Intelligent automatic distribution method, system and equipment for college paper marking experts and medium

Abstract

The invention discloses an intelligent automatic distribution method, system and device for college paper marking experts and a medium. The method comprises the steps of calculating a first attribute weight, a second attribute weight and a third attribute weight; inputting the historical data corresponding to each fourth attribute into a linear regression equation to obtain a fourth attribute weight of each fourth attribute; selecting a preset number of target experts; based on the multiple first attributes, the first attribute weights, the multiple second attributes, the second attribute weights, the multiple third attributes, the third attribute weights, the multiple fourth attributes and the fourth attribute weights, the test questions and the target experts are matched, and a matching success record set is obtained; and performing matching effect evaluation on the successful matching record set to obtain an evaluation result, and if the evaluation result is greater than or equal to a preset threshold value, taking the successful matching record set as a target matching result, and allocating paper marking experts according to the target matching result. The accuracy of matching between the expert and the test paper to be read is improved, and the matching efficiency is improved.

Description

Technical Field

[0001] This application relates to the field of educational technology, and in particular to a method, system, device and medium for intelligent automatic allocation of grading experts in universities. Background Technology

[0002] Currently, universities use expert grading to assess student learning. The effectiveness of automated matching between experts and exam papers is a crucial issue. This is because experts have diverse research areas, and university courses offer a wide variety of exam papers. Manual allocation is not only inefficient but also yields unsatisfactory matching results. Existing automated methods for assigning grading experts fail to provide a reasonable match between experts and exam papers, resulting in suboptimal matching and reduced accuracy. Therefore, finding a reasonable way to match experts with exam papers has become an important research direction. Summary of the Invention

[0003] This application aims to propose a method, system, equipment, and medium for intelligent automatic allocation of examination paper marking experts in universities, which can improve the accuracy and efficiency of matching experts with examination papers to be marked.

[0004] In a first aspect, embodiments of this application provide a method for intelligent automatic allocation of university marking experts, the method comprising: Retrieve multiple first attributes from the set of test papers to be matched, multiple second attributes from the set of test questions to be matched, multiple third attributes from the set of marking experts to be matched, and multiple fourth attributes from the set of historical marking records of experts; Calculate the first attribute weight for each of the first attributes, calculate the second attribute weight for each of the second attributes, and calculate the third attribute weight for each of the third attributes; The historical data corresponding to each of the fourth attributes in the expert historical marking record set is input into the linear regression equation to obtain the fourth attribute weight of each of the fourth attributes; Select a preset number of target experts from the set of experts to be matched; Based on the plurality of first attributes, the weight of the first attributes, the plurality of second attributes, the weight of the second attributes, the plurality of third attributes, the weight of the third attributes, the plurality of fourth attributes, and the weight of the fourth attributes, the test questions and the target experts are matched to obtain a set of successfully matched records; The matching effect of the set of successfully matched records is evaluated to obtain the evaluation result. If the evaluation result is greater than or equal to a preset threshold, the set of successfully matched records is taken as the target matching result, and marking experts are assigned according to the target matching result.

[0005] Compared with the prior art, the first aspect of this application has the following beneficial effects: This method obtains multiple first attributes from the set of test papers to be matched, multiple second attributes from the set of test questions to be matched, multiple third attributes from the set of marking experts to be matched, and multiple fourth attributes from the set of historical marking records of experts. It calculates the weight of the first attribute for each first attribute, the weight of the second attribute for each second attribute, and the weight of the third attribute for each third attribute. It inputs the historical data corresponding to each fourth attribute in the set of historical marking records of experts into a linear regression equation to obtain the weight of each fourth attribute. It selects a preset number of target experts from the set of marking experts to be matched. Based on multiple first attributes, first attribute weights, multiple second attributes, second attribute weights, multiple third attributes, third attribute weights, multiple fourth attributes, and fourth attribute weights, it matches test questions with target experts to obtain a set of successfully matched records. It evaluates the matching effect of the successfully matched record set and obtains the evaluation result. If the evaluation result is greater than or equal to a preset threshold, the successfully matched record set is taken as the target matching result, and marking experts are assigned to the test questions to be matched based on the target matching result. Thus, calculating the attribute weight of each attribute through computation, instead of obtaining attribute weights one by one using linear regression equations, improves computational efficiency and saves computational resources. Matching questions with target experts by selecting a predetermined number from the set of experts to be matched, rather than performing matching calculations with all experts, avoids excessive computation and improves matching efficiency. Evaluating the matching effect of the successfully matched records lays a solid data foundation for improving the accuracy of matching between experts and exam papers. Finally, automatically assigning marking experts to exam papers based on the target matching results improves the accuracy of matching between experts and exam papers, and also increases matching efficiency.

[0006] In some implementations, calculating the first attribute weight for each of the first attributes, calculating the second attribute weight for each of the second attributes, and calculating the third attribute weight for each of the third attributes includes: ; in, This represents the set of test papers to be matched, the set of test questions to be matched, or the set of marking experts to be matched. This represents the first attribute in the set of test papers to be matched, the second attribute in the set of test questions to be matched, or the third attribute in the set of examiners to be matched. Indicates the weight of the first attribute, the weight of the second attribute, or the weight of the third attribute. When equal to 1, This represents the ratio between the number of positive records contained in the current attribute and the total number of records contained in the current attribute. When equal to 2, This represents the ratio between the number of negative records in the current attribute and the total number of records in the current attribute. Indicates the first [item] in the current attribute The value includes the number of all records. This indicates the total number of records contained in the current attribute. Indicates the range of values ​​that the current attribute can take. When equal to 1, Indicates the first [item] in the current attribute The value includes the number of positive records and the number of records in the current attribute. The value is the ratio between the number of records included in the range. When equal to 2, Indicates the first [item] in the current attribute The number of negative records included in the value and the number of records in the current attribute The value represents the ratio between the number of records included in the range.

[0007] In some implementations, the step of matching test questions with target experts based on the plurality of first attributes, the weights of the first attributes, the plurality of second attributes, the weights of the second attributes, the plurality of third attributes, the weights of the third attributes, the plurality of fourth attributes, and the weights of the fourth attributes to obtain a set of successfully matched records includes: Obtain the weights of the first set of test papers to be matched, the weights of the second set of test questions to be matched, the weights of the third set of marking experts to be matched, and the weights of the fourth set of experts' historical marking records. The plurality of first attributes are constructed into a first attribute matrix, the plurality of second attributes are constructed into a second attribute matrix, the plurality of third attributes are constructed into a third attribute matrix, and the plurality of fourth attributes are constructed into a fourth attribute matrix; A first bias is calculated based on the first attribute weight, a second bias is calculated based on the second attribute weight, and a third bias is calculated based on the third attribute weight. Input all historical data from the expert historical grading record set into the linear regression equation to obtain the fourth bias corresponding to all fourth attributes; A first weight matrix is ​​constructed based on the first attribute weight of each first attribute, a second weight matrix is ​​constructed based on the second attribute weight of each second attribute, a third weight matrix is ​​constructed based on the third attribute weight of each third attribute, and a fourth weight matrix is ​​constructed based on the fourth attribute weight of each fourth attribute. Based on the first set weights, the second set weights, the third set weights, the fourth set weights, the first weight matrix, the second weight matrix, the third weight matrix, the fourth weight matrix, the first attribute matrix, the second attribute matrix, the third attribute matrix, the fourth attribute matrix, the first bias, the second bias, the third bias, and the fourth bias, the test questions and the target experts are matched to obtain a set of successfully matched records.

[0008] In some implementations, the matching of test questions and experts based on the first set weights, the second set weights, the third set weights, the fourth set weights, the first weight matrix, the second weight matrix, the third weight matrix, the fourth weight matrix, the first attribute matrix, the second attribute matrix, the third attribute matrix, the fourth attribute matrix, the first bias, the second bias, the third bias, and the fourth bias to obtain a set of successfully matched records includes: ; in, Indicates the degree of matching. Represents the normalized exponential function, Indicates the weight of the first set. Indicates the weight of the second set. Indicates the weight of the third set. Indicates the weight of the fourth set. This represents the first weight matrix. This represents the second weight matrix. This represents the third weight matrix. This represents the fourth weight matrix. Represents the first attribute matrix, This represents the second attribute matrix. Represents the third attribute matrix. This represents the fourth attribute matrix. Indicates the first bias. This indicates the second bias. Indicates the third bias. Indicates the fourth bias. Indicates matrix transpose; Record a set of successfully matched records corresponding to the matching degree, wherein the attributes in the set of successfully matched records include a first attribute, a second attribute, and a third attribute.

[0009] In some implementations, evaluating the matching effect of the successfully matched record set to obtain the evaluation result includes: Select multiple first target attributes from the set of successfully matched records, and select multiple second target attributes from the set of expert historical grading records, wherein the first target attributes and the second target attributes are the same; Based on the first target attribute and the second target attribute, calculate the vector distance between each record in the successfully matched record set and each record in the expert historical grading record set to obtain a vector distance result set; Based on the vector distance result set, the matching effect of the successfully matched record set is evaluated to obtain the evaluation result.

[0010] In some implementations, the step of evaluating the matching effect of the set of successfully matched records based on the set of vector distance results to obtain an evaluation result includes: Get the total number of questions to be matched in the question set; Sort all vector distance results in the vector distance result set to obtain the vector distance sorting result; Divide the vector distance sorting result into several equal parts and set the return value corresponding to each equal part; Based on the multiple vector distances between each record in the set of successfully matched records and all records in the set of expert historical review records, determine which target segment among the several segments the minimum vector distance falls into, and return the return value corresponding to the target segment, wherein the target segment is one of the several segments; Calculate the ratio between the return value corresponding to the target equal segment and the total number of questions to be matched in the question set, and obtain the ratio result; Based on the ratio result, the matching effect of the successfully matched record set is evaluated to obtain the evaluation result.

[0011] In some implementations, after evaluating the matching effect of the set of successfully matched records and obtaining the evaluation result, the method further includes: If the evaluation result is less than the preset threshold, then a preset number of target experts are selected from the set of matching marking experts to obtain new target experts; Based on the plurality of first attributes, the weight of the first attributes, the plurality of second attributes, the weight of the second attributes, the plurality of third attributes, the weight of the third attributes, the plurality of fourth attributes, and the weight of the fourth attributes, the test questions are matched with the new target experts to obtain a new set of successfully matched records; The matching effect of the new set of successfully matched records is evaluated to obtain a new evaluation result. The new set of successfully matched records is taken as the target matching result until the new evaluation result is greater than or equal to a preset threshold.

[0012] Secondly, embodiments of this application also provide an intelligent automatic allocation system for university marking experts, the system comprising: The data acquisition unit is used to acquire multiple first attributes from the set of test papers to be matched, multiple second attributes from the set of test questions to be matched, multiple third attributes from the set of marking experts to be matched, and multiple fourth attributes from the set of historical marking records of experts. The first calculation unit is used to calculate the first attribute weight of each first attribute, the second attribute weight of each second attribute, and the third attribute weight of each third attribute. The weighting unit is used to input the historical data corresponding to each of the fourth attributes in the expert historical marking record set into the linear regression equation to obtain the fourth attribute weight of each of the fourth attributes; The expert selection unit is used to select a preset number of target experts from the set of experts to be matched for marking. An expert matching unit is used to match test questions with target experts based on the plurality of first attributes, the weights of the first attributes, the plurality of second attributes, the weights of the second attributes, the plurality of third attributes, the weights of the third attributes, the plurality of fourth attributes, and the weights of the fourth attributes, to obtain a set of successfully matched records; An expert allocation unit is used to evaluate the matching effect of the set of successfully matched records and obtain an evaluation result. If the evaluation result is greater than or equal to a preset threshold, the set of successfully matched records is taken as the target matching result, and marking experts are assigned according to the target matching result.

[0013] Thirdly, embodiments of this application also provide an electronic device, including at least one control processor and a memory for communicatively connecting to the at least one control processor; the memory stores instructions executable by the at least one control processor, the instructions being executed by the at least one control processor to enable the at least one control processor to execute the intelligent automatic allocation method for university grading experts as described above.

[0014] Fourthly, embodiments of this application also provide a computer-readable storage medium storing computer-executable instructions for causing a computer to execute the above-described intelligent automatic allocation method for university exam marking experts.

[0015] It is understood that the beneficial effects of the second to fourth aspects compared with the related technologies are the same as the beneficial effects of the first aspect compared with the related technologies. Please refer to the relevant description in the first aspect above, which will not be repeated here. Attached Figure Description

[0016] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1 This is a flowchart illustrating an embodiment of the intelligent automatic allocation method for university exam marking experts provided in this application; Figure 2 This is a schematic diagram of the overall process in the best embodiment of the intelligent automatic allocation method for university marking experts provided in this application; Figure 3 This is a schematic diagram of the structure of an embodiment of the intelligent automatic allocation system for university marking experts provided in this application; Figure 4 This is a schematic diagram of the structure of an embodiment of the electronic device provided in this application. Detailed Implementation

[0017] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application.

[0018] In the description of this application, the use of terms such as "first," "second," etc., is for the purpose of distinguishing technical features only and should not be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated or the order of the technical features indicated.

[0019] In the description of this application, it should be understood that the orientation descriptions, such as up, down, etc., are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0020] In the description of this application, it should be noted that, unless otherwise explicitly defined, terms such as "setup," "installation," and "connection" should be interpreted broadly, and those skilled in the art can reasonably determine the specific meaning of the above terms in this application in conjunction with the specific content of the technical solution.

[0021] To address the issue of how to rationally match experts with exam papers to be graded, this application proposes a method, system, equipment, and medium for intelligent automatic allocation of university exam grading experts.

[0022] Reference Figure 1This application provides a flowchart illustrating the intelligent automatic allocation method for university exam marking experts. This method is applied to an electronic device, which can be a server or a mobile terminal, etc. Figure 1 As shown, the intelligent automatic allocation method for university marking experts may include the following steps: Step S101: Obtain multiple first attributes from the set of test papers to be matched, multiple second attributes from the set of test questions to be matched, multiple third attributes from the set of marking experts to be matched, and multiple fourth attributes from the set of expert historical marking records; Step S102: Calculate the first attribute weight of each first attribute, calculate the second attribute weight of each second attribute, and calculate the third attribute weight of each third attribute. Step S103: Input the historical data corresponding to each fourth attribute in the expert historical marking record set into the linear regression equation to obtain the fourth attribute weight of each fourth attribute; Step S104: Select a preset number of target experts from the set of experts to be matched for marking; Step S105: Based on multiple first attributes, first attribute weights, multiple second attributes, second attribute weights, multiple third attributes, third attribute weights, multiple fourth attributes, and fourth attribute weights, match the test questions with the target experts to obtain a set of successfully matched records; Step S106: Evaluate the matching effect of the set of successfully matched records and obtain the evaluation result. If the evaluation result is greater than or equal to the preset threshold, the set of successfully matched records is taken as the target matching result, and the marking experts are assigned according to the target matching result.

[0023] In this embodiment, multiple first attributes from the set of test papers to be matched, multiple second attributes from the set of test questions to be matched, multiple third attributes from the set of marking experts to be matched, and multiple fourth attributes from the set of historical marking records of experts are obtained. The weight of the first attribute for each first attribute, the weight of the second attribute for each second attribute, and the weight of the third attribute for each third attribute are calculated. The historical data corresponding to each fourth attribute in the set of historical marking records of experts are input into a linear regression equation to obtain the weight of each fourth attribute. A preset number of target experts are selected from the set of marking experts to be matched. Based on multiple first attributes, first attribute weights, multiple second attributes, second attribute weights, multiple third attributes, third attribute weights, multiple fourth attributes, and fourth attribute weights, the test questions and target experts are matched to obtain a set of successfully matched records. The matching effect of the set of successfully matched records is evaluated to obtain an evaluation result. If the evaluation result is greater than or equal to a preset threshold, the set of successfully matched records is taken as the target matching result, and marking experts are assigned to the test questions to be matched according to the target matching result. Thus, calculating the attribute weight of each attribute through computation, instead of obtaining attribute weights one by one using linear regression equations, improves computational efficiency and saves computational resources. Matching questions with target experts by selecting a predetermined number from the set of experts to be matched, rather than performing matching calculations with all experts, avoids excessive computation and improves matching efficiency. Evaluating the matching effect of the successfully matched records lays a solid data foundation for improving the accuracy of matching between experts and exam papers. Finally, automatically assigning marking experts to exam papers based on the target matching results improves the accuracy of matching between experts and exam papers, and also increases matching efficiency.

[0024] The primary attributes of the aforementioned set of test papers to be matched can include course category, course classification, test paper classification, and test paper examination intent.

[0025] The multiple second attributes in the above set of questions to be matched can include question type classification and question testing intent.

[0026] The third attributes in the aforementioned set of experts to be matched for marking can include research direction and discipline.

[0027] The aforementioned set of expert historical marking records may include multiple fourth attributes such as course category, course classification, exam paper classification, exam paper examination intent, question type classification, question examination intent, research direction, and subject, that is, all attributes in the historical matching exam paper set, the historical matching exam question set, and the historical matching marking expert set.

[0028] The linear regression equation mentioned above can be one of the statistical analysis methods that uses regression analysis in mathematical statistics to determine the quantitative relationship of interdependence between two or more variables.

[0029] The aforementioned preset quantity can be a value set based on historical experience, and this embodiment does not impose specific limitations on it.

[0030] The aforementioned preset threshold can be a value set based on historical experience, and this embodiment does not impose any specific limitations on it.

[0031] In some implementations, calculating the first attribute weight for each first attribute, calculating the second attribute weight for each second attribute, and calculating the third attribute weight for each third attribute includes: ; in, This represents the set of test papers to be matched, the set of test questions to be matched, or the set of marking experts to be matched. This represents the first attribute in the set of test papers to be matched, the second attribute in the set of test questions to be matched, or the third attribute in the set of examiners to be matched. Indicates the weight of the first attribute, the weight of the second attribute, or the weight of the third attribute. When equal to 1, This represents the ratio between the number of positive records contained in the current attribute and the total number of records contained in the current attribute. When equal to 2, This represents the ratio between the number of negative records in the current attribute and the total number of records in the current attribute. Indicates the first [item] in the current attribute The value includes the number of all records. This indicates the total number of records contained in the current attribute. Indicates the range of values ​​that the current attribute can take. When equal to 1, Indicates the first [item] in the current attribute The value includes the number of positive records and the number of records in the current attribute. The value is the ratio between the number of records included in the range. When equal to 2, Indicates the first [item] in the current attribute The number of negative records included in the value and the number of records in the current attribute The value represents the ratio between the number of records included in the range.

[0032] In this embodiment, the attribute weight of each attribute in the set of test papers to be matched, the set of test questions to be matched, or the set of marking experts to be matched is calculated by computation, instead of obtaining the attribute weights one by one using a linear regression equation. This can improve computational efficiency and save computational resources, because the linear regression equation requires a large amount of data for training before obtaining the result, while direct calculation does not require data training, resulting in high computational efficiency and strong interpretability.

[0033] In some implementations, based on multiple first attributes, first attribute weights, multiple second attributes, second attribute weights, multiple third attributes, third attribute weights, multiple fourth attributes, and fourth attribute weights, the test questions and target experts are matched to obtain a set of successfully matched records, including: Obtain the weights of the first set of test papers to be matched, the weights of the second set of test questions to be matched, the weights of the third set of marking experts to be matched, and the weights of the fourth set of experts' historical marking records. Multiple first attributes are constructed into a first attribute matrix; multiple second attributes are constructed into a second attribute matrix; multiple third attributes are constructed into a third attribute matrix; and multiple fourth attributes are constructed into a fourth attribute matrix. Calculate the first bias based on the weight of the first attribute, calculate the second bias based on the weight of the second attribute, and calculate the third bias based on the weight of the third attribute. Input all historical data from the expert's historical grading records into the linear regression equation to obtain the fourth bias corresponding to all fourth attributes; Construct a first weight matrix based on the first attribute weight of each first attribute, construct a second weight matrix based on the second attribute weight of each second attribute, construct a third weight matrix based on the third attribute weight of each third attribute, and construct a fourth weight matrix based on the fourth attribute weight of each fourth attribute. Based on the weights of the first set, the weights of the second set, the weights of the third set, the weights of the fourth set, the weight matrix of the first set, the weight matrix of the second set, the weight matrix of the third set, the weight matrix of the fourth set, the attribute matrix of the first set, the attribute matrix of the second set, the attribute matrix of the third set, the attribute matrix of the fourth set, the bias of the first set, the bias of the second set, the bias of the third set, and the bias of the fourth set, the test questions and the target experts are matched to obtain the set of successfully matched records.

[0034] In this embodiment, the set weights, weight matrices, attribute matrices, and biases corresponding to the set of test papers to be matched, the set of test questions to be matched, the set of marking experts to be matched, and the set of historical marking records of experts are comprehensively considered to match the test questions with the target experts, which can improve the matching accuracy. Moreover, the matching is done by matching the test questions with the target experts, rather than using all experts for matching calculation, which can avoid excessive computation and improve matching efficiency.

[0035] In some implementations, based on the weights of the first set, the weights of the second set, the weights of the third set, the weights of the fourth set, the weight matrix of the first set, the weight matrix of the second set, the weight matrix of the third set, the weight matrix of the fourth set, the attribute matrix of the first set, the attribute matrix of the second set, the attribute matrix of the third set, the attribute matrix of the fourth set, the bias of the first set, the bias of the second set, the bias of the third set, and the bias of the fourth set, the test questions and experts are matched to obtain a set of successfully matched records, including: ; in, Indicates the degree of matching. Represents the normalized exponential function, Indicates the weight of the first set. Indicates the weight of the second set. Indicates the weight of the third set. Indicates the weight of the fourth set. This represents the first weight matrix. This represents the second weight matrix. This represents the third weight matrix. This represents the fourth weight matrix. Represents the first attribute matrix, This represents the second attribute matrix. Represents the third attribute matrix. This represents the fourth attribute matrix. Indicates the first bias. This indicates the second bias. Indicates the third bias. Indicates the fourth bias. Indicates matrix transpose; The set of successfully matched records corresponding to the matching degree is recorded. The attributes in the set of successfully matched records include the first attribute, the second attribute, and the third attribute.

[0036] In this embodiment, by comprehensively considering the set weights, weight matrices, attribute matrices, and biases corresponding to the set of test papers to be matched, the set of test questions to be matched, the set of marking experts to be matched, and the set of historical marking records of experts, the test questions and target experts can be matched, thereby improving the matching accuracy.

[0037] The above It is a linear regression equation, which uses the fourth attribute matrix. By inputting the equation into a linear regression model, a fourth bias can be trained. and the fourth weight matrix .

[0038] In some implementations, the matching effectiveness is evaluated on the set of successfully matched records to obtain evaluation results, including: Select multiple first target attributes from the set of successfully matched records, and select multiple second target attributes from the set of expert historical marking records, wherein the first target attributes and the second target attributes are the same; Based on the first target attribute and the second target attribute, calculate the vector distance between each record in the successfully matched record set and each record in the expert historical review record set to obtain the vector distance result set; Based on the vector distance result set, the matching effect of the successfully matched record set is evaluated, and the evaluation result is obtained.

[0039] In this embodiment, the vector distance between each record in the successfully matched record set and each record in the expert's historical marking record set is calculated based on the first target attribute and the second target attribute, resulting in a vector distance result set. Based on this vector distance result set, the matching effect of the successfully matched record set is evaluated, yielding an evaluation result. Thus, by evaluating the matching effect, a solid data foundation can be laid for improving the accuracy of matching between experts and the exam papers to be marked. Specifically, if the evaluation result shows that the matching effect of the successfully matched record set is good, it indicates that the experts in the successfully matched record set are matched relatively accurately with the exam papers to be marked.

[0040] In some implementations, the matching effect is evaluated on the set of successfully matched records based on the vector distance result set, and the evaluation result is obtained, including: Get the total number of questions to be matched in the question set; Sort all vector distance results in the vector distance result set to obtain the vector distance sorted result; Divide the vector distance sorting result into several equal parts and set the return value for each part; Based on the multiple vector distances between each record in the successfully matched record set and all records in the expert historical review record set, determine which target segment the minimum vector distance falls into among several segments, and return the return value corresponding to the target segment. The target segment is one of the several segments. Calculate the ratio between the return value corresponding to the target equal segment and the total number of questions to be matched in the question set, and obtain the ratio result; Based on the ratio results, the matching effect of the successfully matched record set is evaluated, and the evaluation results are obtained.

[0041] In this embodiment, by determining which target segment among several segments the minimum vector distance falls into based on multiple vector distances between each record in the successfully matched record set and all records in the expert's historical marking record set, the return value corresponding to the target segment is returned. The target segment is one of several segments. The ratio between the return value corresponding to the target segment and the total number of questions to be matched in the question set is calculated to obtain the ratio result. Based on the ratio result, the matching effect of the successfully matched record set is evaluated to obtain the evaluation result. In this way, the matching effect of the successfully matched record set can be better evaluated, which can lay a good data foundation for improving the accuracy of matching between experts and the questions to be marked.

[0042] In some implementations, after evaluating the matching effectiveness of the set of successfully matched records and obtaining the evaluation results, the method further includes: If the evaluation result is less than the preset threshold, a preset number of target experts will be selected from the set of matching marking experts to obtain new target experts; Based on multiple first attributes, first attribute weights, multiple second attributes, second attribute weights, multiple third attributes, third attribute weights, multiple fourth attributes, and fourth attribute weights, the test questions are matched with new target experts to obtain a new set of successfully matched records. The matching effect of the new set of successfully matched records is evaluated to obtain a new evaluation result. The new evaluation result is used as the target matching result until the new evaluation result is greater than or equal to the preset threshold.

[0043] In this embodiment, if the evaluation result is less than a preset threshold, it indicates that the matching between the experts in the successfully matched record set and the test papers to be graded is not very accurate. Therefore, a preset number of target experts are selected from the set of experts to be matched, and the test questions are matched with new target experts again based on multiple first attributes, first attribute weights, multiple second attributes, second attribute weights, multiple third attributes, third attribute weights, multiple fourth attributes, and fourth attribute weights to obtain a new set of successfully matched records. This process continues until the new evaluation result is greater than or equal to the preset threshold, which can lay a good data foundation for improving the accuracy of matching between experts and test papers to be graded.

[0044] To facilitate understanding by those skilled in the art, a set of preferred embodiments is provided below: Currently, universities use expert grading to assess student learning. The effectiveness of automated matching between experts and exam papers is a crucial issue. This is because experts have diverse research areas, and university courses offer a wide variety of exam papers. Manual allocation is not only inefficient but also yields unsatisfactory matching results. Existing automated methods for assigning grading experts fail to provide a reasonable match between experts and exam papers, resulting in suboptimal matching and reduced accuracy. Therefore, finding a reasonable way to match experts with exam papers has become an important research direction.

[0045] Existing university exam marking expert allocation methods typically assign experts per question per exam paper. This implementation plan primarily addresses the issue of assigning experts based on exam questions. Furthermore, this plan matches experts to specific exam papers, rather than assigning scores based on the difficulty level. This implementation plan considers three aspects: first, the experts' past marking experience and professional expertise; second, the course category, course classification, question type classification, and the exam's assessment intent; and third, it evaluates the effectiveness of the expert matching process. Based on these three aspects, it provides an intelligent automatic marking expert allocation method, resolving the difficulties and unsatisfactory results often encountered in universities' automatic expert matching systems.

[0046] This method offers several advantages: First, it automatically assigns experts to all exam papers based on the questions. Second, it intelligently considers experts' past grading experience through weighted analysis, comprehensively taking into account their professional expertise, course-related attributes, and exam paper-related attributes. Third, it scores each automatic expert assignment using parameterized evaluation criteria. Compared to existing automatic expert assignment methods, this approach is more intelligent, considering both expert-related attributes and course / exam paper attributes, while also providing evaluation criteria for expert assignment. This significantly improves the quality and accuracy of automatic expert assignment in universities, reduces staff workload, and increases user satisfaction.

[0047] Reference Figure 2 The technical solution of this embodiment specifically includes: Part 1: Parameter Settings. The set of exam papers to be matched, C1, contains multiple primary attributes, including course category, course classification, exam paper classification, and exam paper assessment intent. It also contains some other attributes (exam paper ID, total exam score, and exam difficulty rating). The number of records in C1 is greater than 0. The set of exam questions to be matched, C1_1, contains multiple secondary attributes, including question type classification and exam question assessment intent. It also contains some other attributes (exam paper ID, question ID, total question score, and question difficulty rating). The number of records in C1_1 is greater than 0. The set of marking experts to be matched, C2, contains multiple secondary attributes, including research direction and subject area. It also contains some other attributes (expert ID, evaluation standard score, and expert score). The number of records in C2 is greater than 0. The expert historical marking record set C2_1 contains multiple fourth attributes, namely course category, course classification, exam paper classification, exam paper examination intent, question type classification, question examination intent, research direction, and subject. It also contains some other attributes (exam paper ID, question ID, and expert ID). The number of records in C2_1 is greater than 0. The maximum number of explorations, N, has a default value of 2, and N is less than or equal to the total number of records in set C2. The evaluation criterion, P, has a default value of 85%. The lower bound for flexibility, P1, has a default value of 10%, and P - P1 > 0. The maximum number of evaluation iterations, T1, has a default value of 1. In this embodiment, all calculations involving decimals are rounded to one decimal place.

[0048] The attributes of each set (i.e., the set of test papers to be matched C1, the set of test questions to be matched C1_1, the set of marking experts to be matched C2, and the set of historical marking records of experts C2_1) are described as follows: The exam paper ID can be a unique identifier for the exam paper to be matched, and the unique identifier can be used to identify which exam paper it is.

[0049] The corresponding test paper ID can be the test paper ID to which the question to be matched belongs.

[0050] An expert ID can be a unique identifier for an expert, allowing identification of which expert it is.

[0051] The question ID can be a unique identifier for a question, and the question can be identified based on the unique identifier.

[0052] Test paper classification can be done by categorizing all test papers according to certain rules, such as classifying test papers into Chinese language, mathematics, etc.

[0053] Question type classification can be done by classifying all questions according to certain rules, such as classifying all questions into multiple choice questions, fill-in-the-blank questions, and true / false questions.

[0054] Research direction can be the academic field each expert studies, such as mathematics or physics. The affiliated discipline can be the specific subject area an expert researches.

[0055] The difficulty level score can be a rating of how difficult the test paper is.

[0056] The difficulty level score can be a rating of the difficulty of each question.

[0057] The evaluation standard score can be the total score corresponding to the evaluation of each expert's historical marking ability.

[0058] Expert ratings can be scores given to each expert on their historical ability to grade papers.

[0059] The collection of expert historical review records contains historical data.

[0060] The course categories can include five major categories: mathematics, Chinese, politics, English, and physics, or other categories. This embodiment does not make specific limitations on this.

[0061] Courses can be categorized into compulsory and elective courses. For example, mathematics, Chinese, and English can be classified as compulsory courses, while politics and physics can be classified as elective courses. Courses can also be categorized according to other categories. This embodiment does not impose any specific limitations.

[0062] The testing intent of an exam paper can include the testing intent of all questions within the exam paper.

[0063] The purpose of a test question may include assessing a student’s mastery of basic knowledge, logical thinking ability, problem analysis ability, and attention to detail. Different test questions may have different or the same purpose. A test question may have two or more purpose. This embodiment does not make specific limitations on this.

[0064] Part Two: Attribute Weight Calculation Model. To calculate the importance of each attribute in the set of test papers to be matched (C1), the set of test questions to be matched (C1_1), and the set of marking experts to be matched (C2), the information gain function is used to determine the importance of each attribute value in each set. The importance value ranges from 0 to 1, with values ​​closer to 1 indicating greater importance. The attribute weight calculation formula is as follows: ; in, The set of test papers to be matched (C1), the set of test questions to be matched (C1_1), and the set of examiners to be matched (C2) are all considered. C1 includes multiple first attributes (course category, course classification, test paper classification, and test paper assessment intent), C1_1 includes multiple second attributes (question type classification and test question assessment intent), and C2 includes multiple third attributes (research direction and affiliated discipline). This embodiment specifies that a default score for each attribute is positive when it is greater than or equal to 70% of the original value, and negative when it is less than 70% of the original value. The original value in C1 refers to the total score of the test paper, in C1_1 it refers to the total score of the test questions, and in C2 it refers to the evaluation standard score. The 70% threshold can be adjusted according to actual circumstances. For example: If the total score of a certain test paper in C1 is 100 points, then 100 × 70% = 70. A positive result is greater than or equal to 70 points, and a negative result is less than 70 points. When the score of the test paper to be matched is 71 points, it is considered a positive result. If the total score of the test paper in C1_1 is 5 points, then 5 × 70% = 3.5. A positive result is greater than or equal to 3.5 points, and a negative result is less than 3.5 points. When the score of the test paper is 3 points, it is considered a negative result. For calculation Baseline information gain. The maximum value is 2, which is based on The scoring attributes (i.e., the attributes in C1 (exam paper scoring), the attributes in C1_1 (exam question scoring), or the attributes in C2 (expert scoring)) are divided according to the positive and negative standards specified above. refer to One of the attributes, refer to The range of values ​​for each attribute. When When equal to 1, This refers to the number of positive records contained in the current attribute divided by the total number of records contained in the current attribute. This refers to the number of positive records contained in the j-th value of the current attribute / the total number of records contained in the j-th value of the current attribute; when When equal to 2, This refers to the number of negative records contained in the current attribute divided by the total number of records contained in the current attribute. This refers to the number of negative records contained in the j-th value of the current attribute divided by the total number of records contained in the j-th value of the current attribute. For example: Suppose C1 has a total of 5 test paper records, 3 positively graded test papers, and 2 negatively graded test papers, then... , Suppose that the course category dictionary for C1 contains 5 values, then The total number of courses is 5. The courses are categorized into five major categories: Mathematics, Chinese Language, Politics, English, and Physics. Let the value of j in the Mathematics category be 1, and the number of test records in the Mathematics category be 2, with 1 positive record and 1 negative record. Then... That is, the number of test paper records. That is, the total number of test paper records. That is, the number of positive records in the math exam / the total number of records in the math exam. That is, the number of negative records in the math exam / the total number of records in the math exam. The weight of each attribute in sets C1, C1_1, and C2 can be obtained using the above formula.

[0065] The third part is the intelligent expert allocation model. Loop through T1, and within T1, loop through C1, associating the attributes (exam paper ID) of C1_1 with the attributes (exam paper ID) of C1_1, and so on. When matching each question with an expert, to avoid excessive computation, an exploration count parameter N is used to control the number of attempts; that is, N target experts are randomly selected from C2, and then a user-defined function is used to... Calculate the matching score (i.e., matching degree) for each target expert, and select the target expert with the highest matching score as the matching result. If there are multiple target experts with the same highest matching score, then any one of them is randomly selected as the matching result. (Custom function) as follows: ; in, It is a normalized exponential function with a value range of [0-1]. The larger the value, the better the match. This indicates the matrix transpose. This represents the weight of the first set corresponding to C1. This represents the weight of the second set corresponding to C1_1. This indicates the weight of the third set corresponding to C2. This represents the weight of the fourth set corresponding to C2_1. , , and The sum of the weights is 1, and the default weight values ​​are 0.1, 0.2, 0.2, and 0.5. According to... The function calculates the weight of each attribute in C1, C1_1, and C2, and constructs the first weight matrix accordingly. Second weight matrix and the third weight matrix Based on the historical dataset corresponding to the fourth attribute in C2_1 (i.e., course category, course classification, exam paper classification, exam paper assessment intent, question type classification, exam question assessment intent, research direction, and subject), a linear regression equation (i.e., Training yields the fourth weight matrix. . Indicates the first bias. This indicates the second bias. Indicates the third bias. Indicates the fourth bias, where, , and The default values ​​are respectively from , and The average value of each weight in the weight matrix is ​​calculated; for example, by... The weights in the weight matrix are summed, and then the sum is divided by the number of weights in the weight matrix to obtain the average value. Based on the historical dataset in C2_1, a linear regression equation is used to train and obtain the bias. , This represents the attribute value matrix (i.e., the first attribute matrix) of the test paper set C1 to be matched. This represents the attribute value matrix (i.e., the second attribute matrix) of the set of test questions to be matched, C1_1. This represents the attribute value matrix (i.e., the third attribute matrix) of the expert set C2 to be matched. This represents the attribute value matrix (i.e., the fourth attribute matrix) of the expert historical review set C2_1.

[0066] At the end of each iteration of T1 (assuming T1 equals 3, it means T1 has three iterations), the matching degree F is obtained and the corresponding set of successfully matched records S is recorded. The attributes of S include (exam paper ID, question ID, expert ID, course category, course classification, exam paper classification, exam paper examination intent, question type classification, question examination intent, research direction, and subject). The set of successfully matched records S can be recorded by programming software or by techniques known to those skilled in the art. This embodiment does not specifically limit the method used.

[0067] Part Four: Effectiveness Evaluation Model. At the end of each iteration of T1, it is necessary to evaluate whether the matching effect has met the target. Based on the set of successfully matched records S obtained in Part Three, select the same attributes (course category, course classification, exam paper classification, exam paper examination intent, question type classification, question examination intent, research direction, and subject) for S and C2_1, that is, the same first target attribute (corresponding to S) and second target attribute (corresponding to C2_1). Standardize the data corresponding to the first and second target attributes in the sets S and C2_1 to obtain the following results: and The evaluation criterion function is defined as follows: , It requires a distance calculation function. The distance calculation function is as follows: ; in, This represents the total number of attributes in sets S and C2_1. refer to Each record contains One attribute, refer to Each record contains One attribute, Function calculation Each record and each item Vector distance between records express The first in each record One attribute, express The first in each record Each attribute. (Through) ,cycle Each record and Each record can be obtained The set of vector distance results H of the data volume refer to Total number of records in refer to The total number of records in set H. Sort the data in set H (i.e., the vector distance results) from largest to smallest to obtain the vector distance sorting result. Divide the vector distance sorting result into four interval sets H1, H2, H3, and H4, i.e., four equal parts. H1 contains the data in H [0-25%], H2 contains the data in H [25%-50%], H3 contains the data in H [50%-75%], and H4 contains the data in H [75%-100%]. The corresponding return values ​​for H1, H2, H3, and H4 are 0.25, 0.50, 0.75, and 1.0 respectively (if more granular management is needed, H can be divided into more intervals).

[0068] ; Where P refers to the evaluation standard and P1 refers to the lower limit of elasticity. This represents the total number of records (i.e., the total number of items) of the test questions to be matched. refer to and The minimum distance value (i.e., the minimum vector distance) calculated by set computation. This refers to the value returned when the minimum distance value falls within a certain interval set (i.e., the target equal division) of (H1, H2, H3, H4). For example, when the minimum distance value falls into H1, the returned value is 0.25; when it falls into H2, the returned value is 0.50; when it falls into H3, the returned value is 0.75; and when it falls into H4, the returned value is 1.0. For example: P=85%, elasticity lower limit=10%. When evaluating the S set obtained in the third part, since 85%-10%=75%, the evaluation standard is 75%; substituting... and to ,when If the automatic allocation is successful, the set of successfully matched records S is taken as the target matching result, and marking experts are automatically assigned based on the target matching result. If the target matching result is not successful, within the range of T1, loop C1 is re-executed. Based on the attribute (exam paper ID) of C1, the attribute (exam paper ID) of C1_1 is associated with it, and loop C1_1 is re-matched to obtain set S. R is then recalculated based on S until... When the automatic allocation is successful, the T1 loop is considered complete. Regarding T1, if the evaluation criteria are met at any point within the T1 value range, the T1 loop exits. If the evaluation criteria are still not met within the T1 value range, the user is prompted that the matching failed, and it is recommended that the user adjust the P or P1 value and attempt automatic matching again, or that manual allocation of marking experts be used.

[0069] Compared with the prior art, this embodiment has the following advantages: This embodiment's technical solution improves matching accuracy by objectively calculating the weights of each set, ensuring that important factors are fully considered during the prediction process. During the iterative cycle, users can customize the weights of each set and flexibly adjust the calculation strategy to ensure the matching effect remains within a controllable range. Finally, through quantitative evaluation criteria (i.e., the effect evaluation model in Part Four), the effectiveness of each automatic matching of marking experts can be scientifically demonstrated, providing a reliable basis for decision-making. Practical application in multiple universities has shown that, compared with traditional expert evaluation methods, this embodiment's method significantly improves matching efficiency and quality, effectively reducing the error rate of manual marking expert allocation. This innovation not only optimizes the expert allocation process but also greatly improves the satisfaction of automatically assigned marking experts and the overall matching quality, bringing substantial improvements to the field of automatic expert marking allocation.

[0070] Reference Figure 3 This application also provides an intelligent automatic allocation system for university marking experts. The system includes a data acquisition unit 301, a first calculation unit 302, a weight acquisition unit 303, an expert selection unit 304, an expert matching unit 305, and an expert allocation unit 306, wherein: The data acquisition unit 301 is used to acquire multiple first attributes from the set of test papers to be matched, multiple second attributes from the set of test questions to be matched, multiple third attributes from the set of marking experts to be matched, and multiple fourth attributes from the set of historical marking records of experts. The first calculation unit 302 is used to calculate the first attribute weight of each first attribute, the second attribute weight of each second attribute, and the third attribute weight of each third attribute. The weight acquisition unit 303 is used to input the historical data corresponding to each fourth attribute in the expert historical marking record set into the linear regression equation to obtain the fourth attribute weight of each fourth attribute. The expert selection unit 304 is used to select a preset number of target experts from the set of experts to be matched for marking. The expert matching unit 305 is used to match test questions with target experts based on multiple first attributes, first attribute weights, multiple second attributes, second attribute weights, multiple third attributes, third attribute weights, multiple fourth attributes, and fourth attribute weights, to obtain a set of successfully matched records. The expert allocation unit 306 is used to evaluate the matching effect of the successfully matched record set and obtain the evaluation result. If the evaluation result is greater than or equal to the preset threshold, the successfully matched record set is taken as the target matching result, and the marking experts are assigned according to the target matching result.

[0071] It should be noted that since the intelligent automatic allocation system for university marking experts in this embodiment is based on the same inventive concept as the intelligent automatic allocation method for university marking experts described above, the corresponding content in the method embodiment is also applicable to this system embodiment, and will not be described in detail here.

[0072] Reference Figure 4 This application also provides an electronic device, which includes: At least one memory; At least one processor; At least one program; The program is stored in memory, and the processor executes at least one program to implement the above-described intelligent automatic allocation method for university grading experts.

[0073] This electronic device can be any smart terminal, including mobile phones, tablets, personal digital assistants (PDAs), and in-vehicle computers.

[0074] The electronic devices according to embodiments of this application will now be described in detail.

[0075] The processor 1600 can be implemented using a general-purpose central processing unit (CPU), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this disclosure. The memory 1700 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 1700 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1700 and is called and executed by the processor 1600 to execute the intelligent automatic allocation method for university exam marking experts according to the embodiments of this disclosure.

[0076] The input / output interface 1800 is used to implement information input and output. The communication interface 1900 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 2000 transmits information between various components of the device (e.g., processor 1600, memory 1700, input / output interface 1800, and communication interface 1900); The processor 1600, memory 1700, input / output interface 1800 and communication interface 1900 are connected to each other within the device via bus 2000.

[0077] This disclosure also provides a storage medium, which is a computer-readable storage medium storing computer-executable instructions for causing a computer to execute the above-described intelligent automatic allocation method for university grading experts.

[0078] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0079] The embodiments described in this disclosure are for the purpose of more clearly illustrating the technical solutions of this disclosure and do not constitute a limitation on the technical solutions provided by this disclosure. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by this disclosure are also applicable to similar technical problems.

[0080] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this disclosure, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0081] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0082] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or appropriate combinations thereof.

[0083] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0084] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0085] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0086] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0087] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0088] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause an electronic device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks. The embodiments of this application have been described in detail above with reference to the accompanying drawings, but this application is not limited to the above embodiments. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of this application.

[0089] The embodiments of this application have been described in detail above with reference to the accompanying drawings. However, this application is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of this application.

Claims

1. A method for intelligent automatic allocation of university exam marking experts, characterized in that, The method includes: Retrieve multiple first attributes from the set of test papers to be matched, multiple second attributes from the set of test questions to be matched, multiple third attributes from the set of marking experts to be matched, and multiple fourth attributes from the set of historical marking records of experts; Calculate the first attribute weight for each of the first attributes, calculate the second attribute weight for each of the second attributes, and calculate the third attribute weight for each of the third attributes; The historical data corresponding to each of the fourth attributes in the expert historical marking record set is input into the linear regression equation to obtain the fourth attribute weight of each of the fourth attributes; Select a preset number of target experts from the set of experts to be matched; Based on the plurality of first attributes, the weight of the first attributes, the plurality of second attributes, the weight of the second attributes, the plurality of third attributes, the weight of the third attributes, the plurality of fourth attributes, and the weight of the fourth attributes, the test questions and the target experts are matched to obtain a set of successfully matched records; The matching effect of the successfully matched record set is evaluated to obtain the evaluation result. If the evaluation result is greater than or equal to a preset threshold, the successfully matched record set is taken as the target matching result, and marking experts are assigned according to the target matching result.

2. The intelligent automatic allocation method for university exam marking experts according to claim 1, characterized in that, The calculation of the first attribute weight for each of the first attributes, the calculation of the second attribute weight for each of the second attributes, and the calculation of the third attribute weight for each of the third attributes include: ； in, This represents the set of test papers to be matched, the set of test questions to be matched, or the set of marking experts to be matched. This represents the first attribute in the set of test papers to be matched, the second attribute in the set of test questions to be matched, or the third attribute in the set of examiners to be matched. Indicates the weight of the first attribute, the weight of the second attribute, or the weight of the third attribute. When equal to 1, This represents the ratio between the number of positive records contained in the current attribute and the total number of records contained in the current attribute. When equal to 2, This represents the ratio between the number of negative records in the current attribute and the total number of records in the current attribute. Indicates the first [item] in the current attribute The value includes the number of all records. This indicates the total number of records contained in the current attribute. Indicates the range of values ​​that the current attribute can take. When equal to 1, Indicates the first in the current attribute The value includes the number of positive records and the number of records in the current attribute. The value is the ratio between the number of records included in the range. When equal to 2, Indicates the first [item] in the current attribute The number of negative records included in the value and the number of records in the current attribute The value represents the ratio between the number of records included in the range.

3. The intelligent automatic allocation method for university exam marking experts according to claim 1, characterized in that, The method involves matching test questions with target experts based on the plurality of first attributes, the weights of the first attributes, the plurality of second attributes, the weights of the second attributes, the plurality of third attributes, the weights of the third attributes, the plurality of fourth attributes, and the weights of the fourth attributes, to obtain a set of successfully matched records, including: Obtain the weights of the first set of test papers to be matched, the weights of the second set of test questions to be matched, the weights of the third set of marking experts to be matched, and the weights of the fourth set of expert historical marking records. The plurality of first attributes are constructed into a first attribute matrix, the plurality of second attributes are constructed into a second attribute matrix, the plurality of third attributes are constructed into a third attribute matrix, and the plurality of fourth attributes are constructed into a fourth attribute matrix; A first bias is calculated based on the first attribute weight, a second bias is calculated based on the second attribute weight, and a third bias is calculated based on the third attribute weight. Input all historical data from the expert historical grading record set into the linear regression equation to obtain the fourth bias corresponding to all fourth attributes; A first weight matrix is ​​constructed based on the first attribute weight of each first attribute, a second weight matrix is ​​constructed based on the second attribute weight of each second attribute, a third weight matrix is ​​constructed based on the third attribute weight of each third attribute, and a fourth weight matrix is ​​constructed based on the fourth attribute weight of each fourth attribute. Based on the first set weights, the second set weights, the third set weights, the fourth set weights, the first weight matrix, the second weight matrix, the third weight matrix, the fourth weight matrix, the first attribute matrix, the second attribute matrix, the third attribute matrix, the fourth attribute matrix, the first bias, the second bias, the third bias, and the fourth bias, the test questions and the target experts are matched to obtain a set of successfully matched records.

4. The intelligent automatic allocation method for university exam marking experts according to claim 3, characterized in that, The process involves matching test questions and experts based on the weights of the first set, the second set, the third set, the fourth set, the first weight matrix, the second weight matrix, the third weight matrix, the fourth weight matrix, the first attribute matrix, the second attribute matrix, the third attribute matrix, the fourth attribute matrix, the first bias, the second bias, the third bias, and the fourth bias, resulting in a set of successfully matched records, including: ； in, Indicates the degree of matching. Represents the normalized exponential function, Indicates the weight of the first set. Indicates the weight of the second set. Indicates the weight of the third set. Indicates the weight of the fourth set. This represents the first weight matrix. This represents the second weight matrix. This represents the third weight matrix. This represents the fourth weight matrix. Represents the first attribute matrix, This represents the second attribute matrix. Represents the third attribute matrix. This represents the fourth attribute matrix. Indicates the first bias. This indicates the second bias. Indicates the third bias. Indicates the fourth bias. Indicates matrix transpose; Record a set of successfully matched records corresponding to the matching degree, wherein the attributes in the set of successfully matched records include a first attribute, a second attribute, and a third attribute.

5. The intelligent automatic allocation method for university exam marking experts according to claim 1, characterized in that, The evaluation of the matching effect of the successfully matched record set, to obtain the evaluation result, includes: Select multiple first target attributes from the set of successfully matched records, and select multiple second target attributes from the set of expert historical marking records, wherein the first target attributes and the second target attributes are the same; Based on the first target attribute and the second target attribute, calculate the vector distance between each record in the successfully matched record set and each record in the expert historical grading record set to obtain a vector distance result set; Based on the vector distance result set, the matching effect of the successfully matched record set is evaluated to obtain the evaluation result.

6. The intelligent automatic allocation method for university exam marking experts according to claim 5, characterized in that, The step of evaluating the matching effect of the successfully matched record set based on the vector distance result set to obtain the evaluation result includes: Get the total number of questions to be matched in the question set; Sort all vector distance results in the vector distance result set to obtain the vector distance sorting result; Divide the vector distance sorting result into several equal parts and set the return value corresponding to each equal part; Based on the multiple vector distances between each record in the set of successfully matched records and all records in the set of expert historical review records, determine which target segment among the multiple vector distances falls into, and return the return value corresponding to the target segment, wherein the target segment is one of the multiple segments; Calculate the ratio between the return value corresponding to the target equal segment and the total number of questions to be matched in the question set, and obtain the ratio result; Based on the ratio result, the matching effect of the successfully matched record set is evaluated to obtain the evaluation result.

7. The intelligent automatic allocation method for university exam marking experts according to claim 1, characterized in that, After evaluating the matching effect of the successfully matched record set and obtaining the evaluation result, the method further includes: If the evaluation result is less than the preset threshold, then a preset number of target experts are selected from the set of matching marking experts to obtain new target experts; Based on the plurality of first attributes, the weight of the first attributes, the plurality of second attributes, the weight of the second attributes, the plurality of third attributes, the weight of the third attributes, the plurality of fourth attributes, and the weight of the fourth attributes, the test questions are matched with the new target experts to obtain a new set of successfully matched records; The matching effect of the new set of successfully matched records is evaluated to obtain a new evaluation result. The new set of successfully matched records is taken as the target matching result until the new evaluation result is greater than or equal to a preset threshold.

8. A smart automatic allocation system for university exam marking experts, characterized in that, The system includes: The data acquisition unit is used to acquire multiple first attributes from the set of test papers to be matched, multiple second attributes from the set of test questions to be matched, multiple third attributes from the set of marking experts to be matched, and multiple fourth attributes from the set of historical marking records of experts. The first calculation unit is used to calculate the first attribute weight of each first attribute, the second attribute weight of each second attribute, and the third attribute weight of each third attribute. The weighting unit is used to input the historical data corresponding to each of the fourth attributes in the expert historical marking record set into the linear regression equation to obtain the fourth attribute weight of each of the fourth attributes; The expert selection unit is used to select a preset number of target experts from the set of experts to be matched for marking. An expert matching unit is used to match test questions with target experts based on the plurality of first attributes, the weight of the first attributes, the plurality of second attributes, the weight of the second attributes, the plurality of third attributes, the weight of the third attributes, the plurality of fourth attributes, and the weight of the fourth attributes, to obtain a set of successfully matched records; An expert allocation unit is used to evaluate the matching effect of the set of successfully matched records and obtain an evaluation result. If the evaluation result is greater than or equal to a preset threshold, the set of successfully matched records is taken as the target matching result, and marking experts are assigned according to the target matching result.

9. An electronic device, characterized in that, It includes at least one control processor and a memory for communicatively connecting to the at least one control processor; the memory stores instructions executable by the at least one control processor, which, when executed by the at least one control processor, enable the at least one control processor to perform the intelligent automatic allocation method for university grading experts as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions for causing a computer to execute the intelligent automatic allocation method for university exam marking experts as described in any one of claims 1 to 7.

Images