Offline learning material processing method and device, electronic equipment and storage medium

Abstract

Embodiments of the present disclosure disclose a method and device for processing offline learning materials, electronic equipment and a storage medium. The method comprises: obtaining offline learning materials uploaded to an education platform, and splitting the offline learning materials into a plurality of structured objects; determining a first answering process graph of the structured objects based on object structure parameters of the structured objects; determining a second answering process graph of the structured objects, determining an answering difference result based on the first answering process graph and the second answering process graph of the structured objects; generating target review data based on the answering difference result and corresponding historical review data, and updating a learning growth record of the education platform based on the target review data; and the historical review data comprises a learning progress in the education platform for a question type corresponding to the structured objects. The present disclosure can not only improve the conversion efficiency of the individual review closed loop, but also complement the data perception ability of the education platform for the offline learning process.

Description

Technical Field

[0001] This disclosure relates to the field of educational information processing technology, and in particular to a method, apparatus, electronic device, and storage medium for processing offline learning materials. Background Technology

[0002] With the development of educational platforms and educational AI systems, existing educational platforms can usually handle exercises, assessments, and incorrect answers within the platform. However, for external learning materials uploaded to the platform, they can typically only save the question images, query the answers, or provide one-time grading, thus obtaining the processing results of the external learning materials. However, the above processing results are only for the currently uploaded external learning materials and cannot enter the long-term learning loop of the educational platform. In other words, the educational platform still lacks a stable mechanism for processing external learning materials. Summary of the Invention

[0003] This disclosure provides a method, apparatus, electronic device, and storage medium for processing offline learning materials, which can both improve the conversion efficiency of the individual review loop and enhance the data perception capabilities of educational platforms for the offline learning process.

[0004] According to one aspect of this disclosure, a method for processing offline learning materials is provided, the method comprising: The system retrieves offline learning materials uploaded to an education platform and breaks them down into multiple structured objects. Each structured object represents a question. Each structured object is bound to corresponding object structured parameters, which include question type, question stem data, and answer data. Based on the object structuring parameters of the structured object, a first answer process diagram of the structured object is determined; the first answer process diagram is used to describe the sequence of answering steps of the structured object. A second answer process diagram of the structured object is determined, and the answer difference results are determined based on the first and second answer process diagrams of the structured object; the second answer process diagram is the standard answer step sequence of the structured object provided by the education platform; Based on the differences in the answers and the corresponding historical review data, target review data is generated, and the learning progress record of the education platform is updated based on the target review data; the historical review data includes the learning progress of the question type corresponding to the structured object in the education platform; the target review data is used to describe the learning strategy for the question type corresponding to the structured object.

[0005] According to another aspect of this disclosure, an apparatus for processing offline learning materials is provided, the apparatus comprising: The first data processing module is used to acquire offline learning materials uploaded to the education platform and split the offline learning materials into multiple structured objects; each structured object represents a question; each structured object is bound to corresponding object structured parameters; the object structured parameters include question type, question stem data, and answer data; The second data processing module is used to determine a first answer process diagram of the structured object based on the object structure parameters of the structured object; the first answer process diagram is used to describe the answering step sequence of the structured object; The difference determination module is used to determine the second answer process diagram of the structured object, and to determine the answer difference result based on the first answer process diagram and the second answer process diagram of the structured object; the second answer process diagram is the standard answer step sequence of the structured object provided by the education platform; The update module is used to generate target review data based on the answer difference results and the corresponding historical review data, and update the learning progress record of the education platform based on the target review data; the historical review data includes the learning progress of the question type corresponding to the structured object in the education platform; the target review data is used to describe the learning strategy for the question type corresponding to the structured object.

[0006] According to another aspect of this disclosure, an electronic device is provided, the electronic device comprising: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the offline learning material processing method described in any embodiment of this disclosure.

[0007] According to another aspect of this disclosure, a computer-readable storage medium is provided that stores computer instructions for causing a processor to execute and implement the offline learning material processing method described in any embodiment of this disclosure.

[0008] The technical solution of this disclosure involves acquiring offline learning materials uploaded to an education platform and splitting these materials into multiple structured objects. Each structured object represents a question; each structured object is bound to corresponding object structured parameters; these parameters include question type, question stem data, and answer data. This achieves precise splitting of the offline learning materials, facilitating accurate processing of each structured object in the future. Then, based on the object structured parameters, a first answer process diagram describing the answer step sequence of the structured object is determined, converting the object structured parameters of each question into a data format that allows for faster comparison later. Furthermore, a second answer process diagram is determined for each structured object, and the difference in answer results is determined by comparing the first and second answer process diagrams. This enables precise analysis of the differences between the standard answer step sequence provided by the education platform and the answer step sequence of the offline learning materials, thereby improving the education platform's data perception capabilities regarding the offline learning process. Correspondingly, based on the differences in answers and the corresponding historical review data, target review data is generated to describe the learning strategies for the question types corresponding to the structured objects. This is because the historical review data includes the learning progress of the question types corresponding to the structured objects in the education platform. In other words, the target review data generated based on the differences in answers and the corresponding historical review data can not only cover the education platform's own review strategies, but also combine the learning situation of offline learning materials. This can better provide users with accurate review strategies. Furthermore, the learning growth profile of the education platform can be updated based on the target review data to improve the conversion efficiency of the individual review loop and also make up for the education platform's data perception ability of the offline learning process.

[0009] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments of this disclosure, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 This is a flowchart of a method for processing offline learning materials according to an embodiment of this disclosure; Figure 2 This is a flowchart of another method for processing offline learning materials according to an embodiment of this disclosure; Figure 3This is a flowchart of another method for processing offline learning materials according to an embodiment of the present disclosure; Figure 4 This is a schematic diagram of the structure of an offline learning material processing device provided according to an embodiment of the present disclosure; Figure 5 This is a schematic diagram of the structure of an electronic device that implements the offline learning material processing method of the present disclosure according to an embodiment of the present disclosure. Detailed Implementation

[0012] To enable those skilled in the art to better understand the present disclosure, the technical solutions of the present disclosure will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present disclosure, and not all embodiments. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present disclosure.

[0013] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0014] Figure 1 This is a flowchart illustrating a method for processing offline learning materials according to an embodiment of this disclosure. This embodiment is applicable to situations where learning materials from outside the education platform are introduced into the education platform for processing. This method can be executed by an offline learning material processing device, which can be implemented in hardware and / or software and can be configured in any electronic device with network communication capabilities. Figure 1 As shown, the processing methods for this publicly available offline learning material may include: S110. Obtain the offline learning materials uploaded to the education platform and split the offline learning materials into multiple structured objects; each structured object represents a question; each structured object is bound to corresponding object structured parameters; the object structured parameters include question type, question stem data, and answer data.

[0015] Offline learning materials can be understood as learning materials completed after practicing various offline exercises and having those exercises graded. Exercise materials can be understood as various practice and test materials compiled based on teaching content, mainly used for knowledge consolidation, skills training, and learning effect assessment. Common forms of exercise materials can include lesson exercises, special exercises, unit tests, past exam papers, and mock exams. Structured objects can be understood as structured information representing the corresponding question information. For example, if the offline learning material is a single training test paper, the structured object can be the question number information of each question in the training test paper; if the offline learning material consists of multiple training test papers, the test paper identification information for each training test paper is determined, and the structured object can be the test paper identification information and the question number information; the test paper identification information can be understood as the identification information used to distinguish different training test papers.

[0016] Specifically, offline learning materials are segmented into text segments, the boundaries of each question and the corresponding reference data within those boundaries are identified, and an independent structured object is mapped for each question. Based on the reference data within each question's boundaries, the object's structured parameters for each question are determined, and the corresponding object structured parameters are bound to each structured object. The reference data refers to data associated with the corresponding question.

[0017] Optionally, in an embodiment of this disclosure, splitting offline learning materials into multiple structured objects may include: extracting data from the offline learning materials to obtain structured learning evidence; the structured learning evidence includes a question number area, a question stem area, an answer area, an annotation area, and a score area; and determining multiple structured objects and corresponding object structured parameters based on the structured learning evidence.

[0018] Specifically, extracting structured learning evidence from offline learning materials can include: performing region detection on the offline learning materials to obtain multiple question number regions, multiple question stem regions, multiple answer regions, multiple annotation regions, and multiple score regions; determining multiple structured objects based on all question number regions; determining the object's structured parameters based on the data in the question stem regions, answer regions, annotation regions, and score regions; and binding the object's structured parameters to the corresponding structured objects. Correspondingly, region detection on offline learning materials can include: identifying different data types in the offline learning materials, obtaining region detection methods based on the data types, and performing region detection based on the region detection methods. These region detection methods include one or more of the following: Optical Character Recognition (OCR), formula recognition, image recognition, and multimodal recognition models.

[0019] In this embodiment, offline learning materials are extracted to obtain structured learning evidence, enabling precise division of the question number area, question stem area, answer area, annotation area, and score area. Based on the structured learning evidence, multiple structured objects and their corresponding structured parameters can be accurately determined, avoiding the impact of inaccurate question segmentation on subsequent data processing.

[0020] Optionally, in this embodiment of the disclosure, the structured learning evidence further includes data identification confidence level and desensitization status. The desensitization status describes whether the structured learning evidence has been authorized and the degree of data desensitization. Dividing the structured learning evidence into multiple structured objects may include: if the data identification confidence level is greater than or equal to a preset identification confidence level, and the desensitization status is "desensitized," indicating that the structured learning evidence has been authorized and the degree of data desensitization exceeds a preset desensitization threshold, then multiple structured objects and their corresponding object structured parameters are determined based on the structured learning evidence. If the data identification confidence level is greater than or equal to a preset identification confidence level, and the desensitization status is "not desensitized," then the corresponding structured learning evidence is marked as pending desensitization, and the corresponding structured learning evidence is manually verified. If the verification result is "authorization approved," then multiple structured objects and their corresponding object structured parameters are determined based on the structured learning evidence. If the data identification confidence level is less than a preset identification confidence level, it indicates that the identification accuracy of the corresponding structured learning evidence is not high, and it is necessary to return to the offline learning materials to re-extract data and obtain the structured learning evidence.

[0021] The process of determining data recognition confidence can include: determining the data recognition confidence of the structured object based on the regional confidence of each recognition region corresponding to the structured object and the field recognition confidence of the data in the recognition region. The recognition regions include the question number region, question stem region, answer region, annotation region, and score region. Regional confidence can characterize the accuracy of the region division; the higher the regional confidence, the higher the division accuracy. Field recognition confidence can be understood as the recognition accuracy of the data in the corresponding recognition region; the higher the field recognition confidence, the higher the recognition accuracy of the data in the corresponding recognition region. The anonymization status can be determined based on the authorized data of the offline learning materials and the degree of completion of the anonymization processing on the fields representing the data source.

[0022] In this embodiment of the disclosure, the feasibility of using structured learning evidence as subsequent data processing is determined by the confidence level of data identification, and the ability to integrate offline learning materials with the internal resources of the education platform is determined by the anonymization status, thereby ensuring the certainty of multiple structured objects and their corresponding object structured parameters identified based on structured learning evidence.

[0023] Furthermore, optionally, after splitting offline learning materials into multiple structured objects, the method further includes: determining the duplication risk of the structured objects in the target question bank; if the duplication risk is less than a preset duplication risk threshold, then the structured objects are stored in the target question bank. The target question bank can be understood as exercises categorized by question type according to the teaching content by the educational platform. Duplication risk can be understood as the degree of duplication with structured objects of the same question type in the target question bank. This disclosure verifies the duplication risk between the anonymized structured objects and the exercises in the target question bank, thereby avoiding duplication of exercises in the target question bank.

[0024] S120. Based on the object structuring parameters of the structured object, determine the first answer process diagram of the structured object; the first answer process diagram is used to describe the sequence of answering steps of the structured object.

[0025] Specifically, the answer data of the structured object is segmented at the row level, step level, or formula level according to the corresponding question type to obtain multiple answer steps. The answer steps are then arranged in the order of answering to form the first answer process used to describe the answer step sequence of the structured object.

[0026] S130. Determine the second answer process diagram of the structured object, and determine the answer difference results based on the first and second answer process diagrams of the structured object; the second answer process diagram is the standard answer step sequence of the structured object provided by the education platform.

[0027] Specifically, the first and second answer process diagrams of the structured object are compared to determine whether there are any discrepancies between the answer step sequence corresponding to the structured object and the standard answer step sequence, and these discrepancies are identified as the answer difference results. Discrepancies can be conflicting steps, missing steps, or erroneous steps.

[0028] S140. Based on the answer difference results and the corresponding historical review data, generate target review data, and update the learning progress record of the education platform based on the target review data; the historical review data includes the learning progress of the question types corresponding to the structured objects in the education platform; the target review data is used to describe the learning strategies for the question types corresponding to the structured objects.

[0029] Specifically, based on the differences in responses, error types and error cause confidence levels are determined, where error cause confidence levels describe the accuracy of the response difference results. Based on the error types and error cause confidence levels, a first review progress for structured objects is generated. Based on the first review progress and corresponding historical review data, a second review progress for the question types corresponding to the structured objects is determined, and target review data is generated based on the second review progress. Accordingly, generating target review data based on the second review progress may include: different review progresses have pre-defined correlations with different learning strategies; target review data can be determined from these pre-defined correlations based on the second review progress, and then the learning progress profile on the educational platform is updated based on the target review data.

[0030] Optionally, in this embodiment of the disclosure, the object's structured parameters further include annotation data and score data. Generating target review data based on the answer difference results and corresponding historical review data may include: determining the score confidence level of the structured object based on its annotation data, score data, and answer difference results; the score confidence level describes the accuracy of the structured object's score data; if the score confidence level is greater than a preset score confidence level, then target review data is generated based on the answer difference results and corresponding historical review data. If the score confidence level is less than the preset score confidence level, then the structured object corresponding to the score confidence level is manually verified, as well as the accuracy of the structured object and its bound object structured parameters is verified. Once the verification result confirms the data is accurate, target review data is generated based on the answer difference results and corresponding historical review data. In this embodiment of the disclosure, the scoring confidence level used to describe the accuracy of the scoring data of the structured object is accurately determined by the annotation data, scoring data and answer difference results of the structured object. Then, based on the scoring confidence level and the preset scoring confidence level, it is determined whether the answer difference results and the corresponding historical review data can be used to generate target review data, so as to ensure the accuracy of the target review data.

[0031] Optionally, in this embodiment of the disclosure, the object's structured parameters further include annotation data and scoring data. Generating target review data based on answer discrepancies and corresponding historical review data may include: determining annotation conflict results based on the annotation data of the structured object and platform annotation data; the platform annotation data is the annotation data obtained by the education platform from correcting the answer data of the structured object; determining scoring conflict results based on the scoring data of the structured object and platform scoring data; the platform scoring data is the scoring data after the education platform corrects the answer data of the structured object; if the annotation conflict results and scoring conflict results meet preset conditions, then target review data is generated based on the answer discrepancies and corresponding historical review data. If either the annotation conflict result or the scoring conflict result does not meet the preset conditions, then the corresponding structured object is manually verified, as well as the accuracy of the structured object and its bound object structured parameters is verified. Once the verification result confirms that the data is accurate, target review data is generated based on the answer discrepancies and corresponding historical review data.

[0032] Specifically, determining whether annotation conflict results and rating conflict results meet preset conditions may include: determining the confidence level of annotation conflict based on the annotation conflict results, determining the confidence level of rating conflict based on the rating conflict results, and if the confidence level of annotation conflict is less than a preset annotation conflict threshold and the confidence level of rating conflict is less than a preset rating conflict threshold, then the annotation conflict results and rating conflict results are determined to meet the preset conditions; if the confidence level of annotation conflict is greater than or equal to the preset annotation conflict threshold and / or the confidence level of rating conflict is greater than or equal to the preset rating conflict threshold, then the annotation conflict results and rating conflict results are determined not to meet the preset conditions.

[0033] In this embodiment of the disclosure, annotation conflict results are determined based on the annotation data of structured objects and the platform annotation data, and scoring conflict results are determined based on the scoring data of structured objects and the platform scoring data. Furthermore, by judging whether both the annotation conflict results and the scoring conflict results meet preset conditions, it is determined whether the answer difference results and the corresponding historical review data can be used to generate target review data, thereby ensuring the accuracy of the target review data.

[0034] Optionally, generating target review data based on the answer difference results and corresponding historical review data may further include: if the score confidence is greater than the preset score confidence, and the annotation conflict results and score conflict results meet the preset conditions, then target review data is generated based on the answer difference results and corresponding historical review data. This ensures the accuracy of the target review data more precisely by simultaneously determining whether target review data can be generated based on the answer difference results and corresponding historical review data, using the score confidence, annotation conflict results, and score conflict results.

[0035] The technical solution of this disclosure involves acquiring offline learning materials uploaded to an education platform and splitting these materials into multiple structured objects. Each structured object represents a question; each structured object is bound to corresponding object structured parameters; these parameters include question type, question stem data, and answer data. This achieves precise splitting of the offline learning materials, facilitating accurate processing of each structured object in the future. Then, based on the object structured parameters, a first answer process diagram describing the answer step sequence of the structured object is determined, converting the object structured parameters of each question into a data format that allows for faster comparison later. Furthermore, a second answer process diagram is determined for each structured object, and the difference in answer results is determined by comparing the first and second answer process diagrams. This enables precise analysis of the differences between the standard answer step sequence provided by the education platform and the answer step sequence of the offline learning materials, thereby improving the education platform's data perception capabilities regarding the offline learning process. Correspondingly, based on the differences in answers and the corresponding historical review data, target review data is generated to describe the learning strategies for the question types corresponding to the structured objects. This is because the historical review data includes the learning progress of the question types corresponding to the structured objects in the education platform. In other words, the target review data generated based on the differences in answers and the corresponding historical review data can not only cover the education platform's own review strategies, but also combine the learning situation of offline learning materials. This can better provide users with accurate review strategies. Furthermore, the learning growth profile of the education platform can be updated based on the target review data to improve the conversion efficiency of the individual review loop and also make up for the education platform's data perception ability of the offline learning process.

[0036] Figure 2 This is a flowchart illustrating another method for processing offline learning materials provided in this embodiment. The technical solution of this embodiment further optimizes the process of processing offline learning materials based on the above embodiments. This embodiment can be combined with various optional solutions in one or more of the above embodiments. Figure 2 As shown, the processing methods for this publicly available offline learning material include: S210. Obtain the offline learning materials uploaded to the education platform and split the offline learning materials into multiple structured objects; each structured object represents a question; each structured object is bound to corresponding object structured parameters; the object structured parameters include question type, question stem data, and answer data.

[0037] S220. Based on the object structuring parameters of the structured object, determine the first answer process diagram of the structured object; the first answer process diagram is used to describe the sequence of answering steps of the structured object.

[0038] S230. Determine the second answer process diagram of the structured object, and determine the answer difference results based on the first and second answer process diagrams of the structured object; the second answer process diagram is the standard answer step sequence of the structured object provided by the education platform.

[0039] S240. Based on the answer difference results and the corresponding historical review data, generate target review data; the historical review data also includes parameters representing the mastery level of the question types corresponding to the structured objects; the answer difference results also include parameters representing the mastery level of the structured objects; the target review data is used to describe the learning strategies for the question types corresponding to the structured objects.

[0040] Specifically, generating target review data based on the differences in responses and the corresponding historical review data may include steps A1-A4: Step A1: Generate reference review data based on the differences in responses; the reference review data is used to describe the learning strategies of structured objects.

[0041] Learning strategies can be understood as plans for the learning process of structured objects designed to improve learning efficiency and effectiveness. For example, learning strategies may include correcting mistakes, re-practicing, reviewing explanations, supplementing methods, and doing variation problems, as well as recommended rules for variation problems, problems with similar mechanisms, or basic problems.

[0042] Specifically, the different responses can reflect the degree of mastery of the structured object. Different degrees of mastery correspond to different learning strategies. Therefore, the first degree of mastery of the structured object is determined based on the response differences, and reference review data is generated based on the first degree of mastery.

[0043] Step A2: Determine the confidence level of the structured object's score loss based on the difference in responses. The confidence level of the score loss is used to describe the degree of score loss of the structured object.

[0044] Step A3: Based on the confidence level of the points lost and historical review data, determine the review priority; the review priority is used to characterize the urgency of the question types corresponding to the structured objects being reviewed.

[0045] Specifically, based on historical review data, determine the second level of mastery of the question types corresponding to the structured objects, and determine the review priority based on the second level of mastery and the confidence level of lost points.

[0046] Among them, review priority This can be expressed using the following formula: ; in, r 1. r 2. r 3. r 4 andr 5 represents the corresponding weighting coefficients; E 1 indicates the confidence level of the score loss; E 2 represents the frequency of recurrence and interval decay of error types for the corresponding question types of structured objects on the education platform within a preset time period; E 3 indicates the risk of forgetting; E 4 indicates the weight of the question type corresponding to the structured object in the overall teaching and training objectives; E 5 indicates the second level of mastery.

[0047] Step A4: Generate target review data based on review priorities and reference review data.

[0048] Specifically, the review priorities for different question types are associated with corresponding learning strategy adjustment plans. Therefore, the learning strategy adjustment plan for the question type corresponding to the structured object is determined based on the review priority, and the reference review data is adjusted based on the learning strategy adjustment plan to generate target review data.

[0049] In this embodiment, reference review data for describing learning strategies for structured objects is generated based on the answer difference results. Since the answer difference results also include parameters characterizing the mastery level of the structured objects, the confidence level of the structured objects used to describe the degree of point deduction can be accurately determined based on the answer difference results. Since the historical review data also includes parameters characterizing the mastery level of the question types corresponding to the structured objects, the review priority characterizing the urgency of reviewing the question types corresponding to the structured objects can be more accurately determined based on the confidence level of point deduction and the historical review data. Thus, by combining the review priority and the reference review data, target review data is accurately generated.

[0050] Optionally, generating target review data based on review priority and reference review data may include: if the review priority is greater than or equal to a preset priority, then multiple reference review data for the question type of the structured object are aggregated into target review data. This disclosure determines whether it is necessary to aggregate multiple reference review data for the question type of the structured object into target review data based on the judgment result between the review priority and the preset priority, thus forming a thematic review task and achieving targeted review of questions of that question type. Further, generating target review data based on review priority and reference review data may include: if the review priority is greater than or equal to the preset priority, and the number of recurrences of the same type of error on the education platform within a preset time period reaches a preset number, then multiple reference review data for the question type of the structured object are aggregated into target review data. The multiple reference review data for the question type of the structured object can be obtained from a target question bank. This disclosure further combines the judgment result between the review priority and the preset priority with the number of recurrences of the same type of error, thereby enabling a more accurate distinction as to whether targeted review of questions of the corresponding question type of the structured object is necessary.

[0051] S250. Update the learning and growth records of the education platform based on the target review data.

[0052] Optionally, in this embodiment of the disclosure, the learning growth portfolio includes the change range of the growth portfolio corresponding to the target review data; the change range of the growth portfolio is determined based on parameters characterizing the changes in the learning status of the target review data; the method further includes: determining the target training weight, which is used to describe the training objective of the question type corresponding to the target review data; determining the service priority of each target review data based on the target training weight, the change range of the growth portfolio, and the review priority; updating the service strategy of all target review data in the learning growth portfolio according to all service priorities; the service strategy is used to describe the priority of distributing the target review data.

[0053] The learning status of the target review data can be understood as the degree of mastery of the target review data.

[0054] Among them, service priority It can be expressed by the following formula: ; in, b 1. b 2 and b 3 represent the corresponding weighting coefficients; The extent of changes in the growth record; Develop weights for the target.

[0055] This embodiment combines the target training weight, the magnitude of changes in the learning portfolio, and the review priority to determine the service priority of each target review data, ensuring the accuracy of the service priority, thereby ensuring the accuracy of the service strategy for updating all target review data in the learning portfolio according to all service priorities.

[0056] Furthermore, determining the service priority of each target review data based on the target training weight, the change range of the growth record, and the review priority can also include: determining the task load based on the issued target review data; and determining the service priority of each target review data based on the task load, target training weight, change range of the growth record, and review priority.

[0057] Among them, service priority It can be expressed by the following formula: ; in, b 1. b 2 and b 3 and b 4 represents the corresponding weight coefficients; For task load.

[0058] This embodiment further introduces task load, combining target training weight, growth record change range and review priority to determine the service priority of each target review data, thereby further ensuring the accuracy of the service priority of each target review data.

[0059] The technical solution of this disclosure involves acquiring offline learning materials uploaded to an education platform and splitting these materials into multiple structured objects. Each structured object represents a question; each structured object is bound to corresponding object structured parameters; these parameters include question type, question stem data, and answer data. This achieves precise splitting of the offline learning materials, facilitating accurate processing of each structured object in the future. Then, based on the object structured parameters, a first answer process diagram describing the answer step sequence of the structured object is determined, converting the object structured parameters of each question into a data format that allows for faster comparison later. Furthermore, a second answer process diagram is determined for each structured object, and the difference in answer results is determined by comparing the first and second answer process diagrams. This enables precise analysis of the differences between the standard answer step sequence provided by the education platform and the answer step sequence of the offline learning materials, thereby improving the education platform's data perception capabilities regarding the offline learning process. Since historical review data also includes parameters representing the mastery level of the question types corresponding to the structured objects, and the answer difference results also include parameters representing the mastery level of the structured objects, more accurate target review data can be generated based on the answer difference results and the corresponding historical review data. This allows the platform to not only cover its own review strategies but also combine the learning situation of offline learning materials, thus providing users with more precise review strategies. Furthermore, the learning growth profile of the education platform can be updated based on the target review data to improve the conversion efficiency of the individual review loop and also supplement the education platform's data perception capabilities of the offline learning process.

[0060] Figure 3 This is a flowchart illustrating another method for processing offline learning materials provided in this embodiment. The technical solution of this embodiment further optimizes the process of processing offline learning materials based on the above embodiments. This embodiment can be combined with various optional solutions in one or more of the above embodiments. Figure 3 As shown, the processing methods for this publicly available offline learning material include: S310. Obtain the offline learning materials uploaded to the education platform and split the offline learning materials into multiple structured objects; each structured object represents a question; each structured object is bound to corresponding object structured parameters; the object structured parameters include question type, question stem data, and answer data.

[0061] Optionally, after splitting the structured learning evidence into multiple structured objects, and before determining the first answer process diagram of the structured object based on the object structured parameters of the structured object, the method further includes: determining the object credibility of the structured object; the object credibility is used to describe the attribution precision of the object structured parameters bound to the structured object; if the object credibility is greater than or equal to a preset object credibility, the structured object corresponding to the object credibility is retained; if the object credibility is less than the preset object credibility, the structured object corresponding to the object credibility is manually verified.

[0062] Specifically, the structured parameters of the object also include annotation data and score data. Determining the object credibility of a structured object can include: determining the first degree of matching between the structured object and the corresponding question stem data, the second degree of matching between the structured object and the corresponding answer data, the third degree of matching between the structured object and the corresponding annotation data, and the fourth degree of matching between the structured object and the corresponding score data. Based on the first, second, third, and fourth degree of matching, the object credibility of the structured object is determined. Further, if the object credibility is greater than or equal to the preset object credibility, it means that the structured object corresponding to the object credibility can be introduced into the educational platform as internal learning material, and therefore the structured object corresponding to the object credibility is retained. If the object credibility is less than the preset object credibility, it means that the structured object corresponding to the object credibility does not yet meet the requirements for introduction into the educational platform as internal learning material. Therefore, the structured object corresponding to the object credibility is manually verified. If the verification result is confirmed, the structured object corresponding to the object credibility can continue to be retained.

[0063] This disclosure embodiment filters structured objects by assessing their object credibility to ensure the accuracy of structured objects introduced into the education platform, thereby ensuring the accuracy of the education platform's data perception of the offline learning process.

[0064] S320. Based on the question type of the structured object, determine the structured data mapping method of the structured object; the structured data mapping method is used to describe the data transformation method of converting the structured parameters of the object into ordered answer nodes and / or answer edges; based on the structured data mapping method, convert the structured parameters of the structured object into the first answer process diagram.

[0065] The question types can be problem-solving questions, proof questions, or calculation questions.

[0066] Specifically, if the question type is a problem-solving question, the structured data mapping method can be to construct a step tree for the problem-solving question, which includes answer nodes and answer edges. If the question type is a proof question, the structured data mapping method can be to construct proof obligation edges for the proof question, which can be understood as answer edges. If the question type is a calculation question, the structured data mapping method can be to construct derivation nodes for key formulas, which can be understood as answer nodes.

[0067] S330. Determine the second answer process diagram of the structured object, and determine the answer difference results based on the first and second answer process diagrams of the structured object; the second answer process diagram is the standard answer step sequence of the structured object provided by the education platform.

[0068] Specifically, determining the answer difference results based on the first and second answer process diagrams of the structured object may include: determining whether there are conflicting nodes, missing nodes, and erroneous nodes between the first and second answer process diagrams to obtain a first matching result; determining whether there are conflicting edges, missing edges, and erroneous edges between the first and second answer process diagrams to obtain a second matching result; and using the first matching result and / or the second matching result as the answer difference results of the structured object.

[0069] S340. Based on the answer difference results and the corresponding historical review data, generate target review data, and update the learning progress record of the education platform based on the target review data; the historical review data includes the learning progress of the question types corresponding to the structured objects in the education platform; the target review data is used to describe the learning strategies for the question types corresponding to the structured objects.

[0070] Optionally, historical review data also includes learning strategies on the education platform for the question types corresponding to the target review data. Updating the learning progress profile on the education platform based on the target review data may include: determining the data weight based on the confidence level of structured objects; the data weight is used to describe the importance of the offline learning materials corresponding to the structured objects; determining the profile conflict level based on the target review data and historical review data; the profile conflict level is used to describe the deviation between the target review data and the historical learning plan of the education platform; if the data weight is greater than or equal to the preset data weight, and the profile conflict level is less than the preset conflict level, then the learning progress profile on the education platform is updated based on the target review data. If the data weight is less than the preset data weight, then the target review data is manually verified. Furthermore, the method also includes: if the data weight is greater than or equal to the preset data weight, and the profile conflict degree is greater than or equal to the preset conflict degree, then the target review data is marked as pending verification, and a preset round of retesting tasks is issued; if the retesting task is detected to be completed and the retesting results are obtained, and if the retesting results meet the retesting requirements, then the learning and growth profile of the education platform is updated based on the target review data.

[0071] Among them, data weights This can be expressed using the following formula: ; in, a 1. a 2. a 3 and a 4 represents the corresponding weight coefficients; Q 1 indicates the credibility of the source, which can be determined based on the credibility level of the uploading entity corresponding to the offline learning materials, the completeness of the source, and the authorization data; Q 2 indicates recognition quality, which can be determined based on data recognition confidence level, or based on data recognition confidence level, field missing rate, and manual sampling pass rate. Q 3 indicates scoring consistency, which can be determined based on the difference between the score data and the corresponding AI score result, or based on the difference between the score data and the corresponding AI score result, the consistency rate of the same question review, and the pass rate of the manual review. Q 4 indicates consistency with the platform's historical performance, which can be determined based on the degree of conflict among multiple profiles corresponding to structured objects within a preset time period.

[0072] Furthermore, if the target review data is reviewed subsequently, it will be marked as reviewed upon completion. Therefore, when the target review data is detected as reviewed, the review result of the target review data is obtained. The review verification intensity corresponding to the target review data is determined based on the review result, the similarity of review questions, and the review interval decay factor. Further, the data weight is determined based on the data recognition confidence of structured objects and the review verification intensity. And based on the data weight and the current profile conflict degree, it is determined whether to update the learning growth profile of the education platform based on the target review data.

[0073] Among them, data weights This can be expressed using the following formula: ; in, a 1. a 2. a 3. a 4 and a 5 represents the corresponding weighting coefficients;Q 5 indicates the intensity of subsequent review and verification.

[0074] In this embodiment, the ability to update the learning and growth profile of the education platform based on the target review data is determined by the weight of the data and the degree of conflict in the profile. This ensures the accurate updating of the learning and growth profile and avoids the inaccuracy of the target review data introduced from outside affecting the accuracy of the learning and growth profile within the education platform.

[0075] The technical solution of this disclosure involves acquiring offline learning materials uploaded to an education platform and splitting these materials into multiple structured objects. Each structured object represents a question; each structured object is bound to corresponding object structured parameters; these parameters include question type, question stem data, and answer data. This achieves precise splitting of the offline learning materials, facilitating accurate processing of each structured object in the future. Furthermore, based on the question type of each structured object, a structured data mapping method is determined, enabling precise matching of different question types to corresponding structured data mapping methods. This accurately transforms the object structured parameters of the structured objects into ordered answer nodes and / or answer edges, forming a more accurate first answer process diagram. Then, by comparing the first and second answer process diagrams of the structured objects, the answer difference results are determined. This allows for precise analysis of the differences between the standard answer step sequence provided by the education platform and the answer step sequence of the offline learning materials, thereby improving the education platform's data perception capabilities regarding the offline learning process. Correspondingly, based on the differences in answers and the corresponding historical review data, target review data is generated to describe the learning strategies for the question types corresponding to the structured objects. This is because the historical review data includes the learning progress of the question types corresponding to the structured objects in the education platform. In other words, the target review data generated based on the differences in answers and the corresponding historical review data can not only cover the education platform's own review strategies, but also combine the learning situation of offline learning materials. This can better provide users with accurate review strategies. Furthermore, the learning growth profile of the education platform can be updated based on the target review data to improve the conversion efficiency of the individual review loop and also make up for the education platform's data perception ability of the offline learning process.

[0076] Figure 4 This is a schematic diagram of a device for processing offline learning materials according to an embodiment of this disclosure. This embodiment is applicable to situations where learning materials from outside the education platform are introduced into the education platform for processing. The device for processing offline learning materials can be implemented in hardware and / or software, and can be configured in any electronic device with network communication capabilities. Figure 4 As shown, the device for processing offline learning materials disclosed herein includes: The first data processing module 410 is used to acquire offline learning materials uploaded to the education platform and split the offline learning materials into multiple structured objects; each structured object represents a question; each structured object is bound to corresponding object structured parameters; the object structured parameters include question type, question stem data, and answer data; The second data processing module 420 is used to determine a first answer process diagram of the structured object based on the object structure parameters of the structured object; the first answer process diagram is used to describe the answering step sequence of the structured object; The difference determination module 430 is used to determine the second answer process diagram of the structured object, and to determine the answer difference result based on the first answer process diagram and the second answer process diagram of the structured object; the second answer process diagram is the standard answer step sequence of the structured object provided by the education platform; The update module 440 is used to generate target review data based on the answer difference results and the corresponding historical review data, and update the learning progress record of the education platform based on the target review data; the historical review data includes the learning progress of the question type corresponding to the structured object in the education platform; the target review data is used to describe the learning strategy for the question type corresponding to the structured object.

[0077] Based on the above embodiments, optionally, the first data processing module is used to extract data from the offline learning materials to obtain structured learning evidence; the structured learning evidence includes a question number area, a question stem area, an answer area, an annotation area, and a score area; based on the structured learning evidence, multiple structured objects and corresponding object structured parameters are determined.

[0078] Based on the above embodiments, optionally, the structured learning evidence further includes data identification confidence and desensitization status, wherein the desensitization status is used to describe whether the structured learning evidence has been authorized and the degree of data desensitization; the first data processing module further includes a data judgment unit, which is used to determine multiple structured objects and corresponding object structured parameters based on the structured learning evidence if the data identification confidence is greater than or equal to a preset identification confidence and the desensitization status is desensitized.

[0079] Based on the above embodiments, optionally, the offline learning material processing device further includes an update module. The update module is used to determine the object credibility of the structured object after the structured learning evidence is split into multiple structured objects and before the first answer process diagram of the structured object is determined based on the object structured parameters of the structured object. The object credibility is used to describe the attribution precision of the object structured parameters bound to the structured object. If the object credibility is greater than or equal to a preset object credibility, the structured object corresponding to the object credibility is retained. If the object credibility is less than the preset object credibility, the structured object corresponding to the object credibility is manually verified.

[0080] Based on the above embodiments, optionally, the second data processing module is used to determine the structured data mapping method of the structured object based on the question type of the structured object; the structured data mapping method is used to describe the data transformation method of converting the object's structured parameters into ordered answer nodes and / or answer edges; based on the structured data mapping method, the object's structured parameters are converted into a first answer process diagram.

[0081] Based on the above embodiments, optionally, the difference determination module is used to determine whether there are conflicting nodes, missing nodes, and erroneous nodes between the first answer process diagram and the second answer process diagram, and obtain a first matching result; determine whether there are conflicting edges, missing edges, and erroneous edges between the first answer process diagram and the second answer process diagram, and obtain a second matching result; and use the first matching result and / or the second matching result as the answer difference result of the structured object.

[0082] Based on the above embodiments, optionally, the object structured parameters further include annotation data and score data, and the update module includes a first judgment unit, which is used to: determine the scoring confidence of the structured object based on the annotation data, the score data and the answer difference results of the structured object; the scoring confidence is used to describe the accuracy of the score data of the structured object; if the scoring confidence is greater than a preset scoring confidence, then generate target review data based on the answer difference results and the corresponding historical review data.

[0083] Optionally, based on the above embodiments, the object structured parameters further include annotation data and score data. The update module includes a second judgment unit, which is used to determine the annotation conflict result based on the annotation data of the structured object and the platform annotation data. The platform annotation data is the annotation data obtained by the education platform in correcting the answer data of the structured object. Based on the score data of the structured object and the platform score data, a scoring conflict result is determined. The platform score data is the score data after the education platform corrects the answer data of the structured object. If the annotation conflict result and the scoring conflict result meet preset conditions, target review data is generated based on the answer difference result and the corresponding historical review data.

[0084] Based on the above embodiments, optionally, the historical review data further includes parameters characterizing the mastery level of the question type corresponding to the structured object; the answer difference results further include parameters characterizing the mastery level of the structured object; the update module includes a review data generation unit, which is used to generate reference review data based on the answer difference results; the reference review data is used to describe the learning strategy of the structured object; the confidence level of the lost points of the structured object is determined based on the answer difference results, and the confidence level of the lost points is used to describe the degree of lost points of the structured object; the review priority is determined based on the confidence level of the lost points and the historical review data; the review priority is used to characterize the urgency of reviewing the question type corresponding to the structured object; and target review data is generated based on the review priority and the reference review data.

[0085] Optionally, based on the above embodiments, the review data generation unit is further configured to aggregate multiple reference review data of the question type of the structured object into target review data if the review priority is greater than or equal to a preset priority.

[0086] Optionally, based on the above embodiments, the historical review data also includes learning strategies on the education platform for the question types corresponding to the target review data. The update module includes a file update judgment unit, which is used to determine the data weight based on the data identification confidence level of the structured object; the data weight is used to describe the importance of the offline learning materials corresponding to the structured object; based on the target review data and historical review data, the profile conflict degree is determined; the profile conflict degree is used to describe the deviation between the target review data and the historical learning plan of the education platform; if the data weight is greater than or equal to the preset data weight, and the profile conflict degree is less than the preset conflict degree, then the learning growth file of the education platform is updated based on the target review data.

[0087] Based on the above embodiments, optionally, the file update judgment unit is further configured to: if the data weight is greater than or equal to the preset data weight, and the profile conflict degree is greater than or equal to the preset conflict degree, then mark the target review data as pending verification and issue a preset round of retesting tasks; if it is detected that the retesting task has been completed and the retesting result has been obtained, and if the retesting result meets the retesting requirements, then update the learning growth file of the education platform based on the target review data.

[0088] Based on the above embodiments, optionally, the learning growth portfolio includes the change range of the growth portfolio corresponding to the target review data; the change range of the growth portfolio is determined according to parameters characterizing the changes in the learning status of the target review data, and the offline learning material processing device further includes a service strategy update module, which is used to determine the target training weight, the target training weight being used to describe the training objective of the question type corresponding to the target review data; based on the target training weight, the change range of the growth portfolio, and the review priority, the service priority of each target review data is determined; according to all the service priorities, the service strategy of all target review data in the learning growth portfolio is updated; the service strategy is used to describe the priority of distributing the target review data.

[0089] Optionally, based on the above embodiments, the service strategy update module is further configured to: determine the task load based on the issued target review data; and determine the service priority of each target review data based on the task load, the target training weight, the growth record change range, and the review priority.

[0090] Based on the above embodiments, optionally, the offline learning material processing device further includes a storage module, which is used to: after splitting the offline learning material into multiple structured objects, determine the duplication risk of the structured objects in the target question bank; if the duplication risk is less than a preset duplication risk threshold, then store the structured objects in the target question bank.

[0091] The offline learning material processing apparatus provided in this disclosure can execute the offline learning material processing method provided in any embodiment of this disclosure, and has the corresponding functional modules and beneficial effects for executing the method.

[0092] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0093] Figure 5A schematic diagram of an electronic device is shown that can be used to implement the offline learning material processing method of embodiments of this disclosure. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the disclosure described and / or claimed herein.

[0094] like Figure 5 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0095] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0096] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as methods for processing offline learning materials.

[0097] In some embodiments, the method for processing offline learning materials may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via read-only memory (ROM) 12 and / or communication unit 19. When the computer program is loaded into random access memory (RAM) 13 and executed by processor 11, one or more steps of the method for processing offline learning materials described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the method for processing offline learning materials by any other suitable means (e.g., by means of firmware).

[0098] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0099] Computer programs used to implement the methods of this disclosure may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0100] In the context of this disclosure, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium can be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0101] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0102] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0103] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0104] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this disclosure can be achieved, and this is not limited herein.

[0105] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for processing offline learning materials, characterized in that, The method includes: The system retrieves offline learning materials uploaded to an education platform and breaks them down into multiple structured objects. Each structured object represents a question. Each structured object is bound to corresponding object structured parameters, which include question type, question stem data, and answer data. Based on the object structuring parameters of the structured object, a first answer process diagram of the structured object is determined; the first answer process diagram is used to describe the sequence of answering steps of the structured object. A second answer process diagram of the structured object is determined, and the answer difference results are determined based on the first and second answer process diagrams of the structured object; the second answer process diagram is the standard answer step sequence of the structured object provided by the education platform; Based on the differences in the answers and the corresponding historical review data, target review data is generated, and the learning progress record of the education platform is updated based on the target review data; the historical review data includes the learning progress of the question type corresponding to the structured object in the education platform; the target review data is used to describe the learning strategy for the question type corresponding to the structured object.

2. The method according to claim 1, characterized in that, The step of splitting the offline learning materials into multiple structured objects includes: Data is extracted from the offline learning materials to obtain structured learning evidence; the structured learning evidence includes a question number area, a question stem area, a response area, an annotation area, and a score area. Based on the structured learning evidence, multiple structured objects and their corresponding object structured parameters are determined.

3. The method according to claim 2, characterized in that, The structured learning evidence also includes data identification confidence and desensitization status, wherein the desensitization status is used to describe whether the structured learning evidence has been authorized and the degree of data desensitization; The step of determining multiple structured objects and their corresponding object structure parameters based on the structured learning evidence includes: If the data identification confidence level is greater than or equal to the preset identification confidence level, and the de-identification status is de-identified, then based on the structured learning evidence, multiple structured objects and corresponding object structured parameters are determined.

4. The method according to claim 1, characterized in that, The process of determining the first answer process diagram of the structured object based on the object's structured parameters includes: Based on the question type of the structured object, the structured data mapping method of the structured object is determined; the structured data mapping method is used to describe the data transformation method of converting the structured parameters of the object into ordered answer nodes and / or answer edges; Based on the structured data mapping method, the object structured parameters of the structured object are transformed into the first answer process diagram.

5. The method according to claim 4, characterized in that, The determination of answer difference results based on the first and second answer process diagrams of the structured object includes: Determine whether there are conflicting nodes, missing nodes, and erroneous nodes between the first answer process diagram and the second answer process diagram, and obtain the first matching result; Determine whether there are conflicting edges, missing edges, and erroneous edges between the first answer process diagram and the second answer process diagram, and obtain the second matching result; The first matching result and / or the second matching result are used as the response difference results of the structured object.

6. The method according to claim 1, characterized in that, The historical review data also includes parameters characterizing the mastery level of the question types corresponding to the structured objects; the answer difference results also include parameters characterizing the mastery level of the structured objects. The generation of target review data based on the differences in responses and the corresponding historical review data includes: Reference review data is generated based on the differences in the responses; the reference review data is used to describe the learning strategy of the structured object; The confidence level of the structured object's score loss is determined based on the results of the answer differences. The confidence level of the score loss is used to describe the degree of score loss of the structured object. Based on the confidence level of the points lost and the historical review data, a review priority is determined; the review priority is used to characterize the urgency of reviewing the question types corresponding to the structured objects. Based on the review priority and the reference review data, target review data is generated.

7. The method according to claim 3, characterized in that, The historical review data also includes learning strategies on the education platform for the question types corresponding to the target review data. Updating the learning progress record on the education platform based on the target review data includes: Based on the data identification confidence level of the structured object, the data weight is determined; the data weight is used to describe the importance of the offline learning materials corresponding to the structured object. Based on the target review data and historical review data, the degree of profile conflict is determined; the degree of profile conflict is used to describe the deviation between the target review data and the historical learning plan of the education platform. If the weight of the data is greater than or equal to the preset weight of the data, and the degree of conflict in the profile is less than the preset degree of conflict, then the learning and growth profile of the education platform is updated based on the target review data.

8. A device for processing offline learning materials, characterized in that, The device includes: The first data processing module is used to acquire offline learning materials uploaded to the education platform and split the offline learning materials into multiple structured objects; each structured object represents a question; each structured object is bound to corresponding object structured parameters; the object structured parameters include question type, question stem data, and answer data; The second data processing module is used to determine a first answer process diagram of the structured object based on the object structure parameters of the structured object; the first answer process diagram is used to describe the answering step sequence of the structured object; The difference determination module is used to determine the second answer process diagram of the structured object, and to determine the answer difference result based on the first answer process diagram and the second answer process diagram of the structured object; the second answer process diagram is the standard answer step sequence of the structured object provided by the education platform; The update module is used to generate target review data based on the answer difference results and the corresponding historical review data, and update the learning progress record of the education platform based on the target review data; the historical review data includes the learning progress of the question type corresponding to the structured object in the education platform; the target review data is used to describe the learning strategy for the question type corresponding to the structured object.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the method for processing offline learning materials according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the method for processing offline learning materials according to any one of claims 1-7.

Images