions: exam difficulty and market supply and demand, resulting in the assigned value data Eh for skill scarcity. S203. The industry recognition is assessed by a dual-drive approach of enterprise needs and regional orientation, resulting in the assigned value data Fh for industry recognition. S204. After obtaining the certificate value dataset, calculate the credit conversion factor Ph according to the following formula: Where e1, e2, and e3, as well as α, β, and γ, are preset proportional coefficients. The credit conversion coefficient is used to evaluate the credit value corresponding to the personal certificate dataset. The higher the credit conversion coefficient, the higher the credit value corresponding to the personal certificate dataset, and vice versa.

6. The AI-based smart campus certificate and credit data management system according to claim 1, characterized in that, The specific method for automatically generating personalized repair plans is as follows: S301. Obtain the student's name, associated guardian, and semester progress data, wherein the semester progress data includes the student ID, semester, earned credits, total credit requirement, course type, and grades; S302. Establish a timeline based on the start time of the semester, and divide the progress verification nodes based on the time granularity. Link the certificate, course grades and practice records of the same student at the progress verification node. S303. Use STL decomposition to split the time axis into trend terms, which are the long-term direction of credit growth over time. S303. Establish a credit trend evaluation model based on a linear trend model, as shown below: , where t is the time corresponding to the progress verification node, λ0 is a preset constant, λ1 is a preset optimization coefficient, εt is the error term, and Pt is the standard credit conversion coefficient of the current progress verification node; S304. Obtain the credit conversion factor. If the credit conversion factor is less than the standard credit conversion factor, the student is not expected to meet the graduation requirements. S305. Calculate the credit gap as follows: Credit gap = Total credits required for graduation - Credits already earned - Credits of courses currently being taken; And sort the courses according to their priority: Uncompleted required courses will be added directly to the make-up course list; Based on the remaining credit requirements, high-credit and low-difficulty courses are selected from the elective course pool and integrated to obtain a remedial course plan. S306. Convert the repair plan into a CSP model and define the variables, domains, and constraints: The variable is specifically the list of courses taken each semester; The domain is specifically a pool of optional courses, which are courses that meet the prerequisites but have not been taken. The specific constraints are as follows: Credit limit: Total credits per semester ≤ maximum student workload; Time conflict: Course times do not overlap; Prerequisites: If you choose a high-priority course, you must have already completed the corresponding foundation courses; Graduation requirements include: meeting the credit requirements for all required and elective courses; S307. Calculate the overall score for each feasible solution according to the following formula: ,in , and The preset weighting coefficients are used, and the Score is the comprehensive score of the feasible solutions. The feasible solution with the highest comprehensive score is selected as the personalized repair solution.

7. The AI-based smart campus certificate and credit data management system according to claim 1, characterized in that, The specific method for generating progress monitoring cycle signals is as follows: S401. Obtain a personalized repair plan, which includes the plan content and the plan start time. Establish a plan distribution timeline based on the plan start time, parse the plan content, and divide the plan execution time according to the plan content. S402. Based on the execution time of the plan, define the monitoring and reminder nodes on the time axis of the plan distribution, obtain the density of the monitoring and reminder nodes, obtain the preset balanced density threshold, and if the density of the monitoring and reminder nodes is greater than or equal to the balanced density threshold, generate a progress monitoring cycle signal at the monitoring and reminder node.

Images