A Socratic method and device for cultivating comprehensive qualities based on projection interaction

By using projection interaction technology to evaluate and generate progressive questioning logic, the shortcomings of existing intelligent learning devices in the assessment and guidance of quality education are solved, realizing dynamic assessment and effective guidance of users' comprehensive qualities and improving the training effect.

CN123118323APending Publication Date: 2026-11-17北京爱宾果科技有限公司
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611530726.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-30
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing intelligent learning devices suffer from problems in quality education assessment and guidance, such as distorted assessment results, lack of Socratic inspiration and questioning, inability to dynamically generate questions, and inability to precisely reinforce specific cognitive blind spots.

Method used

By using a projection-based interactive method, user response data is obtained, and indices for logical reasoning, information integration, critical thinking, and spatial imagination are calculated. This generates progressively challenging logical guidance content, which is then reinforced through variations of exercises, forming a closed-loop chain of assessment, guidance, and verification.

Benefits of technology

It significantly improves the relevance and effectiveness of skills training, can dynamically assess users' actual skill levels, promotes users' self-correction of cognitive biases, and improves overall skill scores.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123118323A_ABST
    Figure CN123118323A_ABST
Patent Text Reader

Abstract

This application belongs to the field of data processing technology, and specifically relates to a Socratic method and device for cultivating comprehensive qualities based on projection interaction. The method includes: retrieving target exercises from a quality exercise database and projecting them; acquiring user answer data and calculating a comprehensive quality score based on a preset set of evaluation parameters, comprising indices for logical reasoning, information integration, critical thinking, and spatial imagination; if the score does not reach a threshold, generating progressive counter-questions based on the score weaknesses and projecting them to guide the user in correcting cognitive biases; after the guidance, projecting modified exercises of the same type, and recalculating the score based on the consolidated answer data until the target is reached. This invention effectively solves the problems of difficulty in quantifying implicit abilities and lack of targeted guidance in quality education, significantly improving the accuracy of comprehensive quality assessment and teaching effectiveness.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application belongs to the field of data processing technology, and specifically relates to a Socratic method and device for cultivating comprehensive qualities based on projection interaction. Background Technology

[0002] While intelligent learning devices are now widespread, they still face significant technological limitations in assessing and guiding holistic education. Existing systems often rely on rote memorization and test-taking strategies, evaluating learning outcomes solely based on the correctness of answers. This standardized approach fails to capture the underlying thought processes of users. Especially when it comes to interdisciplinary skills like spatial reasoning and logical deduction, traditional assessment methods cannot distinguish the root causes of errors. For example, if a student answers a geometry problem incorrectly, the system often cannot determine whether it's due to a calculation error or an inability to mentally visualize the transition from a three-dimensional solid to a two-dimensional plane. This black-box approach to the thought process distorts assessment results and hinders genuine personalized education.

[0003] Meanwhile, existing guidance mechanisms are mostly didactic, where the system tends to directly display the standard answer or mechanically repeat knowledge points upon detecting an error, lacking Socratic inspiration and questioning. In this model, users are not required to engage in deep independent thinking, but merely passively receive information. This not only fails to effectively correct cognitive biases but also weakens the development of critical thinking. Particularly in cultivating spatial imagination, static graphic explanations cannot dynamically demonstrate the topological transformations of geometric shapes, making it difficult for users to establish accurate mental representations. Furthermore, existing question banks rely heavily on manual pre-setting of variations, lacking the ability to dynamically generate questions based on the user's real-time cognitive state. The rigid push strategy cannot accurately reinforce specific spatial cognitive blind spots such as zero-expansion or single-level expansion. In addition, scoring models often use fixed weighting coefficients, ignoring the natural differences in the weighting of different question types and failing to adaptively optimize based on long-term learning data, making it difficult for the system to accurately assess the user's actual competence level. Summary of the Invention

[0004] To address the aforementioned issues, this application provides a Socratic method and apparatus for cultivating comprehensive qualities based on projection interaction.

[0005] The first aspect of this application is a Socratic method for cultivating comprehensive qualities based on projection interaction, which mainly includes:

[0006] Step S1: Retrieve the target exercise from the pre-built competency exercise database and project it onto the interactive interface;

[0007] Step S2: Obtain the user's initial answer data for the target exercise, including the calculation or reasoning process, and calculate the user's current comprehensive quality score based on the initial answer data. The current comprehensive quality score is a weighted aggregation of logical reasoning index, information integration index, critical thinking index and spatial imagination index.

[0008] Step S3: Determine whether the current comprehensive quality score has reached the preset mastery threshold. If it has not reached the threshold, generate guiding content containing progressive questioning logic based on the index values ​​of the current comprehensive quality score and output it through the projection module to induce the user to correct cognitive biases.

[0009] Step S4: After the guidance process is completed, project the variant exercises of the same type as the target exercises onto the interactive interface, and obtain the user's consolidation answer data for the variant exercises again. Based on the consolidation answer data, recalculate the user's updated comprehensive quality score until the updated comprehensive quality score reaches the preset mastery threshold.

[0010] Preferably, in step S2, calculating the user's current comprehensive quality score based on the initial response data includes:

[0011] Step S21: Determine the derivation steps, knowledge points, interference items, and answer results that hit the evaluation parameter set in the initial answer data. The evaluation parameter set includes a complete thought chain node vector composed of preset logical derivation steps for the target exercise, a knowledge point weight vector composed of each set knowledge point and its weight, an interference vector composed of each set interference item and its weight, and an answer vector composed of different answer results and their weights. Different answer results correspond to different spatial expansion levels.

[0012] Step S22: The ratio of the number of derivation steps of hitting the judgment parameter set to the total number of preset logical derivation steps is used as the logical reasoning index; the sum of the weights of each knowledge point in the hit judgment parameter set is used as the information integration index; the value after deducting the weights of each interference item in the hit judgment parameter set from the preset non-interference index value is used as the critical thinking index; and the weights corresponding to the answer results of hitting the judgment parameter set are used as the spatial imagination index.

[0013] Step S23: Calculate the current comprehensive quality score through weighted aggregation.

[0014] Preferably, step S3 further includes:

[0015] Step S31: Monitor the user's response time and accuracy to the current level's questions in real time;

[0016] Step S32: When the response time exceeds the preset value or the response is inaccurate, add auxiliary prompts at the current level. The auxiliary prompts include highlighting relevant thought chain nodes or visualizing the weight of relevant knowledge points.

[0017] Preferably, the method further includes:

[0018] Step S5: Adjust the weights of the indicators in stages. Based on the user's historical answer data, calculate the correlation coefficient between each index and the unweighted comprehensive quality score. The unweighted comprehensive quality score is the score after adding up all the indices. When the overall correlation between a certain index and the unweighted comprehensive quality score is higher than that of other indices, increase the weight of that index and decrease the weight of other indices proportionally.

[0019] Preferably, in step S5, after every 50 target exercises, a phased adjustment of the indicator weights is performed.

[0020] The second aspect of this application provides a Socratic-style comprehensive quality cultivation device based on projection interaction, mainly comprising:

[0021] The target exercise generation module is used to retrieve target exercises from a pre-built competency exercise database and project them onto the interactive interface;

[0022] The current comprehensive quality score calculation module is used to obtain the user's initial answer data for the target exercise, which includes the calculation or reasoning process, and calculate the user's current comprehensive quality score based on the initial answer data. The current comprehensive quality score is a weighted aggregation of logical reasoning index, information integration index, critical thinking index and spatial imagination index.

[0023] The cognitive guidance module is used to determine whether the current comprehensive quality score has reached the preset mastery threshold. If it has not reached the threshold, it generates guidance content containing progressive questioning logic based on the index values ​​of the current comprehensive quality score and outputs it through the projection module to induce the user to correct cognitive biases.

[0024] The consolidation and answering module is used to project variant exercises of the same type as the target exercises onto the interactive interface after the guidance process is completed, and to obtain the user's consolidation and answering data for the variant exercises again, and to recalculate the user's updated comprehensive quality score based on the consolidation and answering data, until the updated comprehensive quality score reaches the preset mastery threshold.

[0025] Preferably, the current comprehensive quality score calculation module includes:

[0026] The data extraction unit is used to determine the number of derivation steps, knowledge points, interference items, and answer results that hit the evaluation parameter set in the initial answer data. The evaluation parameter set includes a complete thought chain node vector consisting of preset logical derivation steps for the target exercise, a knowledge point weight vector consisting of each set knowledge point and its weight, an interference vector consisting of each set interference item and its weight, and an answer vector consisting of different answer results and their weights. Different answer results correspond to different spatial expansion levels.

[0027] The index calculation unit is used to take the ratio of the number of derivation steps of hitting the evaluation parameter set to the total number of preset logical derivation steps as the logical reasoning index, the sum of the weights of each knowledge point of hitting the evaluation parameter set as the information integration index, the value after deducting the weights of each interference item of hitting the evaluation parameter set from the preset non-interference index value as the critical thinking index, and the weight corresponding to the answer result of hitting the evaluation parameter set as the spatial imagination index.

[0028] The weighted calculation unit is used to calculate the current comprehensive quality score through weighted aggregation.

[0029] Preferably, the cognitive guidance module includes:

[0030] The response monitoring unit is used to monitor the response time and accuracy of the user's response to the current level's questions in real time.

[0031] The auxiliary prompt unit is used to add auxiliary prompts at the current level when the response time exceeds a preset value or the response is inaccurate. The auxiliary prompts include highlighting relevant thought chain nodes or visualizing the weight of relevant knowledge points.

[0032] Preferably, the device further includes:

[0033] The indicator weight adjustment module is used to adjust the indicator weights in stages. Based on the user's historical answer data, it calculates the correlation coefficient between each index and the unweighted comprehensive quality score. The unweighted comprehensive quality score is the score after adding up all the indices. When the overall correlation between a certain index and the unweighted comprehensive quality score is higher than that of other indices, the weight ratio of that index is increased, and the weight ratio of other indices is reduced proportionally.

[0034] Preferably, in the indicator weight adjustment module, a phased indicator weight adjustment is performed after every 50 target exercises have been completed.

[0035] This application improves the accuracy of comprehensive quality assessment and the effectiveness of guidance. Attached Figure Description

[0036] Figure 1This is a flowchart of one implementation method of the Socratic comprehensive quality cultivation method based on projection interaction in this application. Detailed Implementation

[0037] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this application will be described in more detail below with reference to the accompanying drawings. In the drawings, the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The described embodiments are only some, not all, of the embodiments of this application. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. The embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0038] The first aspect of this application provides a Socratic method for cultivating comprehensive qualities based on projection interaction, such as... Figure 1 As shown, it mainly includes:

[0039] Step S1: Retrieve the target exercise from the pre-built competency exercise database and project it onto the interactive interface;

[0040] Step S2: Obtain the user's initial answer data for the target exercise, including the calculation or reasoning process, and calculate the user's current comprehensive quality score based on the initial answer data. The current comprehensive quality score is a weighted aggregation of logical reasoning index, information integration index, critical thinking index and spatial imagination index.

[0041] Step S3: Determine whether the current comprehensive quality score has reached the preset mastery threshold. If it has not reached the threshold, generate guiding content containing progressive questioning logic based on the index values ​​of the current comprehensive quality score and output it through the projection module to induce the user to correct cognitive biases.

[0042] Step S4: After the guidance process is completed, project the variant exercises of the same type as the target exercises onto the interactive interface, and obtain the user's consolidation answer data for the variant exercises again. Based on the consolidation answer data, recalculate the user's updated comprehensive quality score until the updated comprehensive quality score reaches the preset mastery threshold.

[0043] This embodiment presents the core process of quality education based on projection interaction. Step S1 calls a pre-built quality exercise database to ensure that the exercises have their own logical chains. Step S2 quantifies four underlying general abilities—logical reasoning, information integration, critical thinking, and spatial imagination—by parsing the answer data containing calculation or reasoning processes, thus avoiding the one-sidedness of traditional evaluation that only relies on right or wrong results. Step S3 generates progressive counter-questions based on the scoring weaknesses, simulating the Socratic midwifery method, to induce users to self-correct cognitive biases rather than directly providing answers. Step S4 conducts secondary evaluation by introducing modified exercises to verify the guidance effect, forming a closed-loop link of evaluation, guidance, and verification.

[0044] This application overcomes the limitations of traditional learning systems' rote memorization and rote learning approach by concretizing holistic education into a calculable index system. Through interactive projection, the system can visualize abstract thought processes and effectively distinguish between superficial mastery and genuine transfer of knowledge using varied exercises, significantly improving the relevance and effectiveness of holistic education. Detailed explanation follows.

[0045] First, in step S1, a learning machine with a projection and scanning interface retrieves a target exercise from the competency exercise database. For example, the exercise might be: An ant crawls from point A on the bottom surface of a cylinder with a base radius of 3cm and a height of 10cm along its side to point B on the top surface. Find the shortest path.

[0046] The processor controls the projection component to project the question and the three-dimensional cylindrical model onto the desktop or wall.

[0047] Step S2 is used to calculate the current comprehensive quality score.

[0048] In some optional implementations, step S2, calculating the user's current overall quality score based on the initial response data, includes:

[0049] Step S21: Determine the derivation steps, knowledge points, interference items, and answer results that hit the evaluation parameter set in the initial answer data. The evaluation parameter set includes a complete thought chain node vector composed of preset logical derivation steps for the target exercise, a knowledge point weight vector composed of each set knowledge point and its weight, an interference vector composed of each set interference item and its weight, and an answer vector composed of different answer results and their weights. Different answer results correspond to different spatial expansion levels.

[0050] Step S22: The ratio of the number of derivation steps of hitting the judgment parameter set to the total number of preset logical derivation steps is used as the logical reasoning index; the sum of the weights of each knowledge point in the hit judgment parameter set is used as the information integration index; the value after deducting the weights of each interference item in the hit judgment parameter set from the preset non-interference index value is used as the critical thinking index; and the weights corresponding to the answer results of hitting the judgment parameter set are used as the spatial imagination index.

[0051] Step S23: Calculate the current comprehensive quality score through weighted aggregation.

[0052] It should be noted that the "quality education" in this application generally refers to an educational process that, based on the memorization of non-specific subject knowledge, guides individuals to engage in independent thinking, interdisciplinary integration, and cognitive reconstruction, continuously developing fundamental general abilities such as logical reasoning, information integration, critical thinking, logical expression, and spatial imagination. Therefore, when designing quality education exercises, we reject using scores for single-subject knowledge points as a representation of ability. Instead, we map the exercises to several fundamental general ability dimensions. Furthermore, we need to label each exercise with its activation parameters and weights on each indicator dimension, so that any answer can be analyzed as a contribution to the core indicator set, also known as the evaluation parameter set. The system compares the user's answers with each parameter in the evaluation parameter set and determines the activation parameters for each indicator based on the matching results; these are the logical reasoning index, information integration index, critical thinking index, and spatial imagination index. The weights of each indicator dimension can be preset through labeling, highly correlated with the exercise type, or set as a single fixed weight. They can also be set according to the user's current learning stage, or dynamically adjusted as described in the subsequent embodiments of this application.

[0053] Taking the above exercise on calculating the shortest path as an example, the set of evaluation parameters includes:

[0054] Thinking node vector {identifying topological constraints, surface unfolding, identifying geometric relationships in right triangles, and performing algebraic calculations using the Pythagorean theorem};

[0055] Knowledge point weight vector {lateral expansion, weight 0.6; Pythagorean theorem, weight 0.4};

[0056] The interference vector is a straight line passing through the inside of the cylinder, with a weight of 0.9.

[0057] The answer vector is {answer A, weight a, answer B, weight b, answer C, weight c}.

[0058] Suppose the user's initial answer is "to pass directly through the inside of the cylinder, the distance is the square root of 10 squared plus the square root of 6 squared, which is approximately 11.66 centimeters".

[0059] The system identification result is as follows:

[0060] The Pythagorean theorem was used, and the number of knowledge points called was K=1;

[0061] The surface was not unfolded, and the concept of "geometric unfolding" was not utilized.

[0062] It falls into the preset interference item and passes through the inside of the cylinder in a straight line;

[0063] The space transformation was not carried out by side expansion.

[0064] Specifically, in step S21, the system identifies that the user has only completed two valid derivation steps: "identifying right triangles" and "substituting the Pythagorean theorem," meaning the number of derivation steps in the evaluation parameter set is L=2. The system also identifies that the user only called the single knowledge point "Pythagorean theorem," meaning K=1 (number of knowledge points called), and did not call the "lateral expansion" knowledge point. The knowledge point and weight in the evaluation parameter set is "lateral expansion, weight 0.6." The user's answer falls into a preset interference item, meaning the interference item in the evaluation parameter set is "a straight line passes through the inside of a cylinder." The user's answer is A, with a weight of a.

[0065] In step S22, this application provides a specific method for calculating each index based on the parameters obtained in step S21. The correlation between each index and the parameters is explained here. For the logical reasoning index, by using the number of logical deduction steps, thought transition nodes, and reasoning chain breakpoints in the exercises as quantitative parameters, logical reasoning ability becomes a fundamental indicator reflecting students' abstract reasoning and deductive abilities. This embodiment only uses the number of deduction steps to calculate the logical reasoning index. For the information integration index, by statistically analyzing the number of cross-knowledge point calls, association weights, and cross-disciplinary path triggering in the answers, information integration ability is defined as a fundamental indicator reflecting the breadth and integrated application level of students' knowledge networks. For the critical thinking index, by using preset interference items, cognitive conflict situations, and rhetorical questioning guidance, students' identification, refutation, and self-correction behaviors of fallacies are quantified into critical thinking indicators. This embodiment only uses preset interference items for calculation. For the spatial imagination index, spatial imagination can be directly inferred from the answer results. Different answer results correspond to different spatial imagination. For example, expanding the space based on the first level may result in the wrong answer B. Expanding it to two levels may result in the correct answer C. Not expanding it at all may result in other answers. Therefore, spatial imagination can be directly evaluated based on the answer.

[0066] The following explains the calculation methods for each index, including the logical reasoning index. Calculated using the following formula:

[0067] ;

[0068] in, The number of derivation steps for the hit evaluation parameter set. To find the total number of logical deduction steps, substitute the values ​​to obtain the logical reasoning index. It is 0.5.

[0069] Information Integration Index The calculation is based on the number of knowledge point types invoked in the user's answer and the weight of the associated knowledge points in the database. The calculation formula is as follows:

[0070] ;

[0071] in, To determine the number of knowledge point types, The information integration index is obtained by substituting the value into the weight of the i-th knowledge point. It is 0.4.

[0072] Interference Index The calculation formula is:

[0073] ;

[0074] in, The preset "no-interference index" value refers to the value at which the user's answer does not fall into any of the interference items. For example, a value of 1. If it is subsequently found that the user's answer falls into any of the preset interference items, the weight value of the corresponding interference item will be deducted. The number of terms that fall into the interference category. Let be the weight of the j-th interference item. In the example above, the user's answer falls into the preset interference item of a straight line passing through the inside of a cylinder, which has a weight of 0.9. Then, the interference index can be calculated. It is 0.1.

[0075] Spatial Imagination Index Calculated using the following formula:

[0076] ;

[0077] in, The weight corresponding to the answer result that hits the evaluation parameter set. For example, in the previous example, if the user's answer result is A, the corresponding weight a = 0.2, then the spatial imagination index... It is 0.2.

[0078] Then, in step S23, the current comprehensive quality score is calculated by weighted aggregation.

[0079] In some optional implementations, step S2, calculating the user's current overall quality score based on the initial response data, includes:

[0080] Step S21: Determine the derivation steps, knowledge points, interference items, and answer results that hit the evaluation parameter set in the initial answer data. The evaluation parameter set includes a complete thought chain node vector composed of preset logical derivation steps for the target exercise, a knowledge point weight vector composed of each set knowledge point and its weight, an interference vector composed of each set interference item and its weight, and an answer vector composed of different answer results and their weights. Different answer results correspond to different spatial expansion levels.

[0081] Step S22: The ratio of the number of derivation steps of hitting the judgment parameter set to the total number of preset logical derivation steps is used as the logical reasoning index; the sum of the weights of each knowledge point in the hit judgment parameter set is used as the information integration index; the value after deducting the weights of each interference item in the hit judgment parameter set from the preset non-interference index value is used as the critical thinking index; and the weights corresponding to the answer results of hitting the judgment parameter set are used as the spatial imagination index.

[0082] Step S23: Calculate the current comprehensive quality score through weighted aggregation.

[0083] Then, in step S3, when the current comprehensive quality score does not reach the preset mastery threshold, Socratic dialogue guidance is conducted.

[0084] In this embodiment, the system judged the score to be low, and the spatial imagination index was severely insufficient. Based on the distribution of weaknesses, the above exercises and answers, comprehensive quality scores and explanations, etc., were input into the large model as prompts. The large model then generated progressive rhetorical questions: You calculated the straight-line distance through a solid cylinder. If an ant could only crawl on the outer surface, could it still walk like this? Imagine cutting open the side and flattening it, what would the path become?

[0085] In some alternative implementations, step S3 further includes:

[0086] Step S31: Monitor the user's response time and accuracy to the current level's questions in real time;

[0087] Step S32: When the response time exceeds the preset value or the response is inaccurate, add auxiliary prompts at the current level. The auxiliary prompts include highlighting relevant thought chain nodes or visualizing the weight of relevant knowledge points.

[0088] For example, the projection module dynamically demonstrates the transformation process of cutting open the side of a cylinder and flattening it into a rectangle.

[0089] Finally, in step S4, after the guidance is completed, the system automatically generates variation exercises, updates the comprehensive quality score based on the user's answer data, until the user masters the relevant types of exercises.

[0090] In this step, the variation exercises are generated by adjusting the spatial orientation, parameter ratios, or constraints of the geometric objects while keeping the core logical framework unchanged. For example, an automatically generated variation exercise might be: If the height of a cylinder becomes 12cm, what is the shortest path for an ant starting from point A, circling the side once, and reaching point B?

[0091] In some alternative implementations, the method further includes:

[0092] Step S5: Adjust the weights of the indicators in stages. Based on the user's historical answer data, calculate the correlation coefficient between each index and the unweighted comprehensive quality score. The unweighted comprehensive quality score is the score after adding up all the indices. When the overall correlation between a certain index and the unweighted comprehensive quality score is higher than that of other indices, increase the weight of that index and decrease the weight of other indices proportionally.

[0093] In some alternative implementations, in step S5, a phased adjustment of the indicator weights is performed after every 50 target exercises have been completed.

[0094] It should be noted that the correlation coefficient is calculated based on the original index scores of each underlying general ability and the target variable of comprehensive quality evaluation before weighted aggregation, rather than based on the current comprehensive quality score after weighting, to prevent strong coupling between the current comprehensive quality score and the high-weight index.

[0095] Taking a geometry learning scenario as an example, the system first initializes the spatial imagination index with a basic weight of 0.4 based on the learning stage of eighth-grade students. In the first round of responses, the index score was only 0.3, and historical correlation analysis showed that its correlation coefficient with the total score was as high as 0.88. The system then temporarily increased its weight to 0.5 and generated guiding content centered on a spatial folding demonstration.

[0096] It should be noted that the temporarily increased weight must not exceed the preset upper limit, for example, the upper limit is set to 0.6; the temporary increase in weight is only valid within the current guiding cycle, and will revert to the basic weight in the next cycle.

[0097] In an alternative implementation, the system can also determine whether the temporarily increased weight needs to revert to the base weight based on the guidance results. For example, if a user's score on a certain index improves significantly after guidance, the increased weight can be reduced back to the initial weight parameter; otherwise, the increased weight value can be retained. For instance, after guidance, a student's spatial imagination index in modified exercises increases to 0.75, an improvement of 0.45. Based on this, the system determines that the dimension has been significantly improved and automatically lowers its base weight to the base weight of 0.4 in the next cycle.

[0098] The second aspect of this application provides a Socratic-style comprehensive quality cultivation device based on projection interaction, mainly comprising:

[0099] The target exercise generation module is used to retrieve target exercises from a pre-built competency exercise database and project them onto the interactive interface;

[0100] The current comprehensive quality score calculation module is used to obtain the user's initial answer data for the target exercise, which includes the calculation or reasoning process, and calculate the user's current comprehensive quality score based on the initial answer data. The current comprehensive quality score is a weighted aggregation of logical reasoning index, information integration index, critical thinking index and spatial imagination index.

[0101] The cognitive guidance module is used to determine whether the current comprehensive quality score has reached the preset mastery threshold. If it has not reached the threshold, it generates guidance content containing progressive questioning logic based on the index values ​​of the current comprehensive quality score and outputs it through the projection module to induce the user to correct cognitive biases.

[0102] The consolidation and answering module is used to project variant exercises of the same type as the target exercises onto the interactive interface after the guidance process is completed, and to obtain the user's consolidation and answering data for the variant exercises again, and to recalculate the user's updated comprehensive quality score based on the consolidation and answering data, until the updated comprehensive quality score reaches the preset mastery threshold.

[0103] In some optional implementations, the current comprehensive quality score calculation module includes:

[0104] The data extraction unit is used to determine the number of derivation steps, knowledge points, interference items, and answer results that hit the evaluation parameter set in the initial answer data. The evaluation parameter set includes a complete thought chain node vector consisting of preset logical derivation steps for the target exercise, a knowledge point weight vector consisting of each set knowledge point and its weight, an interference vector consisting of each set interference item and its weight, and an answer vector consisting of different answer results and their weights. Different answer results correspond to different spatial expansion levels.

[0105] The index calculation unit is used to take the ratio of the number of derivation steps of hitting the evaluation parameter set to the total number of preset logical derivation steps as the logical reasoning index, the sum of the weights of each knowledge point of hitting the evaluation parameter set as the information integration index, the value after deducting the weights of each interference item of hitting the evaluation parameter set from the preset non-interference index value as the critical thinking index, and the weight corresponding to the answer result of hitting the evaluation parameter set as the spatial imagination index.

[0106] The weighted calculation unit is used to calculate the current comprehensive quality score through weighted aggregation.

[0107] In some alternative implementations, the cognitive guidance module includes:

[0108] The response monitoring unit is used to monitor the response time and accuracy of the user's response to the current level's questions in real time.

[0109] The auxiliary prompt unit is used to add auxiliary prompts at the current level when the response time exceeds a preset value or the response is inaccurate. The auxiliary prompts include highlighting relevant thought chain nodes or visualizing the weight of relevant knowledge points.

[0110] In some alternative embodiments, the apparatus further includes:

[0111] The indicator weight adjustment module is used to adjust the indicator weights in stages. Based on the user's historical answer data, it calculates the correlation coefficient between each index and the unweighted comprehensive quality score. The unweighted comprehensive quality score is the score after adding up all the indices. When the overall correlation between a certain index and the unweighted comprehensive quality score is higher than that of other indices, the weight ratio of that index is increased, and the weight ratio of other indices is reduced proportionally.

[0112] In some optional implementations, the indicator weight adjustment module performs a phased indicator weight adjustment after every 50 target exercises.

[0113] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A Socratic comprehensive quality training method based on projection interaction, characterized in that, include: Step S1: Retrieve the target exercise from the pre-built competency exercise database and project it onto the interactive interface; Step S2: Obtain the user's initial answer data for the target exercise, including the calculation or reasoning process, and calculate the user's current comprehensive quality score based on the initial answer data. The current comprehensive quality score is a weighted aggregation of logical reasoning index, information integration index, critical thinking index and spatial imagination index. Step S3: Determine whether the current comprehensive quality score has reached the preset mastery threshold. If it has not reached the threshold, generate guiding content containing progressive questioning logic based on the index values ​​of the current comprehensive quality score and output it through the projection module to induce the user to correct cognitive biases. Step S4: After the guidance process ends, project the variant exercises of the same type as the target exercises onto the interactive interface, and obtain the user's consolidation answer data for the variant exercises again. Based on the consolidation answer data, recalculate the user's updated comprehensive quality score until the updated comprehensive quality score reaches the preset mastery threshold.

2. The Socratic comprehensive quality training method based on projection interaction according to claim 1, characterized in that, In step S2, calculating the user's current comprehensive quality score based on the initial response data includes: Step S21: Determine the derivation steps, knowledge points, interference items, and answer results that hit the evaluation parameter set in the initial answer data. The evaluation parameter set includes a complete thought chain node vector composed of preset logical derivation steps for the target exercise, a knowledge point weight vector composed of each set knowledge point and its weight, an interference vector composed of each set interference item and its weight, and an answer vector composed of different answer results and their weights. Different answer results correspond to different spatial expansion levels. Step S22: The ratio of the number of derivation steps of hitting the judgment parameter set to the total number of preset logical derivation steps is used as the logical reasoning index; the sum of the weights of each knowledge point in the hit judgment parameter set is used as the information integration index; the value after deducting the weights of each interference item in the hit judgment parameter set from the preset non-interference index value is used as the critical thinking index; and the weights corresponding to the answer results of hitting the judgment parameter set are used as the spatial imagination index. Step S23: Calculate the current comprehensive quality score through weighted aggregation.

3. The Socratic comprehensive quality training method based on projection interaction according to claim 1, characterized in that, Step S3 further includes: Step S31: Monitor the user's response time and accuracy to the current level's questions in real time; Step S32: When the response time exceeds the preset value or the response is inaccurate, add auxiliary prompts at the current level. The auxiliary prompts include highlighting relevant thought chain nodes or visualizing the weight of relevant knowledge points.

4. The Socrates-based comprehensive quality training method based on projection interaction according to claim 1, characterized in that, The method further includes: Step S5: Adjust the weights of the indicators in stages. Based on the user's historical answer data, calculate the correlation coefficient between each index and the unweighted comprehensive quality score. The unweighted comprehensive quality score is the score after adding up all the indices. When the overall correlation between a certain index and the unweighted comprehensive quality score is higher than that of other indices, increase the weight of that index and decrease the weight of other indices proportionally.

5. The Socrates-based comprehensive quality training method based on projection interaction according to claim 4, characterized in that, In step S5, after every 50 target exercises, a phased adjustment of the indicator weights is performed.

6. A Socrates type comprehensive quality training device based on projection interaction, characterized in that, include: The target exercise generation module is used to retrieve target exercises from a pre-built competency exercise database and project them onto the interactive interface; The current comprehensive quality score calculation module is used to obtain the user's initial answer data for the target exercise, which includes the calculation or reasoning process, and calculate the user's current comprehensive quality score based on the initial answer data. The current comprehensive quality score is a weighted aggregation of logical reasoning index, information integration index, critical thinking index and spatial imagination index. The cognitive guidance module is used to determine whether the current comprehensive quality score has reached the preset mastery threshold. If it has not reached the threshold, it generates guidance content containing progressive questioning logic based on the index values ​​of the current comprehensive quality score and outputs it through the projection module to induce the user to correct cognitive biases. The consolidation and answering module is used to project variant exercises of the same type as the target exercises onto the interactive interface after the guidance process is completed, and to obtain the user's consolidation and answering data for the variant exercises again. Based on the consolidation and answering data, the user's updated comprehensive quality score is recalculated until the updated comprehensive quality score reaches the preset mastery threshold.

7. The Socrates-type integrated diathesis training device based on projection interaction according to claim 6, characterized in that, The current comprehensive quality score calculation module includes: The data extraction unit is used to determine the number of derivation steps, knowledge points, interference items, and answer results that hit the evaluation parameter set in the initial answer data. The evaluation parameter set includes a complete thought chain node vector consisting of preset logical derivation steps for the target exercise, a knowledge point weight vector consisting of each set knowledge point and its weight, an interference vector consisting of each set interference item and its weight, and an answer vector consisting of different answer results and their weights. Different answer results correspond to different spatial expansion levels. The index calculation unit is used to take the ratio of the number of derivation steps of hitting the evaluation parameter set to the total number of preset logical derivation steps as the logical reasoning index, the sum of the weights of each knowledge point of hitting the evaluation parameter set as the information integration index, the value after deducting the weights of each interference item of hitting the evaluation parameter set from the preset non-interference index value as the critical thinking index, and the weight corresponding to the answer result of hitting the evaluation parameter set as the spatial imagination index. The weighted calculation unit is used to calculate the current comprehensive quality score through weighted aggregation. 8.The projection interaction-based Socrates-type comprehensive quality training device according to claim 6, wherein, The cognitive guidance module includes: The response monitoring unit is used to monitor the response time and accuracy of the user's response to the current level's questions in real time. The auxiliary prompt unit is used to add auxiliary prompts at the current level when the response time exceeds a preset value or the response is inaccurate. The auxiliary prompts include highlighting relevant thought chain nodes or visualizing the weight of relevant knowledge points. 9.The projection interaction-based Socrates-type comprehensive quality training device according to claim 6, wherein, The device further includes: The indicator weight adjustment module is used to adjust the indicator weights in stages. Based on the user's historical answer data, it calculates the correlation coefficient between each index and the unweighted comprehensive quality score. The unweighted comprehensive quality score is the score after adding up all the indices. When the overall correlation between a certain index and the unweighted comprehensive quality score is higher than that of other indices, the weight ratio of that index is increased, and the weight ratio of other indices is reduced proportionally. 10.The projection interaction-based Socrates-type comprehensive quality training device according to claim 9, wherein, In the indicator weight adjustment module, a phased indicator weight adjustment is performed after every 50 target exercises.