Block chain-based paper marking method and apparatus, and program product

Abstract

The invention provides a paper marking method and device based on a block chain and a program product. A smart contract is executed according to the following steps: receiving a picture generated by a test paper; storing the storage address of the picture, the hash value of the picture and the examinee information into a block chain; the answer content of the examinee is extracted, the answer content serves as a new block and is inserted into the block chain, and the answer content comprises objective question answer content and subjective question answer content; scoring the answer content of the objective questions according to the standard answers, and inserting the obtained scores of the objective questions into the block chain as new blocks; performing a scoring step of mutual correction of AI large model scoring and manual scoring scoring on each question in the subjective question content, determining a final score of the subjective question, generating a new block from the final score, and inserting the new block into the block chain; and after completing the examination paper marking of the answer content of all subjective questions, summarizing and filing the scores of the examination paper. By means of the technical scheme, full-process management of the paper marking process can be achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular, to a marking method, device, and program product based on a blockchain. Background Art

[0002] At present, in the marking process of large-scale examinations, test papers are scanned, and teachers mark the papers by reading the scanned copies. For objective questions, the form of 2B answer sheets has basically achieved automation, but for subjective questions, manual marking is still mainly used, and the entire marking process is carried out in a traditional database system. The traditional database system may have problems of data tampering, and there are problems that it is difficult to trace the data processing process, and the data credibility has not reached the extreme value; moreover, manual marking is greatly affected by subjective factors, and there is also a problem of heavy workload in data archiving. Summary of the Invention

[0003] Embodiments of the present invention provide a marking method, device, and program product based on a blockchain to achieve full-process management of the marking process.

[0004] To achieve the above object, on the one hand, a marking method based on a blockchain is provided. The blockchain includes a central scheduling node, and the central scheduling node is used to run a smart contract. The marking method includes the following steps executed by running the smart contract:

[0005] Receive a picture generated from a test paper;

[0006] Store the storage address of the picture, the hash value of the picture, and the candidate information of the test paper into the blockchain;

[0007] Extract the candidate's answer content from the picture, and insert the answer content as a new block into the blockchain. The answer content includes objective question answer content and subjective question answer content;

[0008] Score the objective question answer content according to the standard answer, and insert the obtained objective question score as a new block into the blockchain;

[0009] For each question in the subjective question content, execute the following scoring steps:

[0010] Insert the first score obtained by marking the answer content of the question through a predetermined AI large model as a new block into the blockchain;

[0011] Insert the second score obtained by marking the answer content of the question by a first marking teacher as a new block into the blockchain;

[0012] Compare the first score with the second score, and determine whether the deviation value between the first score and the second score meets the predetermined condition of small deviation or no deviation; if so, use the second score as the final score for the answer content of this question; otherwise:

[0013] Send the first score and the scoring suggestion of the first score obtained by the AI large model to the first grader, and obtain the third score given by the first grader after re-grading; determine whether the deviation value between the first score and the third score meets the predetermined condition of small deviation or no deviation; if so, use the third score as the final score for the answer content of this question; otherwise, send the answer content of this question to the second grader for grading, obtain the fourth score, and determine whether the third score and the fourth score meet the predetermined condition of no deviation; if so, use the third score as the final score for the subjective question answer content, otherwise, send the answer content of this question to the person in charge of the grading group for grading, obtain the fifth score, and use the fifth score as the final score for the answer content of this question;

[0014] Generate a new block with the final score of the answer content of this question and insert it into the blockchain;

[0015] After completing the grading of all subjective question answer contents, summarize and file the scores of this test paper.

[0016] Preferably, for the grading method, the blockchain further includes: a supervision node, which is used for data disaster recovery backup, data supervision and access of the central scheduling node; only the central scheduling node in the blockchain has the permission to insert a new block into the blockchain.

[0017] Preferably, the grading method further includes:

[0018] Establish an association index between the block representation and the examination information, grader information, and question information; and,

[0019] Create and update the association index when updating the blockchain.

[0020] Preferably, for the grading method, the step of summarizing and filing the scores of this test paper includes:

[0021] According to the association index, find the blocks related to the final scores of all questions of this candidate, add the corresponding scores to get the total score, and insert the total score as a new block into the blockchain for archiving.

[0022] Preferably, for the grading method, after grading, it further includes:

[0023] Submit the answer content of all subjective questions and the corresponding final scores to the AI large model for regression training by the AI large model.

[0024] Preferably, after the marking is completed, the marking method further includes:

[0025] After receiving the score tracing requirement, according to the associated index, find all the blocks related to the corresponding candidates and output the information of each block.

[0026] Preferably, in the marking method, the first score, the second score, the third score, the fourth score and the fifth score are percentage scores; divide one hundred points into a predetermined number of grades; set the conditions of small deviation and no deviation between scores according to the number of grade differences between scores.

[0027] Preferably, in the marking method, divide one hundred points into four grades: excellent, good, passing and failing; among them, when the score is greater than or equal to 90, it is excellent, when the score is greater than or equal to 80, it is good, when the score is greater than or equal to 60, it is passing, and when the score is less than 60, it is failing; among them:

[0028] If the difference between two scores is one grade, it is determined that the deviation is small;

[0029] If two scores are in the same grade, it is determined that there is no deviation, otherwise there is deviation.

[0030] On the other hand, a device for realizing marking is also provided, including a memory and a processor. The memory stores at least one program, and the at least one program is executed by the processor to realize the steps of the marking method as described in any one of the above.

[0031] On the other hand, a computer program product is also provided, including a computer program, characterized in that the computer program realizes the steps of the marking method as described in any one of the above when executed by a processor.

[0032] The above technical solution has the following technical effects:

[0033] The technical solution of the embodiment of the present invention can realize the full-process management of the marking process. By running a smart contract and writing each process in the marking process into the blockchain through a central scheduling node, it provides a marking solution with recognized data that is difficult to tamper with, high data credibility and a data processing process that can be traced; and further, by using a predetermined AI large model to mark and score the answer content of subjective questions and compare it with the manual marking score, the AI correction of the marking result of subjective questions is realized, avoiding the problem that the marking result of subjective questions has a significant deviation due to subjective factors, and enhancing the fairness of marking; Description of the Drawings

[0034] Figure 1 The network topology diagram of the blockchain in the blockchain-based marking method according to an embodiment of the present invention;

[0035] Figure 2 The process schematic diagram of the blockchain-based marking method according to an embodiment of the present invention. Detailed implementation manners

[0036] To further illustrate the embodiments, the present invention provides accompanying drawings. These drawings are part of the disclosure of the present invention, mainly used to illustrate the embodiments, and can be used in conjunction with the relevant descriptions in the specification to explain the operation principle of the embodiments. With reference to these contents, those of ordinary skill in the art should be able to understand other possible implementation manners and the advantages of the present invention. The components in the drawings are not drawn to scale, and similar component symbols are usually used to represent similar components.

[0037] The present invention will be further described below in conjunction with the accompanying drawings and specific implementation manners.

[0038] Embodiment 1:

[0039] An embodiment of the present invention provides a blockchain-based marking method. In this embodiment, the blockchain includes: a central scheduling node, and the central scheduling node is used to run a smart contract. The marking method includes the following steps executed by running the smart contract:

[0040] Receiving the pictures generated by the test papers;

[0041] Storing the storage address of the pictures, the hash values of the pictures and the candidate information of the test papers into the blockchain;

[0042] Extracting the candidate's answer content from the pictures and inserting the answer content as a new block into the blockchain. The answer content includes objective question answer content and subjective question answer content;

[0043] Scoring the objective question answer content according to the standard answers and inserting the obtained objective question scores as new blocks into the blockchain;

[0044] Executing the following scoring steps for each question in the subjective question content:

[0045] Inserting the first score obtained by marking the answer content of this question through a predetermined AI large model as a new block into the blockchain;

[0046] Inserting the second score obtained by marking the answer content of this question through the first marking teacher as a new block into the blockchain;

[0047] Compare the first score with the second score, and determine whether the deviation value between the first score and the second score meets the predetermined condition of small deviation or no deviation; if so, use the second score as the final score for the answer content of this question; otherwise:

[0048] Send the first score and the scoring suggestion of the first score obtained by the AI large model to the first grader, and obtain the third score given by the first grader after re-grading; determine whether the deviation value between the first score and the third score meets the predetermined condition of small deviation or no deviation; if so, use the third score as the final score for the answer content of this question; otherwise, send the answer content of this question to the second grader for grading, obtain the fourth score, and determine whether the third score and the fourth score meet the predetermined condition of no deviation; if so, use the third score as the final score for the subjective question answer content, otherwise, send the answer content of this question to the person in charge of the grading group for grading, obtain the fifth score, and use the fifth score as the final score for the answer content of this question;

[0049] Generate a new block with the final score of the answer content of this question and insert it into the blockchain;

[0050] After completing the grading of all subjective question answer contents, summarize and file the scores of this test paper.

[0051] Among them, the third score, the fourth score, and the fifth score during the grading process and the information of the corresponding graders are all written into the blockchain as new blocks.

[0052] In the embodiment of the present invention, each process during the grading process is written into the blockchain through the central scheduling node, realizing the full-process management of the grading process.

[0053] Preferably, only the central scheduling node has the permission to write new blocks into the blockchain; this avoids tampering during the grading process.

[0054] The embodiment of the present invention realizes grading based on AI and blockchain; specifically, it uses the characteristics of the blockchain to ensure the openness, transparency, non-tamperability, security, and credibility of the test paper and the score; through the design of the blockchain management mechanism of centralized writing and distributed storage, the orderly operation of the blockchain is guaranteed; through the designed grading smart contract of the blockchain, the standardization and contractization of the grading process are guaranteed, realizing the full-process management of the grading process, including: grading of objective questions; and for subjective questions, mutual correction between scoring subjective questions using a pre-selected AI large model and manual grading, improving the fairness of subjective question scoring.

[0055] Embodiment Two:

[0056] The overall process of the grading method of this embodiment of the present invention includes the following steps:

[0057] Scan the test paper to generate a picture of the test paper, and submit the picture to the smart contract running on the central scheduling node. The smart contract starts the processing process of the test paper picture;

[0058] The smart contract extracts the answering content in the test paper through the pre-set text recognition algorithm in the picture;

[0059] The smart contract completes the marking of objective questions;

[0060] The smart contract hands over the process to the AI large model and humans to complete the marking of subjective questions;

[0061] The smart contract conducts summary archiving of the test paper scores;

[0062] The AI large model conducts regression training;

[0063] Trace the scores according to the requirements.

[0064] Next, a detailed description will be given to the blockchain design adopted in the embodiments of the present invention and the blockchain management mechanism for centralized writing and distributed storage.

[0065] Figure 1 This is the blockchain network topology diagram of this embodiment of the present invention. As Figure 1 , the blockchain system of the embodiments of the present invention adopts a distributed network architecture, which includes two types of blockchain network nodes: central scheduling nodes and supervision nodes. Among them, each node maintains a blockchain, that is, a local blockchain. In the case of complete synchronization, the content of all node blockchains is the same.

[0066] The blockchain network is connected to various clients, including: marking clients, AI marking large model services, AI picture text recognition services, test paper scanning services, score tracing clients, etc.

[0067] Blockchain design

[0068] The working mode of the blockchain in the embodiments of the present invention has the following similarities and differences with the traditional blockchain:

[0069] The difference from the traditional Bitcoin blockchain system is that the blockchain in the embodiments of the present invention cancels the proof-of-work mechanism and instead adopts a mode of writing by the central scheduling node and read-only by the supervision node. The writing steps of the blockchain in the embodiments of the present invention include:

[0070] When each client needs to insert a block into the blockchain, it needs to apply to the central scheduling node for writing;

[0071] After receiving the request, the central scheduling node writes the block into the blockchain in sequence according to the order of receiving the requests, and then broadcasts the newly added block to other blockchain network nodes;

[0072] When other blockchain network nodes receive the broadcast, blockchain synchronization is performed.

[0073] Through the above centralized writing steps, network resource consumption is reduced, and the problems of out-of-sync and inconsistency when each node of the traditional blockchain generates a new block are avoided.

[0074] The blockchain of the embodiment of the present invention: According to the characteristics of newly added blocks in the blockchain, the blockchain can only insert new blocks into the blockchain and cannot modify the inserted blocks; this enables new marking content to be inserted only as new blocks, thus ensuring that the entire marking process is tamper-proof and traceable.

[0075] The blockchain of the embodiment of the present invention: Based on the distributed characteristics of the blockchain, the marking scores are distributed and stored in each node of the blockchain network; if the data of any network node is damaged, it can be easily restored from other nodes. This ensures the disaster tolerance and tamper-proof capabilities of the system.

[0076] The blockchain network nodes in the blockchain network of the embodiment of the present invention include:

[0077] The central scheduling node, which is responsible for running the smart contract and scheduling and managing the entire marking process; in the blockchain designed in the embodiment of the present invention, only the central scheduling node has the permission to insert new blocks;

[0078] The supervision node, generally various supervision departments, enrollment departments, etc.; its role is to serve as a disaster tolerance backup of the data of the central scheduling node, as well as data supervision and access; such nodes do not have the permission to actively add new blocks, and only have the permission to receive the instruction of the central scheduling node to synchronize blocks.

[0079] The data structure design of the letter blockchain in the blockchain network of the embodiment of the present invention:

[0080] The data structure of the letter blockchain is similar to that of the traditional blockchain, and it is also a chain structure formed by connecting blocks one after another; among them, a block consists of a block header and a block body. The role of the block header is to verify the accuracy of the data and prevent the block information from being tampered with. The block header includes: 1. The hash value of the previous block header; 2. The timestamp; 3. The hash value of the current block body.

[0081] In the embodiment of the present invention, compared with the block header of the traditional blockchain, since the blockchain is updated by the method of the central scheduling node keeping accounts, the difficulty and Nonce answer fields are cancelled.

[0082] The principle of anti-tampering implementation in the embodiment of the present invention is: The block header contains the hash value of the block header of the previous block, and the hash value of the block header of the previous block in turn contains the hash value of its own block body; to tamper with the content of any block will cause the hash values of the subsequent blocks to not match.

[0083] The function of the block body is to store data. In the embodiments of the present invention, the block body includes: the hash value of the scanned test paper file such as the test paper picture, the storage path of the scanned file, i.e., the picture, candidate information such as candidate identity information, question numbers and corresponding candidate answers, marking teacher information and their scores, AI engine information and their scores, etc.

[0084] An example of a data structure is as follows:

[0085]

[0086]

[0087] The unique identifier of the block

[0088] Each block uses the serial number of the block in the blockchain as the unique identifier, i.e., the block ID; for example, the identifier of the first block is 0, and the identifier of the 12th block is 11, and so on.

[0089] The index of the block

[0090] Establish an associated index between the block ID and information such as candidate numbers, marking teacher numbers, question numbers, etc., to speed up the search speed, and create and update the index simultaneously when updating the blockchain; information such as candidate numbers, marking teacher numbers, question numbers, etc. can quickly find the block where the information is located.

[0091] Figure 2 It is a flow chart of the marking method based on blockchain in an embodiment of the present invention. The embodiments of the present invention design a marking intelligent contract for blockchain to ensure the standardization and contractualization of the marking process and realize the full-process management of the marking process. Among them, all processes and results such as test paper scanning, text recognition of the scanned test paper, AI marking, intelligent contract marking, manual marking, secondary marking, marking review, etc. are stored in the blockchain, so that the whole marking process can be traced.

[0092] The definition of a blockchain intelligent contract is a computer protocol designed to spread, verify, or execute a contract in an information-based manner. It allows for trusted blockchain management without a third party, and these blockchain managements are traceable and irreversible.

[0093] Such as Figure 2 , the marking intelligent contract of the embodiments of the present invention includes:

[0094] 1. Scan the test paper to generate a test paper picture. The picture is submitted to the intelligent contract, and the intelligent contract starts the processing process. The intelligent contract stores the picture storage address, the hash value of the picture, the answer sheet information, and the candidate information into a new block in the blockchain; stores the picture storage address and the hash value into the blockchain to prevent the scanned picture from being tampered with;

[0095] Among them, the above-mentioned answer sheet information, i.e., the relevant information of the test paper, includes: the test questions of the test paper and their corresponding question types, the standard answers corresponding to the test questions with standard answers, the numbers of the test questions, and the numbers identifying the test paper types; among them, the question types include: objective questions and subjective questions; among them, the types of objective questions also include: fill-in-the-blank questions, true or false questions, multiple-choice questions, and / or multiple-answer questions; the types of subjective questions also include: essay questions; the types of test papers include: the subjects to which the test papers belong. In a specific implementation, predetermined fields or numbers are used to represent the test paper types and / or question types.

[0096] In a specific implementation, before marking the papers, the following information digitization preparation work for the test questions and standard answers is to be carried out, and information is entered into the smart contract as the basis for the smart contract to mark the papers:

[0097] 1) When the question type is an objective question, there is a standard answer; there is no standard answer for subjective questions; it is input to the smart contract through a question type field such as "type";

[0098] 2) Each test paper has its unique number according to the test paper type, for example, Chinese: 001, Mathematics: 002, etc.;

[0099] 3) The questions of each test paper include the small question types, such as: true or false questions, multiple-choice questions, multiple-answer questions, essay questions, etc.; the large question types: subjective questions and objective questions, and the small question types can be classified into the corresponding large question types; the question number, the unique code of this question.

[0100] 2. The smart contract extracts the content of the answer sheet through the text recognition algorithm of the picture; in a specific implementation, a variety of predetermined text recognition algorithms are used, and these text recognition algorithms are existing algorithms. The answer content of the scanned answer sheet is recognized and output through each algorithm, and the results are compared; if the recognition results of each algorithm are the same, the candidate's answer content is generated as a block body to generate a new block and inserted into the blockchain; if they are different, it is handed over to a dedicated person for verification, and after verification and processing, the answer content is generated as a new block and inserted into the blockchain;

[0101] 3. The smart contract completes the marking of objective questions; objective questions include: multiple-choice questions, multiple-answer questions, fill-in-the-blank questions, true or false questions, etc., and there are clear answers; the smart contract compares the candidate's answer sheet with the standard answer; if they are the same, the candidate gets points, if they are different, the candidate does not get points; the scores of the objective questions are used as new blocks and inserted into the blockchain.

[0102] In a specific implementation, when marking the papers, the smart contract enables the marking process corresponding to the test paper number according to the test paper number in the answer sheet information; in the marking process, the smart contract obtains the type of the question according to the question number, and distinguishes the marking processes of subjective questions and objective questions according to whether the question type is a subjective question or an objective question.

[0103] 4. The smart contract hands over the process to the predetermined AI big model to complete the marking of subjective questions. In a specific implementation, the existing relatively mature AI big model is used as the AI ​​marking big model. The marking big model marks the papers, and after the marking is completed, the percentage score of the question is fed back to the smart contract, and the smart contract inserts the score into the blockchain as a new block.

[0104] 5. The smart contract will hand over the process to the examiner for examination;

[0105] After the examiner has finished grading the papers, he / she will feed back the actual score of the question to the smart contract, which will insert the score into the blockchain as a new block.

[0106] 6. The smart contract calculates the deviation between the AI ​​score and the manual score according to the deviation comparison method; in a specific implementation, the deviation comparison method includes the following steps:

[0107] The AI ​​big model generates a score of 100 for this question, recorded as aiScore. When it is greater than or equal to 90, it is excellent in the first tier; when it is greater than or equal to 80, it is good in the second tier; when it is greater than or equal to 60, it is passing in the third tier; and when it is less than 60, it is failing in the fourth tier;

[0108] The full score of this question is recorded as fullScore, and the score of the examiner is recorded as teacherRealScore. The score of this question is converted into a percentage system by dividing teacherRealScore by fullScore and multiplying by 100, and the teacher's score in percentage is recorded as teacherScore; the same as above is also divided into four levels: excellent, good, pass, and fail.

[0109] If the difference between the two scores is greater than or equal to two, it is considered to be a large deviation; for example, excellent and passing, good and failing are all considered to be a large deviation;

[0110] If the two ratings differ by one level, the deviation is assessed as small;

[0111] If the two scores are in the same range, they are considered unbiased, otherwise they are considered biased.

[0112] The percentage system, the grade division method, the small deviation condition and the no deviation condition of the above-mentioned scoring deviation comparison method are all exemplary. In other implementations, other scoring systems, division methods and deviation determination conditions can be selected;

[0113] 7. For answer sheets with little or no deviation between AI and manual marking, the smart contract uses the teacher’s actual score teacherRealScore as the final score for the question, generates a new block and stores it in the blockchain;

[0114] 8. For answer sheets with large deviations between AI marking and manual marking, the smart contract submits the AI score and suggestions to the original marker for judgment; if the score after adoption has a small deviation or no deviation from the AI score, the modified score is used as the final score for this question and a new block is generated and stored in the blockchain.

[0115] 9. If there are still large deviations in the judgment of the original teacher, the smart contract submits the answer sheet to other teachers for secondary marking.

[0116] 10. If there is no deviation between the secondary marking and the original marker, the score of the original marker is used as the final score for this question and a new block is generated and stored in the blockchain.

[0117] 11. If there is a deviation between the secondary marking and the original marker, it is distributed to the person in charge of the marking group for judgment to obtain the score, which is used as the final score for this question and a new block is generated and stored in the blockchain.

[0118] The steps from 4 to 11 above are the methods of the smart contract for mutual correction between AI and manual marking of subjective questions. In the embodiments of the present invention, AI marking is used as an auxiliary correction, and the final score is based on the score of the original marker or the person in charge of the marking group, ensuring the objectivity and fairness of the marking process.

[0119] 12. Through the steps from 4 to 11 for each subjective question, the marking of all subjective questions in this test paper is completed in sequence.

[0120] 13. After the smart contract completes the marking of all subjective questions in this test paper, it summarizes and archives the scores of this answer sheet.

[0121] The smart contract quickly finds all the blocks of the final scores of all questions involving this candidate according to the index mentioned above, adds up the scores to get the total score, inserts it as a new block into the blockchain for archiving.

[0122] 14. After the scores are announced, all the answer sheets and final scores of subjective questions are handed over to the AI large model for regression training.

[0123] 15. If there is a need for score traceability, the smart contract quickly finds all the blocks involving this candidate according to the index mentioned above and outputs the information of each block.

[0124] Embodiment 3:

[0125] The present invention also provides a device for realizing marking, including a memory and a processor. The memory stores at least one segment of program, and the at least one segment of program is executed by the processor to realize the steps of any of the above method embodiments. In a specific implementation, the device for realizing marking is the central scheduling node in the above method embodiments.

[0126] Further, the above device may be a computer unit, which may be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The computer unit may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above composition structure of the computer unit is only an example of the computer unit, and does not constitute a limitation on the computer unit. It may include more or fewer components than the above, or combine some components, or different components. For example, the computer unit may further include input / output devices, network access devices, buses, etc., and the embodiments of the present invention do not limit this.

[0127] Further, as an executable solution, the so-called processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the computer unit, and connects various parts of the entire computer unit through various interfaces and lines.

[0128] The memory can be used to store the computer program and / or modules. The processor realizes various functions of the computer unit by running or executing the computer program and / or modules stored in the memory, and by calling the data stored in the memory. The memory may mainly include a program storage area and a data storage area. Among them, the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created according to the use of the mobile phone, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, at least one magnetic disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0129] Embodiment 4:

[0130] The present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the above method in the embodiments of the present invention are realized.

[0131] If the modules / units integrated in the computer unit are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such understanding, to implement all or part of the processes in the above-described embodiment methods of the present invention, it can also be completed by a computer program instructing relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-described method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), and software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction.

[0132] Embodiment Five:

[0133] The present invention also provides a computer program product, including a computer program, and when the computer program is executed by a processor, the steps of the marking method as described above are implemented.

[0134] Although the present invention is specifically shown and described in combination with the preferred implementation embodiments, those skilled in the art should understand that various changes can be made to the present invention in terms of form and details without departing from the spirit and scope of the present invention defined by the appended claims, and all of them are within the protection scope of the present invention.

Claims

1. A blockchain-based examination paper marking method, characterized in that: The blockchain includes: a central scheduling node, the central scheduling node is used to run the smart contract, and the marking method includes the following steps of running the smart contract: Receive the image generated by the test paper; The storage address of the image, the hash value of the image and the examinee information of the test paper are stored in the blockchain; Extract the examinee's answer content from the image, and insert the answer content as a new block into the blockchain, wherein the answer content includes the answer content of the objective question and the answer content of the subjective question; Scoring the objective questions according to the standard answers, and inserting the obtained objective questions scores into the blockchain as new blocks; For each of the subjective questions, the following scoring steps are performed: The first score obtained by grading the answer content of the question through the predetermined AI big model is inserted into the blockchain as a new block; The second score obtained by the first examiner from examining the answer to the question is inserted into the blockchain as a new block; The first score is compared with the second score, and it is determined whether the deviation between the first score and the second score satisfies a predetermined small deviation condition or no deviation condition; if so, the second score is used as the final score of the answer content of the question; otherwise: Send the first score and the scoring suggestion of the first score obtained by the AI ​​big model to the first examiner, and obtain the third score scored by the first examiner after re-examination; judge whether the deviation value between the first score and the third score meets the predetermined small deviation condition or no deviation condition; if so, use the third score as the final score of the answer content of the question; otherwise, send the answer content of the question to the second examiner for examination, obtain a fourth score, and judge whether the third score and the fourth score meet the predetermined no deviation condition, if so, use the third score as the final score of the answer content of the subjective question, otherwise, send the answer content of the question to the person in charge of the examination group for examination, obtain a fifth score, and use the fifth score as the final score of the answer content of the question; Generate a new block with the final score of the answer to the question and insert it into the blockchain; After completing the grading of all the subjective questions, the scores of the test paper will be summarized and archived.

2. The marking method according to claim 1, characterized in that: The blockchain also includes: a supervision node, which is used to serve as the data disaster recovery backup, data supervision and review of the central scheduling node; in the blockchain, only the central scheduling node has the authority to insert new blocks into the blockchain.

3. The marking method according to claim 1, characterized in that: Also includes: Establish the association index between the block representation and the examination information, the marking teacher information, and the examination question information; And, the associated index is created and updated when the blockchain is updated.

4. The method for marking papers according to claim 3, characterized in that: The steps to compile and archive the test results include: According to the associated index, the block involving the final scores of all questions of the examinee is found, the corresponding scores are added up to obtain the total score, and the total score is inserted into the blockchain as a new block for archiving.

5. The marking method according to claim 1, characterized in that: After the examination, it also includes: The answers to all subjective questions and the corresponding final scores are submitted to the AI ​​big model for regression training.

6. The marking method according to claim 3, characterized in that: After the examination, it also includes: After receiving the score tracing request, all blocks related to the corresponding examinee are found according to the associated index, and the information of each block is output.

7. The method for marking papers according to claim 1, characterized in that: The first score, the second score, the third score, the fourth score and the fifth score are percentage scores; one hundred points are divided into a predetermined number of multiple levels; and the small deviation condition and the no deviation condition between the scores are set according to the number of different levels between the scores.

8. The method for marking papers according to claim 7, characterized in that: The 100 points are divided into four levels: excellent, good, pass, and fail; when the score is greater than or equal to 90, it is excellent; when the score is greater than or equal to 80, it is good; when the score is greater than or equal to 60, it is pass; when the score is less than 60, it is fail; among them: If the two scores differ by one level, the deviation is determined to be small; If the two scores are in the same range, it is determined to be unbiased, otherwise it is determined to be biased.

9. A device for marking papers, characterized in that: The method comprises a memory and a processor, wherein the memory stores at least one program, and the at least one program is executed by the processor to implement the steps of the paper marking method according to any one of claims 1 to 8.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the examination paper marking method according to any one of claims 1 to 8 are implemented.

Images