College student information quality intelligent evaluation system and method

Abstract

The invention relates to the technical field of intelligent evaluation, and particularly discloses an intelligent evaluation system and method for college student information quality, and the system comprises an evaluation behavior collection module which collects dominant operation data and implicit operation data of each student when the evaluation is completed; the dominant operation analysis module analyzes interdisciplinary coverage data, a tool combination application chain and a knowledge migration logic chain of each student based on the dominant operation data; the digital-intelligent characteristic evaluation module evaluates the digital-intelligent characteristic ability evaluation result of each student; the implicit operation analysis module evaluates a basic operation capability evaluation result of each student; a multi-dimensional result output module carries out subject comparative analysis and digital intelligence capability short plate positioning on the digital intelligence characteristic capability evaluation result and the basic operation capability evaluation result of each student and outputs the results in a thermodynamic diagram form to obtain an information quality evaluation thermodynamic diagram of all students; the information quality level of students can be comprehensively known, and a powerful basis is provided for targeted teaching and personalized cultivation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent assessment, in particular to a college student information literacy intelligent assessment system and method. BACKGROUND

[0002] In today's information explosion era, information literacy has become one of the key abilities of college students. It not only affects the development of students in the academic field, but also plays a decisive role in students' future entry into society and adaptation to the rapidly changing information environment. Students with good information literacy can more efficiently acquire, analyze and utilize information, thereby occupying an advantage in learning, research and future career development. With the popularization of higher education and the change of education concept, colleges and universities pay more and more attention to the cultivation of students' comprehensive ability, and information literacy education has become an important part of the higher education system. In order to accurately measure the information literacy level of students and provide basis for teaching improvement and individualized development of students, it is crucial to develop a scientific, comprehensive and intelligent college student information literacy assessment system and method. With the rapid development of artificial intelligence technology, its application in the field of education assessment has broad prospects. Intelligent assessment system can use advanced data analysis algorithms to mine the information hidden behind students' operation data, realize accurate assessment of students' information literacy, and promote the intelligent and scientific development of education assessment.

[0003] However, the existing college student information literacy assessment technology cannot deeply understand the comprehensive performance of students in the information processing process, and thus it is difficult to evaluate students by using a comprehensive information literacy integration assessment task, resulting in inaccurate evaluation results and not truly reflecting the actual ability level of students. The existing technology cannot output the evaluation results in the form of a direct heat map, making the evaluation results not clear enough, which is not conducive for teachers and students to obtain effective information from the results.

[0004] Therefore, the present application proposes a college student information literacy intelligent assessment system and method. SUMMARY

[0005] This invention provides an intelligent assessment system and method for information literacy among university students. By integrating information literacy assessment tasks, it simultaneously collects explicit and implicit operational data from students during assessment, comprehensively acquiring behavioral information about students throughout the assessment process. Based on explicit operational data, it derives interdisciplinary coverage data, tool combination application chains, and knowledge transfer logic chains, deeply analyzing students' performance in knowledge application and logical thinking. By assessing students' unique mathematical and intellectual abilities based on these chains, it can accurately identify students' strengths in the field of digital intelligence. Combining explicit analysis results with implicit operational data to assess basic operational abilities improves the evaluation of students' fundamental skills. The assessment results of digital and intellectual characteristics and basic operational abilities are analyzed comparatively across disciplines to pinpoint weaknesses in mathematical and intellectual abilities, and output in the form of heatmaps, visually presenting the information literacy assessment results of all students. This facilitates universities' comprehensive understanding of students' information literacy levels and provides a strong basis for targeted teaching and personalized training. This assessment system can deeply analyze various operational data of students in integrated information literacy assessment tasks, evaluating their information literacy from multiple dimensions. This helps university teachers comprehensively understand students' strengths and weaknesses, thereby adjusting teaching strategies, optimizing curriculum design, and improving teaching quality. At the same time, for students themselves, the assessment results help them clearly understand their own information literacy status, identify areas for improvement, and target their efforts to enhance their abilities.

[0006] This invention provides an intelligent assessment system for information literacy of college students, comprising:

[0007] The assessment behavior collection module is used to assess each student by integrating assessment tasks with each student's information literacy, and to collect explicit and implicit operational data of each student when completing the assessment.

[0008] The explicit action analysis module is used to analyze each student's cross-disciplinary coverage data, tool combination application chain, and knowledge transfer logic chain based on the explicit action data of each student when completing the assessment.

[0009] The Digital Intelligence Feature Assessment Module is used to evaluate each student's digital intelligence feature abilities based on their interdisciplinary coverage data, tool combination application chain, and knowledge transfer logic chain.

[0010] The implicit operation analysis module is used to evaluate each student's basic operational ability based on each student's interdisciplinary coverage data, tool combination application chain, knowledge transfer logic chain, and implicit operation data when completing the assessment.

[0011] The multi-dimensional results output module is used to conduct subject-specific comparative analysis of the assessment results of each student's mathematical intelligence characteristic ability and basic operation ability, and to locate the shortcomings of mathematical intelligence ability. The results are output in the form of a heat map, which yields a heat map of the information literacy assessment of all students.

[0012] Preferably, the evaluation behavior collection module comprises:

[0013] The evaluation question bank construction submodule is configured to select multiple interdisciplinary nodes from the professional background data of each student and generate an information quality integration evaluation task for each student.

[0014] The evaluation behavior collection submodule is configured to evaluate the corresponding student based on the information quality integration evaluation task, synchronously collect direct input operation data of each student in completing the evaluation as explicit operation data through screen recording data, operation log records of the participating device, and eye tracking data of the participating object, and perform in-depth reasoning on the direct input operation data of each student in completing the evaluation to obtain implicit operation data of each student in completing the evaluation.

[0015] Preferably, the explicit operation analysis module comprises:

[0016] The interdisciplinary coverage data extraction submodule is configured to analyze database access records, literature source identifiers, and keyword cross-domain distribution characteristics in the explicit operation data of each student in completing the evaluation, count the types of interdisciplinary resources called, the number of interdisciplinary term co-occurrences, and take the ratio of the total number of interdisciplinary resource types to the total number of resource types as the interdisciplinary resource coverage rate, and take the ratio of the number of interdisciplinary term co-occurrences to the total number of terms as the knowledge node association density, and take the interdisciplinary resource coverage rate and the knowledge node association density of each student as the interdisciplinary coverage data of each student.

[0017] The tool combination application chain restoration submodule is configured to identify the tool combination type based on the tool startup log, the function call sequence, and the data flow path in the explicit operation data, mark the operation time length proportion and the parameter adjustment times of each tool to restore the tool use logic, and obtain the tool combination application chain.

[0018] The knowledge migration logic chain analysis submodule is configured to extract the discipline attribution of core arguments and supporting arguments and the argumentation logic path in the analysis report text, argumentation process records, and conclusion derivation steps in the explicit operation data, comb the migration path from the discipline attribution of core arguments and supporting arguments to the target task scene based on a logic relationship identification algorithm, merge the argumentation logic path and all migration paths to obtain the knowledge migration logic chain.

[0019] Preferably, the digital feature evaluation module comprises:

[0020] The interdisciplinary integration ability scoring submodule is configured to evaluate the interdisciplinary coverage data of each student from two dimensions of interdisciplinary breadth and interdisciplinary depth, and then perform weighted operation to obtain the interdisciplinary integration ability score of each student.

[0021] The intelligent tool collaboration ability scoring submodule is configured to score the tool combination application chain of each student from three dimensions of tool adaptability, operation collaboration, and combination application innovation, and then perform weighted operation to obtain the intelligent tool collaboration ability score of each student.

[0022] The knowledge transfer innovation ability scoring submodule is configured to score the knowledge transfer logic chain of each student from three dimensions of logical coherence, domain adaptability, and value conversion degree, and then perform weighted operation to obtain the knowledge transfer innovation ability score of each student.

[0023] The digital characteristic ability comprehensive evaluation submodule is configured to perform weighted operation on the cross-disciplinary integration ability score, the intelligent tool collaboration ability score, and the knowledge transfer innovation ability score of each student to obtain the digital characteristic ability evaluation result of each student.

[0024] Preferably, the cross-disciplinary integration ability scoring submodule comprises:

[0025] The cross-disciplinary breadth scoring unit is configured to obtain the cross-disciplinary breadth score based on the cross-disciplinary resource coverage rate in the cross-disciplinary coverage data of each student.

[0026] The cross-disciplinary depth scoring unit is configured to take the matching score of the knowledge node association density in the cross-disciplinary coverage data of each student and the cross-disciplinary association standard under the target task scenario as the cross-disciplinary depth score.

[0027] The cross-disciplinary integration ability scoring unit is configured to perform weighted operation on the cross-disciplinary breadth score and the cross-disciplinary depth score of each student to obtain the cross-disciplinary integration ability score of each student.

[0028] Preferably, the intelligent tool collaboration ability scoring submodule comprises:

[0029] The tool adaptability analysis unit is configured to analyze the matching degree of all application tools in the tool combination application chain of each student with the target task scenario as the tool adaptability.

[0030] The operation collaboration analysis unit is configured to score the data flow conversion efficiency between tools in the tool combination application chain of each student to obtain the operation collaboration degree.

[0031] The application innovation degree analysis unit is configured to statistically obtain the proportion of unconventional tool combinations in the tool combination application chain of each student based on the conventional tool combination library as the combination application innovation degree.

[0032] The intelligent tool collaboration ability scoring unit is configured to perform weighted operation on the tool adaptability, the operation collaboration degree, and the combination application innovation degree of each student to obtain the intelligent tool collaboration ability score of each student.

[0033] Preferably, the knowledge transfer innovation ability scoring sub-module comprises:

[0034] a logical coherence scoring unit configured to obtain a logical coherence score of each student based on a number of logical discontinuities in the logical chain of knowledge transfer of the student;

[0035] a domain adaptability scoring unit configured to obtain a domain adaptability score of each student based on an association strength between the logical chain of knowledge transfer of the student and the target task scenario;

[0036] a value conversion degree scoring unit configured to obtain a value conversion degree score of each student based on a contribution degree of a conclusion formed by the logical chain of knowledge transfer of the student to a solution of the target task;

[0037] a knowledge transfer innovation ability scoring unit configured to obtain a knowledge transfer innovation ability score of each student by performing a weighted operation on the logical coherence score, the domain adaptability score, and the value conversion degree score of the student.

[0038] Preferably, the implicit operation analysis module comprises:

[0039] a step division sub-module configured to determine all key operation steps and invalid operation steps based on the cross-disciplinary coverage data, the tool combination application chain, and the logical chain of knowledge transfer of each student;

[0040] an implicit operation analysis sub-module configured to obtain, as an operation hesitation degree, a proportion of steps in all key operation steps whose pause duration exceeds a preset pause duration, and obtain, as an operation redundancy degree, a ratio of the total number of invalid operation steps to the total number of operation steps, and obtain, as a tool dependency degree, an average proportion of the number of tool prompt invocations;

[0041] a basic ability evaluation sub-module configured to obtain a basic operation ability evaluation result of each student by performing a weighted operation on the operation hesitation degree, the operation redundancy degree, and the tool dependency degree of the student.

[0042] Preferably, the multi-dimensional result output module comprises:

[0043] a disciplinary deviation analysis sub-module configured to take, as a disciplinary deviation of a corresponding ability type of a corresponding student, a ratio of a difference between the corresponding ability type evaluation result of the student and an average score of the corresponding ability type of the corresponding professional background to a standard deviation of the corresponding ability type of the corresponding professional background;

[0044] a digital capability positioning sub-module configured to filter out a short board ability type and / or an advantage ability type of each student based on the disciplinary deviation of the digital characteristic ability type and the disciplinary deviation of the basic operation ability type of the student;

[0045] The heat map generation submodule is configured to visualize the subject deviation degrees of the digital characteristic ability types and the subject deviation degrees of the basic operation ability types of all students in the form of a two-dimensional matrix heat map, and mark based on the short board ability types and / or the advantage ability types of all students, to obtain the information quality evaluation heat map of all students.

[0046] The present application provides a college student information quality intelligent evaluation method, comprising:

[0047] Each student is evaluated by using the information quality integrated evaluation task of each student, and the explicit operation data and implicit operation data of each student during the evaluation are synchronously collected;

[0048] Based on the explicit operation data of each student during the evaluation, the cross-disciplinary coverage data, tool combination application chain, and knowledge transfer logic chain of each student are analyzed;

[0049] Based on the cross-disciplinary coverage data, tool combination application chain, and knowledge transfer logic chain of each student, the digital characteristic ability evaluation result of each student is evaluated;

[0050] Based on the cross-disciplinary coverage data, tool combination application chain, and knowledge transfer logic chain of each student, and the implicit operation data during the evaluation, the basic operation ability evaluation result of each student is evaluated;

[0051] The digital characteristic ability evaluation result and the basic operation ability evaluation result of each student are subjected to subject comparison analysis and digital ability short board positioning, and are output in the form of a heat map, to obtain the information quality evaluation heat map of all students.

[0052] The beneficial effects generated by the present application relative to the prior art are: by integrating the information quality evaluation task, the explicit and implicit operation data of the students during the completion of the evaluation are synchronously collected, and the behavior information of the students in the evaluation process is comprehensively obtained. Based on the explicit operation data, the cross-disciplinary coverage data, the tool combination application chain and the knowledge transfer logical chain are derived, and the performance of the students in the knowledge application and thinking logic is deeply analyzed. According to these chains, the numerical and intelligent characteristic ability of the students is evaluated, and the strengths of the students in the numerical and intelligent field can be accurately positioned. Combined with the explicit analysis result and the implicit operation data, the basic operation ability is evaluated, and the basic ability of the students is improved. The evaluation results of the numerical and intelligent characteristics and the basic operation ability are compared and analyzed in the disciplines and the numerical and intelligent ability short board is positioned, and the information quality evaluation situation of all the students is output in the form of a heat map, which is convenient for the colleges and universities to comprehensively understand the information quality level of the students, and provides a strong basis for targeted teaching and individualized cultivation. The evaluation system can evaluate the information quality of the students from multiple dimensions by deeply analyzing various operation data of the students in the information quality integrated evaluation task, which helps the teachers of colleges and universities to comprehensively understand the advantages and disadvantages of the students, and then adjust the teaching strategy, optimize the curriculum setting and improve the teaching quality. At the same time, for the students themselves, the evaluation results can help them clearly understand their information quality situation, clearly understand the direction of effort, and improve their ability in a targeted manner.

[0053] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application. The objects and other advantages of the present application can be realized and attained by the structure particularly pointed out in the written description and claims hereof.

[0054] The technical solutions of the present application will be further described in detail below by means of the drawings and examples. BRIEF DESCRIPTION OF DRAWINGS

[0055] The accompanying drawings are included to provide a further understanding of the present application, and constitute a part of the specification, together with the embodiments of the present application, to explain the present application, and do not constitute a limitation on the present application. In the drawings:

[0056] Figure 1 It is a schematic diagram of the information quality intelligent evaluation system for college students in the embodiments of the present application;

[0057] Figure 2 It is a schematic diagram of the explicit operation analysis module in the embodiments of the present application;

[0058] Figure 3 It is a schematic diagram of the numerical and intelligent characteristic evaluation module in the embodiments of the present application;

[0059] Figure 4 It is a schematic diagram of the implicit operation analysis module in the embodiments of the present application. DETAILED DESCRIPTION

[0060] The preferred embodiments of the present application are described below in conjunction with the accompanying drawings, in which it is understood that the preferred embodiments described herein are merely intended to illustrate and explain the present application, and are not intended to limit the present application.

[0061] As shown in Figure 1 The present application provides an implementation of an intelligent evaluation system for college students' information literacy, comprising:

[0062] An evaluation behavior collection module is configured to evaluate each student by using an information literacy integrated evaluation task of each student, and to synchronously collect explicit operation data and implicit operation data of each student when completing the evaluation;

[0063] An explicit operation analysis module is configured to analyze interdisciplinary coverage data, tool combination application chains, and knowledge transfer logical chains of each student based on the explicit operation data of each student when completing the evaluation;

[0064] A digital intelligence feature evaluation module is configured to evaluate a digital intelligence feature ability evaluation result of each student based on the interdisciplinary coverage data, tool combination application chains, and knowledge transfer logical chains of each student;

[0065] An implicit operation analysis module is configured to evaluate a basic operation ability evaluation result of each student based on the interdisciplinary coverage data, tool combination application chains, knowledge transfer logical chains, and implicit operation data of each student when completing the evaluation;

[0066] A multi-dimensional result output module is configured to perform subject comparison analysis and digital intelligence short board positioning on the digital intelligence feature ability evaluation result and the basic operation ability evaluation result of each student, and to output the same in the form of a heat map to obtain an information literacy evaluation heat map of all students.

[0067] In this embodiment, the information literacy integrated evaluation task is generated based on a plurality of interdisciplinary nodes selected according to the professional background data of the students. These tasks are used to comprehensively examine the students' abilities in multidisciplinary knowledge application, information processing tool use, and knowledge interdisciplinary transfer, such as asking engineering students to complete relevant information processing tasks based on the interdisciplinary node of "computer science and materials science".

[0068] In this embodiment, the tool combination application chain is a chain that shows the tool usage and logical relationship of the students.

[0069] In this embodiment, the knowledge transfer logical chain is a logical thought that presents the interdisciplinary transfer of the students' knowledge.

[0070] In this embodiment, the information literacy evaluation heat map of all students: adopt a two-dimensional matrix heat map, the horizontal axis is the subdivision index, the vertical axis is the ability level (excellent, good, standard, to be improved), the color depth represents the subject deviation degree, the specific short board description is marked in the corresponding cell of the short board ability type, the ability radar chart on the right side of the heat map shows the score distribution profile of the six indexes, and the average level profile of the professional is marked with a dashed line for comparison.

[0071] In order to comprehensively obtain the evaluation behavior information of students, an evaluation behavior collection module is proposed, which includes:

[0072] The evaluation question bank construction submodule is used to select multiple interdisciplinary nodes in the professional background data of each student, and generate an information literacy integrated evaluation task for each student.

[0073] The evaluation behavior collection submodule is used to evaluate the corresponding students based on the information literacy integrated evaluation task, and simultaneously collect the direct input operation data of each student in completing the evaluation as explicit operation data through screen recording data, operation log records of the participating device, and eye tracking data of the participating object. In addition, the direct input operation data of each student in completing the evaluation is further inferred to obtain the implicit operation data of each student in completing the evaluation.

[0074] In this embodiment, professional background data refers to a series of information related to the student's major, such as the discipline, curriculum, research direction, etc.

[0075] In this embodiment, the interdisciplinary node is selected from the professional background data, which connects different knowledge fields of different disciplines, such as the intersection of computer science and sociology, which provides a cross-disciplinary knowledge combination point for the evaluation task.

[0076] In this embodiment, multiple interdisciplinary nodes are selected from the professional background data of each student to generate an information literacy integrated evaluation task for each student.

[0077] A multi-dimensional (n-dimensional) subject feature vector space is constructed, with dimensions covering data structure complexity, ethical risk coefficient, model abstraction degree, and other core features of multiple disciplines (by analyzing the keyword distribution and weight in the subject paper, these features are converted into standardized values between 0 and 1 to determine the value), and each dimension is standardized to a value between 0 and 1.

[0078] A feature vector is generated for each subject node, and each dimension value of the vector is determined by the following method: subject modeling is performed on the top journal papers of the subject in the past 5 years, the keyword distribution of multiple feature dimensions is extracted, the weight of each dimension keyword is calculated and normalized (for example, the algorithm complexity dimension value of computer science is 0.87, and the ethical risk coefficient dimension value is 0.32).

[0079] Build an n x n interdisciplinary correlation matrix, where each element represents the correlation strength between two disciplines. Calculate the correlation strength by taking the inner product of the corresponding discipline feature vectors and dividing by the product of the vector magnitudes (range -1 to 1, positive values indicate synergistic correlation, negative values indicate conflicting correlation).

[0080] Map student professional background data into an n-dimensional personal feature vector: extract professional ability dimension scores from course grades (e.g., 90 points in the "Artificial Intelligence" course corresponds to a "model abstraction degree" dimension of 0.9), and extract digital tool application proficiency from research project experience (e.g., using Python to complete a project corresponds to a "tool adaptability" dimension of 0.75).

[0081] Calculate a 1 x n professional matching degree matrix, where each element is the inner product of the personal feature vector and the corresponding discipline feature vector (representing the matching degree).

[0082] Matrix operation logic for interdisciplinary node screening:

[0083] Initial selection: retain 3 core discipline nodes with a matching degree ≥ a preset matching degree threshold (e.g., 0.6);

[0084] Correlation calculation: build a correlation sub-matrix for these 3 nodes, and extract synergistic correlation pairs with a correlation strength ≥ a first preset correlation strength threshold (e.g., 0.5) (e.g., the correlation pair between computer science and statistics);

[0085] Conflict detection: exclude conflicting correlation pairs with a correlation strength ≤ a second preset correlation strength threshold (e.g., -0.3) (e.g., a strong conflict between pure mathematics and applied law);

[0086] Final selection: select 2 optimal interdisciplinary nodes from the correlation sub-matrix through the maximum weight path algorithm, requiring a path weight sum ≥ 1.2. That is, from the 3 core discipline nodes, select the two nodes with the highest correlation strength as interdisciplinary nodes.

[0087] Generate a 5-dimensional task basis vector for the selected interdisciplinary nodes, including 5 complexity dimensions: interdisciplinary parameter coupling degree, tool chain collaboration difficulty, data interference strength, AI result discrimination requirement, and ethical decision weight.

[0088] Generate personalized task vectors by linear transformation of the 5x5 student ability short board matrix (specifically: multiply the 5x5 student ability short board matrix with the 5-dimensional task basis vector to make the complexity dimensions of the task (such as AI result discrimination requirements, cross-disciplinary parameter coupling degree, etc.) proportional to the student's short board ability (such as weak AI discrimination ability, increase the complexity of the corresponding dimension), finally generate personalized task vectors that fit the student's ability short board. Each element in the matrix represents the mapping coefficient of the basic complexity to the short board ability (such as "AI discrimination ability" as a short board, the corresponding dimension mapping coefficient is set to 1.5), and the task vector is the product of the basis vector and the short board matrix.

[0089] Generate an n x m task parameter matrix (n is the number of task steps 5-8, m is the parameter dimension of each step): use Hadamard product to calculate parameter correlation (i.e. multiply corresponding elements), reduce dimension by singular value decomposition, and retain the parameter combination corresponding to the first 3 singular values to balance the independence and correlation of parameters.

[0090] Construct a k x 5 historical task-ability correlation matrix (k is the number of historical tasks) to record the discrimination of each task on the 5 numerical intelligence abilities.

[0091] Calculate the cosine similarity matrix of the new task vector and the historical task matrix. If the maximum similarity is ≤0.7 (to ensure novelty), and the L2 norm of the task vector is between 1.5-3.0 (to ensure moderate difficulty), it is determined as an effective task; otherwise, regenerate by adjusting the short board matrix.

[0092] Through multiple mathematical operations such as eigenvector space modeling, correlation matrix operation, and task vector transformation, quantitative decision-making of cross-disciplinary node selection and accurate generation of assessment tasks are realized, avoiding subjective experience-dominated function stacking, and improving the technical depth and complexity of the system.

[0093] In this embodiment, the participating equipment: the hardware used by students to participate in the evaluation, such as computers, tablets, used to record student evaluation operation behavior.

[0094] In this embodiment, the direct input operation data of each student when completing the evaluation is synchronously collected through screen recording data, operation log recording of the participating equipment, and eye tracking data of the participating object: the screen recording function of the participating equipment is used to record interface operation changes, operation log recording software tools are used to record operation steps, and eye tracking is used to record line of sight movement, etc. At the same time, the direct input operation data of the student during the evaluation is obtained.

[0095] In this embodiment, the direct input operation data of each student during the completion of the test is deeply inferred to obtain the implicit operation data of each student during the completion of the test: according to the collected direct input operation data, the implicit information such as operation thinking, knowledge mastery is inferred from operation pause, repetition times, dependence on prompts and the like.

[0096] As shown in Figure 2 To deeply analyze the explicit operation behavior of students, an explicit operation analysis module is proposed, including:

[0097] The interdisciplinary coverage data extraction submodule is used to analyze the database access records, literature source identification and keyword cross-domain distribution characteristics in the explicit operation data of each student during the completion of the test, to count the types of interdisciplinary resources called, the number of interdisciplinary term co-occurrences, and to take the ratio of the total number of interdisciplinary resource types to the total number of resource types as the interdisciplinary resource coverage rate, and to take the ratio of the number of interdisciplinary term co-occurrences to the total number of terms as the knowledge node association density, and to take the interdisciplinary resource coverage rate and the knowledge node association density of each student as the interdisciplinary coverage data of each student.

[0098] The tool combination application chain restoration module is used to identify the tool combination type based on the tool startup log, function call sequence and data flow path in the explicit operation data, and to mark the operation time length proportion and parameter adjustment times of each tool to restore the tool use logic, and to obtain the tool combination application chain.

[0099] The knowledge transfer logic chain analysis submodule is used to extract the discipline attribution of core arguments and supporting arguments and the argumentation logic path in the analysis report text, argumentation process record and conclusion derivation steps in the explicit operation data, and to comb the transfer path from the discipline attribution of core arguments and supporting arguments to the target task scene based on a logic relationship identification algorithm, and to merge the argumentation logic path and all transfer paths to obtain the knowledge transfer logic chain.

[0100] In this embodiment, the database access record refers to the record generated by the student during the completion of the test task, including the database name, access time, specific content and the like, which can reflect the source database of the student's information acquisition.

[0101] In this embodiment, the literature source identification is an identification of the literature source referenced or referred to by the student during the test, including the author of the literature, the published journal or book, the publication year and the like, which is used to clarify the origin of the literature used by the student.

[0102] In this embodiment, the keyword cross-domain distribution feature refers to the distribution characteristics of the keywords used by the student in completing the evaluation task in different subject domains, such as the number, proportion, and association between keywords in different subjects, which can reflect the student's integration and application of multidisciplinary knowledge.

[0103] In this embodiment, by analyzing the database access records, literature source identification, and keyword cross-domain distribution features in the explicit operation data of each student during the evaluation, the types of cross-disciplinary resources called and the number of cross-disciplinary term co-occurrences are counted: the database access records, literature source identification, and keyword cross-domain distribution features in the explicit operation data of the student during the evaluation are analyzed and interpreted. From the database access records and literature source identification, it is determined which disciplines the resources called come from, so as to count the types of cross-disciplinary resources; by analyzing the keyword cross-domain distribution feature, the number of times different subject keywords appear at the same time, i.e., the number of cross-disciplinary term co-occurrences, is determined.

[0104] In this embodiment, the tool startup log records the relevant information of the student in starting various tool software during the evaluation, such as the tool name, startup time, etc., which is used to understand the order of the student's use of tools.

[0105] In this embodiment, the function call sequence records the order of the student's call of various functions during the use of tools, such as in the data analysis tool, the data filtering function is called first, and then the data sorting function is called, etc., which reflects the specific process of the student's use of tool functions.

[0106] In this embodiment, the data flow path refers to the path of data transmission from one tool or link to another tool or link in the evaluation task, such as from the data collection tool to the data processing tool, and then to the data visualization tool, which shows the flow of data between different tools.

[0107] In this embodiment, the tool combination type is identified based on the tool startup log, function call sequence, and data flow path in the explicit operation data: the tool startup order is understood from the tool startup log in the explicit operation data of the student during the evaluation, the tool function use process is known from the function call sequence, and the data transmission direction is clear from the data flow path, and these information is integrated to determine the type of the tool combination used by the student, such as "data collection tool-data processing tool-data visualization tool" combination.

[0108] In this embodiment, the operation time length proportion of each tool and the parameter adjustment times are marked to restore the tool usage sequence, and the tool combination application chain is obtained: the proportion of the usage time length of each tool in the entire evaluation task operation process to the total operation time length is recorded, and the number of parameter adjustments during tool usage is recorded. Through these data and tool startup logs, function call sequences, and other information, the usage sequence of the tools is sorted out, and then the tool combination application chain is formed, and the specific situation and logical relationship of the students using the tools in the evaluation task are presented completely.

[0109] In this embodiment, the analysis report text is a text written by the student after completing the evaluation task, which analyzes the relevant problems, and contains information such as the elaboration of the problem, the analysis process, and the conclusions drawn.

[0110] In this embodiment, the argumentation process record records the argumentation steps and methods performed by the student to support his / her point of view during the analysis process.

[0111] In this embodiment, the conclusion derivation step refers to the specific steps of the student deriving the final conclusion from the known conditions and analysis process.

[0112] In this embodiment, the discipline attribution of the core argument and supporting argument is that in the student's analysis report, argumentation process, etc., the core argument is the main point of view, and the supporting argument is the basis for supporting the point of view. The discipline attribution of the core argument and supporting argument is to clarify which discipline field these arguments and arguments come from, for example, the core argument comes from physics, and the supporting argument quotes data from the mathematics discipline.

[0113] In this embodiment, the argumentation logic path refers to the logical thinking and context followed by the student in the argumentation process, from presenting the argument to using the argument to reasoning, and finally reaching a conclusion, which reflects the rationality and coherence of the argumentation process.

[0114] In this embodiment, the discipline attribution of the core argument and supporting argument and the argumentation logic path are extracted from the analysis report text, argumentation process record, and conclusion derivation step in the explicit operation data: the analysis report text, argumentation process record, and conclusion derivation step in the explicit operation data of the student evaluation are extracted. From these contents, find out the core argument and supporting argument, and determine their discipline field, and sort out the logical path followed by the entire argumentation process.

[0115] In this embodiment, the logical relationship recognition algorithm is used to identify the migration path from the discipline attribution field of the core argument and supporting argument to the target task scene: a special logical relationship recognition algorithm is used to analyze the discipline attribution information of the extracted core argument and supporting argument, and the path of how these discipline knowledge is applied from the original discipline field to the current target task scene is combed, for example, how to migrate the principles of physics to the scene of solving practical engineering problems.

[0116] In this embodiment, the argument logical path and all migration paths are merged to obtain the knowledge migration logical chain: the argument logical path and all migration paths from the discipline attribution field of the core argument and supporting argument to the target task scene are integrated together to form the knowledge migration logical chain, which fully presents the logical thinking and process of the student's cross-disciplinary migration in the knowledge application process.

[0117] As shown in Figure 3 To comprehensively evaluate the digital intelligence characteristic ability of students, a digital intelligence characteristic evaluation module is proposed, which includes:

[0118] The cross-disciplinary integration ability scoring sub-module is used to evaluate the cross-disciplinary coverage data of each student from the two dimensions of cross-disciplinary breadth and cross-disciplinary depth, and then weighted operation is performed to obtain the cross-disciplinary integration ability score of each student.

[0119] The intelligent tool collaboration ability scoring sub-module is used to evaluate the tool combination application chain of each student from the three dimensions of tool adaptability, operation collaboration, and combination application innovation, and then weighted operation is performed to obtain the intelligent tool collaboration ability score of each student.

[0120] The knowledge migration innovation ability scoring sub-module is used to evaluate the knowledge migration logical chain of each student from the three dimensions of logical coherence, field adaptability, and value conversion degree, and then weighted operation is performed to obtain the knowledge migration innovation ability score of each student.

[0121] The digital intelligence characteristic ability comprehensive evaluation sub-module is used to perform weighted operation on the cross-disciplinary integration ability score, intelligent tool collaboration ability score, and knowledge migration innovation ability score of each student to obtain the digital intelligence characteristic ability evaluation result of each student.

[0122] In this embodiment, the cross-disciplinary integration ability score, intelligent tool collaboration ability score, and knowledge migration innovation ability score of each student are weighted and operated to obtain the digital intelligence characteristic ability evaluation result of each student: according to the pre-set weight, i.e., the cross-disciplinary integration ability score accounts for 30%, the intelligent tool collaboration ability score accounts for 40%, and the knowledge migration innovation ability score accounts for 30%, weighted calculation is performed to obtain the digital intelligence characteristic ability evaluation result of each student.

[0123] To accurately assess the students' interdisciplinary integration ability, an interdisciplinary integration ability scoring sub-module is proposed, including:

[0124] an interdisciplinary breadth scoring unit for obtaining an interdisciplinary breadth score based on an interdisciplinary resource coverage rate in the interdisciplinary coverage data of each student;

[0125] an interdisciplinary depth scoring unit for taking a matching score of a knowledge node association density in the interdisciplinary coverage data of each student and an interdisciplinary association standard under a target task scenario as an interdisciplinary depth score;

[0126] an interdisciplinary integration ability scoring unit for performing a weighted operation on the interdisciplinary breadth score and the interdisciplinary depth score of each student to obtain an interdisciplinary integration ability score of each student.

[0127] In this embodiment, the interdisciplinary breadth score obtained based on the interdisciplinary resource coverage rate in the interdisciplinary coverage data of each student is determined according to a preset correspondence relationship, different interdisciplinary resource coverage rate ranges correspond to different score intervals, and thus the interdisciplinary breadth score is obtained. For example, when the interdisciplinary resource coverage rate is ≥ 60%, 80-100 points are corresponded; 40%-60% corresponds to 60-79 points; and < 40% corresponds to < 60 points. In this way, the performance of students in the interdisciplinary breadth is quantified by the interdisciplinary resource coverage rate, and is presented in the form of scores.

[0128] In this embodiment, the interdisciplinary association standard under the target task scenario: for a specific information literacy integration assessment task scenario, a standard for measuring the reasonable association degree between knowledge nodes of different disciplines is preset. It is a reference value for judging whether the close degree of association between different disciplines of knowledge exhibited by students in the task completion process meets the requirements. For example, for a target task scenario involving the interdisciplinary of “computer science and sociology”, it is possible to set the association proportion between certain concepts of the computer discipline and the related concepts of sociology to 35% as the standard.

[0129] In this embodiment, the matching score of the knowledge node association density in the interdisciplinary coverage data of each student and the interdisciplinary association standard under the target task scenario is taken as the interdisciplinary depth score: the knowledge node association density, i.e., the ratio of the number of interdisciplinary term co-occurrences to the total number of terms, is obtained from the interdisciplinary coverage data of each student. The knowledge node association density is compared with the interdisciplinary association standard under the target task scenario to calculate the matching score. The calculation method is to divide the knowledge node association density by the interdisciplinary association standard under the target task scenario to obtain the matching score (the upper limit is 1).

[0130] In this embodiment, the interdisciplinary breadth score and the interdisciplinary depth score of each student are weighted to obtain the interdisciplinary integration ability score of each student: the scores are calculated according to the established weight ratio. Among them, the interdisciplinary breadth score accounts for 60%, and the interdisciplinary depth score accounts for 40%.

[0131] In order to effectively evaluate the intelligent tool collaboration ability of students, an intelligent tool collaboration ability scoring sub-module is proposed, which includes:

[0132] A tool adaptability analysis unit is configured to analyze the matching degree of all application tools in the tool combination application chain of each student with the target task scene as the tool adaptability;

[0133] An operation collaboration analysis unit is configured to score the data flow efficiency between tools in the tool combination application chain of each student to obtain the operation collaboration degree;

[0134] An application innovation degree analysis unit is configured to statistically obtain the proportion of unconventional tool combinations in the tool combination application chain of each student based on the conventional tool combination library as the combination application innovation degree;

[0135] An intelligent tool collaboration ability scoring unit is configured to weight the tool adaptability, operation collaboration degree, and combination application innovation degree of each student to obtain the intelligent tool collaboration ability score of each student.

[0136] In this embodiment, the matching degree of all application tools in the tool combination application chain of each student with the target task scene is analyzed: for each tool used by each student in the tool combination application chain, whether the tool is suitable for completing the target task scene and the degree of suitability are considered. For example, in a data mining target task scene, the Python data mining tool used by the student is analyzed to determine whether its functions, features, and other characteristics match the requirements of the data mining task, such as whether it has appropriate data processing algorithms and can process data in the corresponding format, to determine the matching of the tool and the target task scene.

[0137] In this embodiment, the tool adaptability is the result of the above analysis of the matching degree of all application tools in the tool combination application chain of each student with the target task scene, which is represented by a quantitative value. This value represents the degree of adaptation of the tool to the task scene. For example, in a data mining task scene, the matching degree of the tool combination Python+Tableau is 90%, which is the tool adaptability of the tool combination. The higher the value, the more suitable the tool is for the target task scene.

[0138] In this embodiment, the operation coordination degree is a quantitative index score of the data flow efficiency between tools in the tool combination application chain, and is used to intuitively reflect the coordination effect between tools in terms of data flow. As described above, the operation coordination degree is obtained by scoring the data flow efficiency according to a specific rule. A higher operation coordination degree means that the data flow between tools is efficient and smooth, which helps students to complete tasks more efficiently.

[0139] In this embodiment, the operation coordination degree is a quantitative index score of the data flow efficiency between tools in the tool combination application chain, and is used to intuitively reflect the coordination effect between tools in terms of data flow. As described above, the operation coordination degree is obtained by scoring the data flow efficiency according to a specific rule. A higher operation coordination degree means that the data flow between tools is efficient and smooth, which helps students to complete tasks more efficiently.

[0140] In this embodiment, the conventional tool combination library is a database that pre-establishes and stores common tool combination modes. This database contains tool combination modes that are frequently used by college students in various subject fields or various task scenarios. For example, in the field of data analysis, common tool combinations such as Excel+SPSS, and in the field of document processing, Word+PowerPoint, are included in the conventional tool combination library as a reference basis for determining whether a tool combination is conventional.

[0141] In this embodiment, the proportion of unconventional tool combinations in the tool combination application chain of each student is calculated based on the conventional tool combination library: the tool combination application chain of each student is compared with the combination modes in the conventional tool combination library, the number of tool combinations that do not belong to the combination modes recorded in the conventional tool combination library is counted, and then the proportion of the number of unconventional tool combinations to the total number of tool combinations in the tool combination application chain of the student is calculated. For example, if there are 5 tool combination modes in the tool combination application chain of a student, and 2 of them do not appear in the conventional tool combination library, then the proportion of unconventional tool combinations is 2÷5×100% = 40%.

[0142] In this embodiment, the combination application innovation degree is the proportion of unconventional tool combinations in the tool combination application chain of each student, which is calculated based on the conventional tool combination library. This proportion is used to measure the innovation degree of students in tool combination application. The higher the proportion, the more unconventional tool combinations the student uses, and the stronger the innovation in tool combination application.

[0143] In this embodiment, the tool adaptation degree, operation coordination degree, and combined application innovation degree of each student are weighted to obtain the intelligent tool coordination ability score of each student: the score is calculated according to preset weights, wherein the tool adaptation degree weight accounts for 50%, the operation coordination degree weight accounts for 30%, and the combined application innovation degree weight accounts for 20%.

[0144] In order to accurately evaluate the knowledge transfer innovation ability of students, a knowledge transfer innovation ability scoring sub-module is proposed, which includes:

[0145] A logical coherence score unit is configured to obtain a logical coherence score based on the number of logical discontinuities in the knowledge transfer logical chain of each student.

[0146] A domain adaptation score unit is configured to obtain a domain adaptation score based on the association strength between the knowledge transfer logical chain of each student and the target task scene.

[0147] A value conversion degree score unit is configured to obtain a value conversion degree score based on the contribution of the conclusion formed by the knowledge transfer logical chain of each student to the solution of the target task.

[0148] A knowledge transfer innovation ability score unit is configured to obtain a knowledge transfer innovation ability score of each student by weighting the logical coherence score, the domain adaptation score, and the value conversion degree score of each student.

[0149] In this embodiment, the logical coherence score is obtained based on the number of logical discontinuities in the knowledge transfer logical chain of each student: in the knowledge transfer logical chain of the student, the number of logical discontinuities refers to the number of logical discontinuities and reasoning interruptions in the process from the core argument to the supporting argument and then to the conclusion derivation. According to the pre-set scoring rules, 15 points are deducted for each occurrence of logical discontinuity. Taking a full score of 100 points as an example, assuming that the number of logical discontinuities in the knowledge transfer logical chain of a student is 2, then the logical coherence score is 100-15x2=70 points. In this way, the logical coherence score is obtained according to the number of logical discontinuities, which measures the degree of logical coherence in the knowledge transfer process of the student.

[0150] In this embodiment, the domain adaptation score is obtained based on the association strength between the knowledge transfer logical chain of each student and the target task scene: the correlation degree, i.e. the association strength, between the knowledge domain involved in the knowledge transfer logical chain of each student and the target task scene is analyzed. For example, if the "blockchain technology" is transferred to the supply chain management scene, the association strength may be evaluated as 85%, and if it is transferred to the literary analysis scene, the association strength may be only 30%. According to this association strength, the score is calculated according to the preset scoring standard, and the higher the association strength, the higher the score, thereby obtaining the domain adaptation score, which reflects the adaptation of the knowledge transferred by the student to the target task scene.

[0151] In this embodiment, the contribution degree of the conclusion formed based on the knowledge transfer logical chain of each student to the solution of the target task obtains a value conversion degree score: for the conclusion finally formed by the knowledge transfer logical chain of each student, the size of the role played in solving the target task, i.e. the solution contribution degree, is judged. If the conclusion obtained by the student through knowledge transfer can put forward new ideas or optimization schemes for the target task, the contribution degree is higher, and 80-100 points can be obtained; if it is only to repeat the conclusions in the original field, the contribution to the solution of the target task is small, and the score is less than 50 points. The value conversion degree score is obtained according to the judgment of the solution contribution degree to evaluate the actual value of the conclusion generated by the knowledge transfer of the student.

[0152] In this embodiment, the logical coherence score, the field adaptability score, and the value conversion degree score of each student are weighted and operated to obtain the knowledge transfer innovation ability score of each student: the weighted calculation is performed according to the weight ratio of 4:3:3, i.e.

[0153] Knowledge transfer innovation ability score = logical coherence score x 40% + field adaptability score

[0154] x 30% + value conversion degree score x 30%.

[0155] As shown in Figure 4 In order to comprehensively evaluate the basic operation ability of the student, an implicit operation analysis module is proposed, which includes:

[0156] A step division sub-module is used to determine all key operation steps and invalid operation steps based on the cross-disciplinary coverage data, tool combination application chain, and knowledge transfer logical chain of each student.

[0157] An implicit operation analysis sub-module is used to calculate the proportion of steps with a stop time longer than a preset stop time in all key operation steps as operation hesitation, and the ratio of all invalid operation steps to the total number of operation steps as operation redundancy, and the average proportion of tool prompt call times as tool dependence.

[0158] A basic ability evaluation sub-module is used to perform weighted operation on the operation hesitation, operation redundancy, and tool dependence of each student to obtain a basic operation ability evaluation result.

[0159] In this embodiment, all key operation steps and invalid operation steps are determined based on the interdisciplinary coverage data of each student, the tool combination application chain, and the knowledge transfer logical chain. By analyzing the interdisciplinary coverage data of each student, the acquisition and use of multidisciplinary resources are understood, and it is determined which operations are key to integrating interdisciplinary knowledge and which are invalid. According to the tool combination application chain, the operation steps that play a key role in completing the task and the invalid operation steps that have no practical significance are determined. In combination with the knowledge transfer logical chain, the key operation steps in the process of knowledge transfer from one field to the target task scenario and the opposite invalid steps are sorted out. For example, in the interdisciplinary coverage data, if accessing a key database is an important step in obtaining information, then the operation step of accessing the database is a key operation step. In the tool combination application chain, if unnecessary data format conversion operations are repeated multiple times and have no substantial help for the final task, it is an invalid operation step.

[0160] In this embodiment, the step ratio of key operation steps with a pause duration exceeding a preset pause duration is calculated based on the implicit operation data of each student. The implicit operation data contains the pause information of the student during the operation process. For all the key operation steps that have been determined, the number of steps with a pause duration exceeding the preset duration is counted, and then the ratio of these steps in all key operation steps is calculated. For example, the preset pause duration is 10 seconds, a student has 20 key operation steps, and 5 steps have a pause duration exceeding 10 seconds. The step ratio of steps with a pause duration exceeding the preset pause duration is 5 ÷ 20 × 100% = 25%. This ratio reflects the degree of hesitation of the student in the key operation steps, and is called operation hesitation.

[0161] In this embodiment, the average proportion of tool tip invocation times is: During the student's completion of the evaluation task, the implicit operation data records the number of times the student invokes the tool tip. The average proportion of tool tip invocation times is the average value obtained by dividing the total number of times the student invokes the tool tip during the entire operation process by the total number of operation steps. For example, a student performs a total of 100 operations, and the total number of times the tool tip is invoked is 20. The average proportion of tool tip invocation times is 20 ÷ 100 × 100% = 20%. This proportion reflects the degree of dependence of the student on the tool tip, i.e., the tool dependence.

[0162] In this embodiment, the operation hesitancy, operation redundancy and tool dependency of each student are weighted to obtain the basic operation ability evaluation result: assuming that the weights of operation hesitancy, operation redundancy and tool dependency are 40%, 40% and 20% respectively, each index is mapped to 0-100 points according to the quantitative value of the implicit operation characteristics (for example, if the operation hesitancy is 0, the corresponding item is 100 points, and for every 10% increase in hesitancy, 10 points are deducted; if the operation redundancy is 0, the corresponding item is 100 points, and for every 10% increase in redundancy, 10 points are deducted; if the tool dependency is 0, the corresponding item is 100 points, and for every 10% increase in dependency, 10 points are deducted). Taking operation hesitancy of 20%, operation redundancy of 15% and tool dependency of 10% as an example, the basic operation ability evaluation result = (100-2x10) x 40% + (100-1.5x10) x 40% + (100-1x10) x 20% = 80x40% + 85x40% + 90x20% = 32 + 34 + 18 = 84 points. The final score is combined with the grade threshold (≥ 80 points is "skilled", 60-79 points is "basically mastered", and < 60 points is "to be strengthened") to generate the basic operation ability evaluation result, which comprehensively evaluates the basic operation ability of the student.

[0163] In order to intuitively present the information quality evaluation result of the student, a multi-dimensional result output module is proposed, which includes:

[0164] A subject deviation analysis submodule is used to take the difference between the number intelligence characteristic ability evaluation result or the basic operation ability evaluation result of each student and the average score of the corresponding ability type of the corresponding professional background and the ratio of the standard deviation of the corresponding ability type of the corresponding professional background as the subject deviation of the corresponding ability type of the corresponding student.

[0165] A number intelligence positioning submodule is used to filter out the short board ability type and / or advantage ability type of each student based on the subject deviation of the number intelligence characteristic ability type and the subject deviation of the basic operation ability type of each student.

[0166] A heat map generation submodule is used to visualize the subject deviation of the number intelligence characteristic ability type and the subject deviation of the basic operation ability type of all students in the form of a two-dimensional matrix heat map, and mark based on the short board ability type and / or advantage ability type of all students to obtain the information quality evaluation heat map of all students.

[0167] In this embodiment, the difference between each student's digital characteristic ability assessment result or basic operation ability assessment result and the average score of the corresponding ability type of the corresponding professional background, divided by the standard deviation of the corresponding ability type of the corresponding professional background: for each student, subtract the average score of the corresponding ability type of the corresponding professional background from the student's digital characteristic ability assessment result (or basic operation ability assessment result), and divide the difference by the standard deviation of the corresponding ability type of the corresponding professional background. For example, a student majoring in engineering has a digital characteristic ability assessment score of 80, the average score of the digital characteristic ability of the major is 70, and the standard deviation is 5. Then the calculation result of the student's digital characteristic ability is (80-70) ÷ 5 = 2. This calculation result is used to measure the deviation of the student from the average level of students in the same major in this ability type, which is called subject deviation. A positive number indicates that the student is better than the average level in this aspect, and a negative number indicates that the student is lower than the average level.

[0168] In this embodiment, the short board ability type and / or the advantage ability type of each student are screened based on the subject deviation of the digital characteristic ability type and the subject deviation of the basic operation ability type of each student: after calculating the subject deviation of the digital characteristic ability type and the subject deviation of the basic operation ability type of each student in the previous step, the set standard is used for screening. For example, it is stipulated that the ability type with a subject deviation ≥ 1.2 and a score ≥ excellent threshold (for example, the excellent threshold of the digital characteristic ability of the major is 85) is the advantage ability type; the ability type with a subject deviation ≤ -0.8 or a score ≤ short board warning line (for example, the short board warning line of the digital characteristic ability of the major is 65) is the short board ability type. Accordingly, the short board ability type and / or the advantage ability type of each student in the digital characteristic ability and the basic operation ability can be determined.

[0169] In this embodiment, the discipline deviation degrees of the digital characteristic ability types and the discipline deviation degrees of the basic operation ability types of all students are visualized in the form of a two-dimensional matrix heat map, and are visualized and marked based on the short board ability types and / or the advantage ability types of all students to obtain the information quality evaluation heat map of all students: taking the two-dimensional matrix heat map as the display mode, the horizontal axis is set as the subdivision index, i.e., the digital characteristic ability and the basic operation ability; the vertical axis is the ability level, including excellent, good, up to standard, and to be improved. The discipline deviation degree of each student on each ability type is represented by color depth. For example, red represents significantly better than the average level (greater discipline deviation degree and positive), blue represents significantly lower than the average level (smaller discipline deviation degree and negative), and white represents close to the average level (close to 0 discipline deviation degree). At the same time, the cells corresponding to the short board ability types and / or the advantage ability types of the students in the heat map are visualized and marked, such as marking the specific short board description, such as "cross-disciplinary integration ability-knowledge node association density deficiency", in the cell corresponding to the short board ability type. In addition, an "ability radar chart" is added on the right side of the heat map to intuitively display the score distribution profile of the six indexes, and the professional average level profile is marked with a dashed line for comparison, and finally the information quality evaluation heat map of all students is formed to intuitively and comprehensively display the information quality evaluation situation of the students.

[0170] The present application provides an implementation of a college student information quality intelligent evaluation method, comprising:

[0171] Each student is evaluated by using the information quality integration evaluation task of each student, and the explicit operation data and implicit operation data of each student when completing the evaluation are synchronously collected;

[0172] Based on the explicit operation data of each student when completing the evaluation, the cross-disciplinary coverage data, tool combination application chain, and knowledge transfer logic chain of each student are analyzed;

[0173] Based on the cross-disciplinary coverage data, tool combination application chain, and knowledge transfer logic chain of each student, the digital characteristic ability evaluation result of each student is evaluated;

[0174] Based on the cross-disciplinary coverage data, tool combination application chain, and knowledge transfer logic chain of each student and the implicit operation data when completing the evaluation, the basic operation ability evaluation result of each student is evaluated;

[0175] The digital characteristic ability evaluation result and the basic operation ability evaluation result of each student are subjected to discipline comparison analysis and digital ability short board positioning and are output in the form of a heat map to obtain the information quality evaluation heat map of all students.

[0176] It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the spirit or scope of the application. Thus, it is intended that the present application cover modifications and variations of this application provided they come within the scope of the application and their equivalent technology.

Claims

1. A smart assessment system for information literacy of college students, characterized in that, include: The assessment behavior collection module is used to integrate assessment tasks based on each student's information literacy to assess each student, and simultaneously collects explicit and implicit operational data of each student when completing the assessment. The explicit action analysis module is used to analyze each student's cross-disciplinary coverage data, tool combination application chain, and knowledge transfer logic chain based on the explicit action data of each student when completing the assessment. The Digital Intelligence Feature Assessment Module is used to evaluate each student's digital intelligence feature abilities based on their interdisciplinary coverage data, tool combination application chain, and knowledge transfer logic chain. The implicit operation analysis module is used to evaluate each student's basic operational ability based on each student's interdisciplinary coverage data, tool combination application chain, knowledge transfer logic chain, and implicit operation data when completing the assessment. The multi-dimensional results output module is used to conduct subject-specific comparative analysis of the assessment results of each student's mathematical intelligence characteristic ability and basic operation ability, and to locate the shortcomings of mathematical intelligence ability. The results are output in the form of a heat map, which yields a heat map of the information literacy assessment of all students.

2. The intelligent assessment system for information literacy of college students according to claim 1, characterized in that, The assessment behavior collection module includes: The assessment question bank construction submodule is used to select multiple interdisciplinary nodes from each student's professional background data to generate an integrated assessment task for each student's information literacy. The assessment behavior collection submodule is used to assess corresponding students based on information literacy integration assessment tasks. At the same time, it collects each student's direct input operation data when completing the assessment as explicit operation data through screen recording data of the participating devices, operation log records, and eye-tracking data of the participants. It also performs in-depth reasoning on each student's direct input operation data when completing the assessment to obtain each student's implicit operation data when completing the assessment.

3. The intelligent assessment system for information literacy of college students according to claim 1, characterized in that, The explicit operation analysis module includes: The interdisciplinary coverage data extraction submodule is used to analyze the database access records, literature source identifiers, and keyword cross-domain distribution characteristics in the explicit operational data of each student when completing the assessment. It then counts the types of interdisciplinary resources accessed and the number of times interdisciplinary terms are co-occurred. The ratio of the total number of interdisciplinary resource types to the total number of resource types is taken as the interdisciplinary resource coverage rate, and the ratio of the number of times interdisciplinary terms are co-occurred to the total number of terms is taken as the knowledge node association density. The interdisciplinary resource coverage rate and knowledge node association density of each student are taken as the interdisciplinary coverage data of each student. The tool combination application chain reconstruction submodule is used to identify the tool combination type based on the tool startup log, function call sequence and data flow path in explicit operation data, and to mark the operation time ratio and parameter adjustment number of each tool to reconstruct the logical sequence of tool use and obtain the tool combination application chain. The knowledge transfer logic chain parsing submodule is used to extract the subject affiliation of core arguments and supporting evidence, as well as the logical path of argumentation, from the analysis report text, argumentation process records, and conclusion derivation steps in explicit operational data. Based on the logical relationship recognition algorithm, it sorts out the migration path from the subject affiliation of core arguments and supporting evidence to the target task scenario, and merges the logical path of argumentation with all migration paths to obtain the knowledge transfer logic chain.

4. The intelligent assessment system for information literacy of college students according to claim 1, characterized in that, The digital intelligence-focused assessment module includes: The interdisciplinary integration ability scoring submodule is used to evaluate each student's interdisciplinary coverage data from two dimensions: interdisciplinary breadth and interdisciplinary depth, and then perform weighted calculations to obtain each student's interdisciplinary integration ability score. The intelligent tool collaboration ability scoring submodule is used to evaluate each student's tool combination application chain from three dimensions: tool adaptability, operational collaboration, and combination application innovation, and then perform weighted calculations to obtain each student's intelligent tool collaboration ability score. The knowledge transfer and innovation ability scoring submodule is used to evaluate each student's knowledge transfer logic chain from three dimensions: logical coherence, domain adaptability, and value conversion degree, and then perform weighted calculations to obtain each student's knowledge transfer and innovation ability score. The comprehensive assessment submodule for digital intelligence-specific abilities is used to perform weighted calculations on each student's scores in interdisciplinary integration ability, intelligent tool collaboration ability, and knowledge transfer and innovation ability to obtain the assessment results for each student's digital intelligence-specific abilities.

5. The intelligent assessment system for information literacy of college students according to claim 4, characterized in that, The interdisciplinary integration ability assessment submodule includes: The interdisciplinary breadth scoring unit is used to obtain an interdisciplinary breadth score based on the interdisciplinary resource coverage rate in each student's interdisciplinary coverage data; The interdisciplinary depth scoring unit is used to determine the interdisciplinary depth score based on the matching score between the knowledge node association density in each student's interdisciplinary coverage data and the interdisciplinary association standard in the target task scenario. The interdisciplinary integration ability scoring unit is used to calculate each student's interdisciplinary integration ability score by weighting their interdisciplinary breadth and interdisciplinary depth scores.

6. The intelligent assessment system for information literacy of college students according to claim 4, characterized in that, The intelligent tool collaboration capability scoring submodule includes: The tool compatibility analysis unit is used to analyze the matching degree between all application tools in each student's tool combination application chain and the target task scenario as the tool compatibility degree. The operational collaboration analysis unit is used to score the data flow efficiency between tools in each student's tool combination application chain to obtain the operational collaboration degree. The application innovation analysis unit is used to calculate the proportion of unconventional tool combinations in each student's tool combination application chain based on the conventional tool combination library, as the combination application innovation. The Intelligent Tool Collaboration Ability Scoring Unit is used to calculate each student's Intelligent Tool Collaboration Ability Score by weighting their tool compatibility, operational collaboration, and innovative combination applications.

7. The intelligent assessment system for information literacy of college students according to claim 4, characterized in that, The knowledge transfer and innovation ability assessment submodule includes: The logical coherence scoring unit is used to obtain a logical coherence score based on the number of logical breaks in each student's knowledge transfer logical chain. The domain suitability scoring unit is used to obtain a domain suitability score based on the strength of the association between each student's knowledge transfer logic chain and the target task scenario; The value conversion score unit is used to obtain a value conversion score based on the contribution of the conclusions formed by each student's knowledge transfer logic chain to the solution of the target task. The knowledge transfer and innovation ability scoring unit is used to calculate each student's knowledge transfer and innovation ability score by weighting their logical coherence score, domain adaptability score, and value conversion score.

8. The intelligent assessment system for information literacy of college students according to claim 1, characterized in that, The implicit operation analysis module includes: The step-by-step sub-module is used to identify all key and invalid operation steps based on each student's interdisciplinary coverage data, tool combination application chain, and knowledge transfer logic chain. The implicit operation analysis submodule is used to calculate the percentage of steps with pause times exceeding the preset pause time in all key operation steps based on each student's implicit operation data as operation hesitation degree, and to calculate the ratio of all invalid operation steps to the total number of operation steps as operation redundancy degree, and to calculate the average percentage of tooltip calls as tool dependency degree. The basic competency assessment submodule is used to perform weighted calculations on each student's operational hesitation, operational redundancy, and tool dependence to obtain the basic operational competency assessment results.

9. The intelligent assessment system for information literacy of college students according to claim 1, characterized in that, The multidimensional results output module includes: The subject deviation analysis submodule is used to take the ratio of the difference between each student's assessment results of mathematical intelligence characteristic ability or basic operation ability and the average score of the corresponding professional background and corresponding ability type to the standard deviation of the corresponding professional background and corresponding ability type as the subject deviation of the corresponding student's corresponding ability type. The Digital Intelligence Ability Positioning Submodule is used to filter out each student's weak ability type and / or strong ability type based on the subject deviation of each student's digital intelligence characteristic ability type and the subject deviation of their basic operation ability type. The heatmap generation submodule is used to visualize the subject deviation of all students' digital intelligence characteristic ability types and basic operation ability types in the form of a two-dimensional matrix heatmap, and to visualize and mark the weak ability types and / or strong ability types of all students to obtain the information literacy assessment heatmap of all students.

10. A method for intelligently assessing the information literacy of college students, characterized in that, include: Each student's information literacy is assessed using an integrated assessment task, and explicit and implicit operational data of each student are collected simultaneously when completing the assessment. Based on the explicit operational data of each student when completing the assessment, we can analyze each student's interdisciplinary coverage data, tool combination application chain, and knowledge transfer logic chain. The assessment results of each student's digital intelligence-specific abilities are evaluated based on the interdisciplinary coverage data, the application chain of tools, and the knowledge transfer logic chain for each student. The basic operational ability assessment results for each student are evaluated based on the interdisciplinary coverage data, the tool combination application chain, the knowledge transfer logic chain, and the implicit operational data when completing the assessment. The results of each student's mathematical intelligence-specific ability assessment and basic operational ability assessment are compared and analyzed across subjects to identify weaknesses in mathematical intelligence ability. The results are then output in the form of a heat map to obtain a heat map of all students' information literacy assessment.

Images