Artificial intelligence paper marking system based on edge calculation

Abstract

The invention discloses an artificial intelligence paper marking system based on edge calculation, and relates to the technical field of machine paper marking, the artificial intelligence paper marking system comprises a cloud processing terminal, the cloud processing terminal is in communication connection with a data acquisition module, an edge calculation module, a data analysis module and a paper marking module; the data acquisition module is used for setting acquisition equipment to carry out pre-acquisition simulation, if the pre-acquisition simulation is passed, formal acquisition of the paper marking related data is continued, and otherwise, the acquisition equipment is debugged; carrying out load analysis on paper marking related data through an edge calculation module so as to judge whether a current edge server can carry out calculation processing or not, and executing different calculation operations according to a judgment result; obtaining a paper marking data set generated by calculation operation through a data analysis module, and carrying out labeling analysis on the paper marking data set through an artificial intelligence algorithm so as to establish a corresponding paper marking model; and the paper marking module carries out paper marking operation through the paper marking model so as to generate a paper marking result.

Description

Technical Field

[0001] The present invention relates to the technical field of machine marking, and in particular to an artificial intelligence marking system based on edge computing. Background Art

[0002] Traditional test marking methods usually rely on manual labor, which is time-consuming and labor-intensive, and is prone to inaccurate scoring or high subjectivity. With the development of artificial intelligence technology, it has become possible to achieve automated marking using technologies such as machine learning and deep learning.

[0003] However, there are the following problems in automated marking: when the acquisition equipment corresponding to automated marking collects the corresponding target test paper, there may be problems such as unclear target test paper and malfunction of the acquisition equipment, which will lead to inaccurate collected data. The collected data cannot be processed in a timely and effective manner, and it is not possible to establish a marking model with a high degree of fit to generate marking results. How to solve these problems is what we urgently need to consider at present. Summary of the invention

[0004] In order to solve the above problems, the purpose of the present invention is to provide an artificial intelligence marking system based on edge computing.

[0005] The object of the present invention can be achieved by the following technical solutions: an artificial intelligence marking system based on edge computing, comprising a cloud processing terminal, wherein the cloud processing terminal is communicatively connected with a data acquisition module, an edge computing module, a data analysis module and a marking module; The data collection module sets several types of collection devices and performs pre-collection simulation on the collection devices. If the pre-collection simulation is passed, the formal collection of the examination paper related data will continue. Otherwise, the collection devices will be debugged. The edge computing module is used to perform load analysis on the examination paper related data, and to determine whether the current edge server can perform computing processing through load analysis, and to perform different computing operations according to the judgment result; The data analysis module obtains the marking data set generated by the calculation operation, annotates and analyzes the marking data set through the artificial intelligence algorithm, and then establishes a corresponding marking model; The marking module performs marking operations through the marking model to generate marking results, and sends the marking results to a set user platform, and students and teachers log in to the user platform to perform relevant operations on the marking results.

[0006] Furthermore, the process of setting up several types of acquisition equipment and performing pre-acquisition simulation includes: Several types of acquisition equipment are provided, including scanners, optical readers and fill-in-light equipment. The scanner scans the target test paper, and then converts the target test paper from paper data to electronic file data. The optical reader automatically reads and recognizes the words and numbers on the target test paper through optical character recognition technology, and converts them into electronic file data. The fill-in-light equipment is used to perform corresponding fill-in-light operations according to the lighting environment of the target test paper. Set the initial value of the fill light parameter of the fill light device, and then perform pre-collection simulation of the scanner and the optical reader respectively. The electronic file data obtained by the scanner is marked as data one, and the electronic file data obtained by the optical reader is marked as data two. The file clarity ratio of data one and data two is obtained and recorded as R1 and R2 respectively. The clarity thresholds of the target test papers corresponding to data one and data two are preset and recorded as α and β respectively.

[0007] Furthermore, the process of formally collecting the data related to the examination paper includes: If R1≥α and R2≥β, the pre-collection simulation is judged to be passed, and the collection device that passes the pre-collection simulation is marked as a qualified device. The collection device sets the data collection period and data collection rate, which are recorded as T 采 and V 采 , data acquisition rate V 采 Set a warning value range, denoted as Ω, during the data collection period T 采 , the collection device conducts formal collection to obtain the relevant data of the examination paper, and enters the relevant data of the examination paper into the preset blank text, and then converts the blank text into the data analysis text, and sends the data analysis text to the edge computing module; When V 采 ∉Ω, indicating that the current acquisition device is abnormal, and a maintenance warning is generated and sent to the cloud processing terminal. The administrator of the cloud processing terminal arranges relevant maintenance personnel to repair the acquisition device; When V 采 ∈Ω, indicating that the current acquisition device has no abnormality and no maintenance is required.

[0008] Furthermore, the process of debugging the acquisition device includes: If R1≥α and R2≥β are not satisfied, it means that the pre-acquisition simulation has failed. The lighting environment of the target test paper is judged by the fill light device, and then different fill light operations are performed according to the judgment results. The acquisition device is debugged through the fill light operation. The fill light operation includes light enhancement and light reduction. When the lighting environment is judged to be in dark light, light enhancement is performed. When the lighting environment is judged to be strong light, light reduction is performed. Then, through the fill light operation, R1≥α and R2≥β are established, indicating that the pre-acquisition simulation has passed and the debugging of the acquisition device is completed.

[0009] Furthermore, the process of determining whether the current edge server can perform computing processing through load analysis and performing different computing operations according to the determination result includes: The data analysis text is obtained and distributed to the edge server set by the edge computing module. The edge server is set with different server states, including "idle state", "occupied state" and "fault state". When the edge server is in "idle state", the edge server receives the data analysis text and deconstructs the corresponding marking-related data for load analysis. When the edge server is in "occupied state" and "fault state", the edge server corresponding to the current status code does not receive the data analysis text. The computing resources required for the edge server to process the marking data set are obtained, and the value of the computing resource is marked as Cal 需 , obtain the real-time computing resources in the edge server, denoted as Cal 实 , and then according to Cal 需 and Cal 实 Conduct load analysis; The calculation operation includes calculation operation 1 and calculation operation 2. The corresponding calculation operation is performed according to different judgment results. If Cal 需 ≥Cal 实 , then it is determined that the current edge server cannot perform calculations on the grading-related data, and the corresponding calculation operation is performed; if Cal 需 <Cal 实 , it is determined that the current edge server can satisfy the corresponding calculation and processing of the marking data, and the calculation operation 2 is executed accordingly, and the marking data set is generated through the calculation operation.

[0010] Furthermore, the process of labeling and analyzing the marking data set through artificial intelligence algorithms and then establishing a corresponding marking model includes: A marking data set is obtained, which includes several test paper images. The several test paper images are numbered and denoted as x, where x=1, 2, 3, ..., k, where k is a natural number greater than 0. A cutting parameter corresponding to the test paper image numbered x is set, and then the test paper image numbered x is cut into several fixed-area to-be-annotated regions through the cutting parameter, denoted as Z(x, y), which represents the yth to-be-annotated region of the test paper image numbered x, where y=1, 2, 3, ..., g, where g is a natural number greater than 0. An annotation analysis of the Z(x, y) to-be-annotated region is performed through an artificial intelligence algorithm, and then a corresponding annotation text is generated, where the annotation text records comparison keywords of the test paper image, a model building program is set, and the annotation texts corresponding to the several to-be-annotated regions are imported into the model building program, and then a regional marking sub-model is generated, and the regional marking sub-models are summarized and spliced, and then a corresponding marking model is established.

[0011] Furthermore, the process of performing the marking operation to generate the marking results and sending them to the user platform includes: Obtain the operation authority of the marking model, set the answer data and score values ​​of several regional marking sub-models corresponding to the marking model through the operation authority, and convert the answer data and score values ​​of the regional marking sub-models into matching text. The matching text is used to record the matching keywords generated by the answer data and the score values, and the comparison keywords and matching keywords are matched one by one through the marking operation to generate the corresponding marking results, and a secure transmission channel is established to send the marking results to the set user platform through the secure transmission channel.

[0012] Furthermore, the process of students and teachers logging into the user platform to perform related operations on the marking results includes: The user platform is associated with different types of verification forms, including student verification forms and teacher verification forms. The student verification form records information about all students allowed to log in to the user platform, and the teacher verification form records information about all teachers allowed to log in to the user platform. Several students and teachers enter their corresponding student information and user information on the user platform, and then the student verification form conducts the student login review, and the teacher verification form conducts the teacher login review. When their respective login reviews are passed, students and teachers are granted relevant operations on the marking results, including browsing, printing, downloading and forwarding operations.

[0013] Compared with the prior art, the present invention has the following beneficial effects: 1. Carry out pre-collection simulation on the acquisition equipment. If it passes the pre-collection simulation, proceed with the formal collection of marking-related data. Otherwise, debug the acquisition equipment, including different fill-light operations of the fill-light equipment. Then, obtain a clear target test paper through the fill-light operation. Determine whether the data acquisition rate of the acquisition equipment is within the set warning value range, and then determine whether the acquisition equipment is abnormal. The abnormal acquisition equipment should be repaired in time, which lays a good foundation for obtaining accurate data.

[0014] 2. By performing load analysis on the marking-related data, we can determine whether the current edge server can perform computing and processing, and perform different computing operations based on the judgment results, thereby realizing timely processing of the marking-related data; obtaining the test paper image corresponding to the marking data set, and cutting the test paper image into several areas to be labeled according to the set cutting parameters, and performing labeling analysis on the areas to be labeled through the artificial intelligence algorithm, and then generating the corresponding annotation text, and the set model building program generates the regional marking sub-model corresponding to each annotation text, and the regional marking sub-models are summarized and spliced ​​to establish the marking model, thereby improving the fitting degree and refinement of the marking model. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a schematic diagram of the present invention. DETAILED DESCRIPTION

[0016] like Figure 1 As shown, an artificial intelligence marking system based on edge computing includes a cloud processing terminal, and the cloud processing terminal is communicatively connected with a data acquisition module, an edge computing module, a data analysis module and a marking module; The data collection module sets several types of collection devices and performs pre-collection simulation on the collection devices. If the pre-collection simulation is passed, the formal collection of the examination paper related data will continue. Otherwise, the collection devices will be debugged. The edge computing module is used to perform load analysis on the examination paper related data, and to determine whether the current edge server can perform computing processing through load analysis, and to perform different computing operations according to the judgment result; The data analysis module obtains the marking data set generated by the calculation operation, annotates and analyzes the marking data set through the artificial intelligence algorithm, and then establishes a corresponding marking model; The marking module performs marking operations through the marking model to generate marking results, and sends the marking results to a set user platform, and students and teachers log in to the user platform to perform relevant operations on the marking results.

[0017] It should be further explained that, in the specific implementation process, several types of acquisition devices are set up, and the process of pre-acquisition simulation of the acquisition devices includes: Setting up several types of acquisition equipment, including scanners, optical readers and fill light equipment; The scanner scans the target test paper, thereby converting the target test paper from paper data to electronic file data. The optical reader automatically reads and recognizes the text, numbers and other information on the target test paper through optical character recognition technology, and converts them into electronic file data. The fill light device is used to perform corresponding fill light operations according to the lighting environment of the target test paper. The initial value of the fill light parameter of the fill light device is set, and then the pre-collection simulation of the scanner and the optical reader is performed respectively, the electronic file data obtained by the scanner is marked as data one, and the electronic file data obtained by the optical reader is marked as data two; Obtain the file clarity ratio of data 1, denoted as R1, R1 = S1 / S1`, where S1 is the area of ​​the clear area corresponding to data 1, S1` is the total area of ​​the target test paper corresponding to data 1, and preset the clarity threshold of the target test paper corresponding to data 1, denoted as α; Obtain the file clarity ratio of data 2, denoted as R2, R2 = S2 / S2`, where S2 is the area of ​​the clear area corresponding to data 2, S2` is the total area of ​​the target test paper corresponding to data 2, and the clarity threshold of the target test paper corresponding to data 2 is preset, denoted as β; Determine whether the pre-collection simulation passes and perform corresponding operations; If R1≥α, and R2≥β, the pre-collection simulation is judged to be passed; Otherwise, the lighting environment of the target test paper is judged by the fill light device, and then different fill light operations are performed according to the judgment results. The acquisition device is debugged through the fill light operation. The fill light operation includes light enhancement and light reduction. When the lighting environment is judged to be in dark light, light enhancement is performed. When the lighting environment is judged to be strong light, light reduction is performed. Then, through the fill light operation, R1≥α and R2≥β are established, indicating that the pre-acquisition simulation passes and the debugging of the acquisition device is completed.

[0018] It should be further explained that, in the specific implementation process, after the pre-collection simulation, the process of formally collecting the examination paper related data includes: The acquisition device that has passed the pre-acquisition simulation is marked as a qualified device. The qualified device generates a remote access instruction before starting the formal acquisition. The remote access instruction is used to obtain all operation permissions corresponding to the qualified device, and the remote access instruction is transmitted to the cloud processing terminal, and then the cloud processing terminal remotely controls the qualified device through the remote access instruction; The acquisition device sets the data acquisition period and data acquisition rate, which are respectively denoted as T 采 and V 采 , where T 采 =[t1, t2], t1 is the start time of the data collection period, t2 is the end time of the data collection period, the data collection rate V 采 Set a warning value range, denoted as Ω, Ω = [min, max], where min is V 采 The minimum allowable value, max is V 采 The maximum permissible value of During the data collection period, the collection device conducts formal collection to obtain the examination-related data, enters the examination-related data into a preset blank text, converts the blank text into a data analysis text, and sends the data analysis text to the edge computing module; When V 采 ∉Ω, indicating that the current acquisition device is abnormal, and a maintenance warning is generated and sent to the cloud processing terminal. The administrator of the cloud processing terminal arranges relevant maintenance personnel to repair the acquisition device; When V 采 ∈Ω, indicating that the current acquisition device has no abnormality and no maintenance is required.

[0019] It should be further explained that, in the specific implementation process, the load analysis of the examination paper related data is performed, and then the load analysis is used to determine whether the current edge server can perform the calculation processing. The process of performing different calculation operations according to the judgment result includes: The edge computing module obtains the data analysis text and distributes the data analysis text to the edge server corresponding to the edge computing module, wherein the edge server is provided with different server states, including "idle state", "occupied state" and "fault state", and different server states have corresponding status codes; The status codes include "00", "01" and "11"; When the status code is "00", it means that the edge server is in "idle state". The edge server receives the data analysis text and deconstructs the corresponding marking-related data for load analysis. When the status code is "01", it means that the edge server is in "occupied state", and the edge server corresponding to the current status code does not receive the data analysis text; When the status code is "11", it means that the edge server is in a "fault state". All operating parameters corresponding to the edge server at this time are exported, and then a maintenance report is generated and sent to the administrator. The administrator performs maintenance on the edge server through remote control and returns the maintenance information to the set fault database. Get the computing resources required by the edge server to process the marking data set, and mark the value of the computing resource as Cal 需 , obtain the real-time computing resources in the edge server, denoted as Cal 实 , and then according to Cal 需 and Cal 实 Conduct load analysis; The calculation operation includes a calculation operation 1 and a calculation operation 2, and corresponding calculation operations are performed according to different judgment results; If Cal 需 ≥Cal 实, it is determined that the current edge server cannot perform calculation processing on the marking-related data, and the corresponding calculation operation 1 is executed; the content of the calculation operation 1 is: uploading the marking-related data to the cloud processing terminal, and the cloud processing terminal schedules its own computing resources, and generates the corresponding marking data set after the required computing resources are met; If Cal 需 <Cal 实 , it is determined that the current edge server can perform corresponding calculations on the marking data, and the calculation operation 2 is performed accordingly; the content of the calculation operation 2 is: the current edge server directly performs calculations and generates a marking data set; The examination paper data set is transmitted to the data analysis module.

[0020] It should be further explained that, in the specific implementation process, the process of obtaining the marking data set generated by the calculation operation, annotating and analyzing it through the artificial intelligence algorithm, and then establishing the corresponding marking model includes: Obtain a marking data set transmitted by the edge computing module, where the marking data set is a number of test paper images, number the test paper images, record the number as x, x=1, 2, 3, ..., k, where k is a natural number greater than 0, set the cutting parameters corresponding to the test paper image numbered x, the cutting parameters include cutting length and cutting width, which are recorded as L[x] and W[x] respectively; Then, the test paper image numbered x is cut into a number of corresponding fixed-area to-be-marked regions, recorded as Z(x, y), by using the cutting parameters, which means: the yth to-be-marked region of the test paper image numbered x, wherein y=1, 2, 3, ..., g, g is a natural number greater than 0, and the Z(x, y) to-be-marked region is subjected to the labeling analysis by the artificial intelligence algorithm, and then the corresponding labeling text is generated, wherein the labeling text records the comparison keywords of the test paper image; Set up a model building program and configure a secure environment for the model building program, and then convert the real-time environment of the model building into a secure environment. In the secure environment, import the annotation texts corresponding to several areas to be annotated into the model building program, and then generate regional marking sub-models, summarize and splice the regional marking sub-models, and then establish the corresponding marking model.

[0021] It should be further explained that, in the specific implementation process, the marking operation is performed to generate the marking results, and the marking results are sent to the user platform. The process in which students and teachers log in to the user platform to perform related operations on the marking results includes: Obtaining the operation authority of the marking model, setting the answer data and score values ​​of several regional marking sub-models corresponding to the marking model through the operation authority, and converting the answer data and score values ​​of the regional marking sub-models into matching texts, wherein the matching texts are used to record matching keywords generated by the answer data and the score values; Perform the marking operation, the content of which is: perform a one-by-one matching comparison of the comparison keywords and the matching keywords, thereby generating corresponding marking results, and establishing a secure transmission channel, and sending the marking results to a set user platform through the secure transmission channel; The user platform is associated with different types of verification forms, including student verification forms and teacher verification forms. The student verification form records information about all students allowed to log in to the user platform, and the teacher verification form records information about all teachers allowed to log in to the user platform. Several students and teachers enter their corresponding student information and user information on the user platform, and then the student verification form is used to check the student's login, and the teacher verification form is used to check the teacher's login; After the login review is passed, students and teachers are granted relevant operations on the marking results, including browsing, printing, downloading and forwarding operations.

[0022] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An artificial intelligence examination system based on edge computing, including a cloud processing terminal, characterized in that: The cloud processing terminal is communicatively connected to a data acquisition module, an edge computing module, a data analysis module and a paper marking module; The data collection module sets several types of collection devices and performs pre-collection simulation on the collection devices. If the pre-collection simulation is passed, the formal collection of the examination paper related data will continue. Otherwise, the collection devices will be debugged. The edge computing module is used to perform load analysis on the examination paper related data, and to determine whether the current edge server can perform computing processing through load analysis, and to perform different computing operations according to the judgment result; The data analysis module obtains the marking data set generated by the calculation operation, annotates and analyzes the marking data set through the artificial intelligence algorithm, and then establishes a corresponding marking model; The marking module performs marking operations through the marking model to generate marking results, and sends the marking results to a set user platform, and students and teachers log in to the user platform to perform relevant operations on the marking results.

2. According to claim 1, an artificial intelligence examination system based on edge computing is characterized in that: The process of setting up several types of acquisition equipment and performing pre-acquisition simulation includes: Several types of acquisition equipment are provided, including scanners, optical readers and fill-in-light equipment. The scanner scans the target test paper, and then converts the target test paper from paper data to electronic file data. The optical reader automatically reads and recognizes the words and numbers on the target test paper through optical character recognition technology, and converts them into electronic file data. The fill-in-light equipment is used to perform corresponding fill-in-light operations according to the lighting environment of the target test paper. Set the initial value of the fill light parameter of the fill light device, and then perform pre-collection simulation of the scanner and the optical reader respectively. The electronic file data obtained by the scanner is marked as data one, and the electronic file data obtained by the optical reader is marked as data two. The file clarity ratio of data one and data two is obtained and recorded as R1 and R2 respectively. The clarity thresholds of the target test papers corresponding to data one and data two are preset and recorded as α and β respectively.

3. The artificial intelligence marking system based on edge computing according to claim 2 is characterized in that: The process of formally collecting the data related to the examination paper includes: If R1≥α and R2≥β, the pre-collection simulation is judged to be passed, and the collection device that passes the pre-collection simulation is marked as a qualified device. The collection device sets the data collection period and data collection rate, which are recorded as T 采 and V 采 , data acquisition rate V 采 Set a warning value range, denoted as Ω, during the data collection period T 采 , the collection device conducts formal collection to obtain the relevant data of the examination paper, and enters the relevant data of the examination paper into the preset blank text, and then converts the blank text into the data analysis text, and sends the data analysis text to the edge computing module; When V 采 ∉Ω, indicating that the current acquisition device is abnormal, and a maintenance warning is generated and sent to the cloud processing terminal. The administrator of the cloud processing terminal arranges relevant maintenance personnel to repair the acquisition device; When V 采 ∈Ω, indicating that the current acquisition device has no abnormality and no maintenance is required.

4. The artificial intelligence examination system based on edge computing according to claim 3 is characterized in that: The process of debugging the acquisition device includes: If R1≥α and R2≥β are not satisfied, it means that the pre-acquisition simulation has failed. The lighting environment of the target test paper is judged by the fill light device, and then different fill light operations are performed according to the judgment results. The acquisition device is debugged through the fill light operation. The fill light operation includes light enhancement and light reduction. When the lighting environment is judged to be in dark light, light enhancement is performed. When the lighting environment is judged to be strong light, light reduction is performed. Then, through the fill light operation, R1≥α and R2≥β are established, indicating that the pre-acquisition simulation has passed and the debugging of the acquisition device is completed.

5. According to claim 4, an artificial intelligence marking system based on edge computing is characterized in that: The process of determining whether the current edge server can perform computing processing through load analysis and performing different computing operations according to the determination result includes: Get the data analysis text and distribute it to the edge server set by the edge computing module. The edge server is set with different server states, including "idle state", "occupied state" and "fault state". When the edge server is in "idle state", the edge server receives the data analysis text and deconstructs the corresponding marking-related data for load analysis. When the edge server is in "occupied state" and "fault state", the edge server corresponding to the current status code does not receive the data analysis text. Get the computing resources required by the edge server to process the marking data set. The value of the computing resource is marked as Cal. 需 , obtain the real-time computing resources in the edge server, denoted as Cal 实 , and then according to Cal 需 and Cal 实 Conduct load analysis; The calculation operation includes calculation operation 1 and calculation operation 2. The corresponding calculation operation is performed according to different judgment results. If Cal 需 ≥Cal 实 , then it is determined that the current edge server cannot perform calculations on the grading-related data, and the corresponding calculation operation is performed; if Cal 需 <Cal 实 , it is determined that the current edge server can satisfy the corresponding calculation and processing of the marking data, and the calculation operation 2 is executed accordingly, and the marking data set is generated through the calculation operation.

6. The artificial intelligence examination system based on edge computing according to claim 5 is characterized in that: The process of labeling and analyzing the marking data set through artificial intelligence algorithms and then establishing the corresponding marking model includes: A marking data set is obtained, which includes several test paper images. The several test paper images are numbered and denoted as x, where x=1, 2, 3, ..., k, where k is a natural number greater than 0. A cutting parameter corresponding to the test paper image numbered x is set, and then the test paper image numbered x is cut into several fixed-area to-be-annotated regions through the cutting parameter, denoted as Z(x, y), which represents the yth to-be-annotated region of the test paper image numbered x, where y=1, 2, 3, ..., g, where g is a natural number greater than 0. An annotation analysis of the Z(x, y) to-be-annotated region is performed through an artificial intelligence algorithm, and then a corresponding annotation text is generated, where the annotation text records comparison keywords of the test paper image, a model building program is set, and the annotation texts corresponding to the several to-be-annotated regions are imported into the model building program, and then a regional marking sub-model is generated, and the regional marking sub-models are summarized and spliced, and then a corresponding marking model is established.

7. The artificial intelligence examination system based on edge computing according to claim 6 is characterized in that: The process of performing the marking operation to generate the marking results and sending them to the user platform includes: Obtain the operation authority of the marking model, set the answer data and score values ​​of several regional marking sub-models corresponding to the marking model through the operation authority, and convert the answer data and score values ​​of the regional marking sub-models into matching text. The matching text is used to record the matching keywords generated by the answer data and the score values, and the comparison keywords and matching keywords are matched one by one through the marking operation to generate the corresponding marking results, and a secure transmission channel is established to send the marking results to the set user platform through the secure transmission channel.

8. The artificial intelligence examination system based on edge computing according to claim 7 is characterized in that: The process of students and teachers logging into the user platform to perform operations related to the marking results includes: The user platform is associated with different types of verification forms, including student verification forms and teacher verification forms. The student verification form records information about all students allowed to log in to the user platform, and the teacher verification form records information about all teachers allowed to log in to the user platform. Several students and teachers enter their corresponding student information and user information on the user platform, and then the student verification form conducts the student login review, and the teacher verification form conducts the teacher login review. When their respective login reviews are passed, students and teachers are granted relevant operations on the marking results, including browsing, printing, downloading and forwarding operations.

Images