A visual data interaction method based on distance education

Abstract

The application discloses a visual data interaction method based on distance education and relates to the technical field of distance education data processing, and comprises the following steps: a distance education platform reads a standard analysis node sequence of a current knowledge point and student answer records, analyzes the student answer records into a plurality of answer action records with time sequence relationship, compares input objects, processing objects and output objects of the answer action records with the standard analysis nodes node by node, substitutes output objects corresponding to candidate deviated analysis nodes into subsequent analysis nodes to perform a carrying check, and forms a deviated analysis node record.The data received by the teacher end in the application is no longer limited to whether the answer result is correct, but can correspond to an analysis node sequence, a knowledge calling field and a subsequent carrying state, so that subsequent preposition path retrieval and interaction record generation are facilitated.

Description

Technical Field

[0001] This invention relates to the field of distance education data processing technology, specifically a visual data interaction method based on distance education. Background Technology

[0002] Distance education platforms are typically used for course resource publishing, online student responses, learning behavior recording, remote teacher tutoring, and feedback on learning outcomes. During distance education, the platform can record students' answers, submission times, resource access behavior, practice responses, and interactive feedback behavior for the current knowledge point, and organize the student's learning process based on course content configuration data and question analysis templates. For course tasks that involve multiple analysis steps and dependencies on prerequisite knowledge points, students' errors in answering the current knowledge point may not necessarily originate solely from the current analysis step itself, but may also be related to insufficient understanding of prerequisite knowledge points, interrupted access to prerequisite knowledge fragments, incomplete learning behavior, and knowledge connection segments that still deviate after historical remediation. Existing data processing methods for distance education can typically grade students' answers, record the steps involved in answering, and display the student's error location or result in the current question to the teacher. These methods can meet the needs of routine online grading and teaching feedback, and can also provide teachers with basic data to understand students' current learning status. However, in learning tasks with multi-level knowledge dependencies, simply providing prompts based on the surface error of the current knowledge point makes it difficult to clearly distinguish the relationship between errors in the current parsing node, failures in subsequent parsing nodes, obstruction of previous knowledge points, and new downstream deviations after remediation. The interactive basis obtained by the teacher is likely to remain at the level of the current error step. Therefore, in the context of distance education, there is a need for a data interaction method that can uniformly process the current knowledge point answering process, standard parsing node sequence, preceding knowledge point connection segments, learning behavior data, and state changes after interaction execution. This would enable the platform to further determine the preceding obstruction segment corresponding to the deviation when the transmission of the current knowledge point deviates, and to link the teacher's interaction records with the subsequent answer update results, thereby providing a more complete data basis for the teacher's visual interaction. Therefore, this invention proposes a visual data interaction method based on distance education. Summary of the Invention

[0003] The purpose of this invention is to provide a visual data interaction method based on distance education to solve the problems mentioned in the background art.

[0004] This invention can be achieved through the following technical solution: a visual data interaction method based on distance education, comprising: Step 1: The remote education platform reads the standard parsing node sequence and student answer records of the current knowledge point, parses the student answer records into multiple answer action records with temporal relationship, compares the input object, processing object and output object of the answer action record with the standard parsing node node by node, and substitutes the output object corresponding to the candidate deviation parsing node into the subsequent parsing node to perform the inheritance verification, forming the transmission deviation parsing node record; Step 2: Based on the knowledge call field in the transmission deviation resolution node record, the remote education platform retrieves the prerequisite knowledge point connection segment in the course knowledge dependency table, and prunes it through dependency continuity verification and historical remediation verification results to form a candidate prerequisite transmission path set. The candidate prerequisite transmission path set is written into the upstream and downstream dependency relationship between the prerequisite knowledge point connection segments. Step 3: The remote education platform reads the learning behavior data of the candidate prerequisite knowledge points along the candidate prerequisite transmission path set, maps the learning behavior data to the corresponding knowledge segment identifier, and forms a candidate prerequisite obstruction segment record based on the continuity and overlap of the learning behavior data on the knowledge segment identifier. Step 4: The remote education platform calculates the path acceptance component, behavior overlap component, and historical verification component based on the candidate pre-blocking segment record and the candidate pre-transmission path set, and generates the pre-blocking transmission quantity by weighting. The target pre-blocking segment record is determined based on the ranking of the pre-blocking transmission quantity and the upstream and downstream dependencies. Step 5: The remote education platform generates a visual interaction record for the teacher based on the records of the target pre-blockage segments, the transmission deviation analysis nodes, and the candidate pre-transmission paths where the target pre-blockage segments are located, and updates the corresponding interaction status based on the visual interaction record for the teacher.

[0005] A further technical improvement of the present invention is that the acceptance verification in step one includes: The remote education platform reads the output object corresponding to the candidate deviation parsing node and converts the output object into input parameters that can be read by subsequent parsing nodes according to the preset node input format; The remote education platform selects at least two consecutive subsequent parsing nodes according to the arrangement order of the standard parsing node sequence, substitutes the input parameters into the selected consecutive subsequent parsing nodes in sequence, and calls the node processing relationship of each consecutive subsequent parsing node to perform calculation replay, thereby obtaining the replay calculation result and replay processing relationship order corresponding to each consecutive subsequent parsing node. The remote education platform compares the replay calculation results with the node output results of the corresponding consecutive subsequent parsing nodes, and compares the order of replay processing relationships with the order of node processing relationships in the standard parsing node sequence. When the replay calculation results of each consecutive subsequent parsing node are consistent with the output results of the corresponding node, and the replay processing order is consistent with the node processing order in the standard parsing node sequence, the candidate deviating parsing node is not written into the propagated deviating parsing node record. When at least one of the following conditions is met: the replay calculation result is inconsistent with the output result of the corresponding node, or the replay processing order is inconsistent with the node processing order in the standard parsing node sequence, the corresponding consecutive subsequent parsing node is determined as the failure to accept the task, and the candidate deviation parsing node and the failure to accept the task are written into the propagation deviation parsing node record.

[0006] A further technical improvement of the present invention lies in: a method for obtaining historical remedial verification results, comprising: The remote education platform reads the historical transmission deviation analysis node records corresponding to the connection segments of the preceding knowledge points in the candidate preceding transmission path set, and sorts them according to the occurrence time of the obstruction handling task to form a historical obstruction handling sequence. The remote education platform sequentially reads the pre-treatment conduction deviation analysis node record and the post-treatment re-identified conduction deviation analysis node record for each historical obstruction treatment record in the historical obstruction treatment sequence. Historical obstruction treatment records that do not contain the pre-treatment conduction deviation analysis node in the post-treatment re-identified conduction deviation analysis node record are included in the historical release record. Historical obstruction treatment records that still contain the pre-treatment conduction deviation analysis node in the post-treatment re-identified conduction deviation analysis node record are included in the historical continuity record. Historical obstruction treatment records that contain newly added conduction deviation analysis nodes located downstream of the pre-treatment conduction deviation analysis node in the post-treatment re-identified conduction deviation analysis node record are included in the historical downstream migration record. The remote education platform generates a time decay coefficient based on the time interval between the occurrence time of the obstruction handling task and the current processing time for each historical obstruction handling record. The number of historical unblocking records, historical continuous records, and historical downstream migration records are multiplied by the corresponding time decay coefficient to form the decayed unblocking amount, decayed continuous amount, and decayed migration amount. The remote education platform calculates the historical verification value of the connection segment based on the attenuation removal amount, attenuation duration amount, and attenuation migration amount, and writes the historical verification value of the connection segment into the historical remediation verification result.

[0007] A further technical improvement of the present invention lies in the method of forming candidate pre-arrest fragment records, including: The remote education platform reads the knowledge fragment identifier, behavior type field and behavior occurrence sequence corresponding to the learning behavior data under the same candidate prerequisite knowledge point, and forms a standard knowledge traversal path within the candidate prerequisite knowledge point based on the course knowledge dependency table. The distance education platform forms an actual knowledge fragment access sequence according to the order of behavior occurrence, and compares the actual knowledge fragment access sequence with the standard knowledge traversal path position by position to determine the path fork point; When none of the consecutive knowledge fragment identifiers after the path fork point fall into the standard knowledge traversal path, and the number of consecutive deviations reaches the preset path deviation length threshold, the corresponding knowledge fragment identifiers will be excluded from the continuous behavior record. When a knowledge fragment identifier after a path fork point falls back into the standard knowledge traversal path, the knowledge fragment identifier that falls back into the path will be included in the continuous behavior record. The remote education platform compares the behavior type field of adjacent learning behavior data with the preset behavior succession relationship. When the behavior type field meets the preset behavior succession relationship and the knowledge fragment identifiers are adjacent, a continuous behavior record is formed. The remote education platform reads the overlapping hit results of multiple learning behaviors of each knowledge segment identifier in the continuous behavior record. When multiple learning behaviors overlap and the previous knowledge segment identifier is included in the continuous behavior record but the next knowledge segment identifier is not included in the continuous behavior record, the current knowledge segment identifier is written into the candidate pre-blocking segment record.

[0008] A further technical improvement of the present invention is that the overlapping hit results of multiple learning behaviors are formed in the following way: The remote education platform reads multiple learning behavior data corresponding to the same knowledge segment identifier in the continuous behavior record and the knowledge call field in the transmission deviation parsing node record, and reads the behavior start time, end time, behavior belonging segment boundary and behavior associated knowledge call field of each learning behavior data; The remote education platform compares the boundaries of the behavior-related fragments with the boundaries of the knowledge fragments, compares the overlap of the start and end times of the behavior in terms of time intervals, and compares the consistency of the behavior-related knowledge call fields with the knowledge call fields in the transmission deviation parsing node records. When the boundaries of at least two types of learning behavior data fall into the same knowledge segment boundary, the behavior time intervals overlap and the fields are consistent, the corresponding knowledge segment identifier is written into the multi-type learning behavior overlap hit result; the remote education platform determines the corresponding knowledge segment identifier as the knowledge segment identifier that meets the overlap judgment based on the multi-type learning behavior overlap hit result, and uses it to form a candidate pre-blocking segment record.

[0009] A further technical improvement of the present invention is that the target pre-blocking segment record is determined in the following manner: The remote education platform reads the number of times the candidate pre-blocking fragment records are repeated in the records of multiple current knowledge point transmission deviation parsing nodes, and forms the blockage stability based on the number of repetitions; The distance education platform determines the relative position between the candidate preceding blocking fragment record and the path bifurcation point based on the order of the knowledge fragment identifier corresponding to the candidate preceding blocking fragment record in the standard knowledge traversal path. When the difference in pre-block conduction between adjacent candidate pre-block segments is lower than a preset conduction difference threshold, the remote education platform reads the block stability and relative position of the corresponding candidate pre-block segment record. When the candidate pre-blocking segment record is located upstream of the path bifurcation point and the blocking stability reaches the preset blocking stability threshold, the path carrying component weight and the historical verification component weight of the corresponding candidate pre-blocking segment record are multiplied by the preset gain coefficient respectively. If any of the following conditions are met: the candidate preceding blocking segment record is located downstream of the path bifurcation point, or the blocking stability is lower than the preset blocking stability threshold, the weight of the path receiving component of the corresponding candidate preceding blocking segment record will be multiplied by the preset weight reduction coefficient. The remote education platform recalculates the pre-blocking transmission volume based on the adjusted path carrying component weight, behavior overlap component weight, and historical verification component weight, and determines the target pre-blocking segment record based on the recalculated pre-blocking transmission volume.

[0010] A further technical improvement of the present invention is that: step four determines the target pre-blocking segment record based on the pre-blocking conduction volume ranking and upstream-downstream dependency, including: The distance education platform sorts candidate pre-blocking segment records based on pre-blocking conductance, forming a conductance sorting sequence; The remote education platform reads candidate pre-blocking segment records sequentially according to the conduction volume sorting sequence, and determines whether the currently read candidate pre-blocking segment record is located in the upstream path segment of subsequent candidate pre-blocking segment records based on the upstream and downstream dependencies in the candidate pre-conduction path set. If the currently read candidate pre-blocking segment record is located in the upstream path segment of the subsequent candidate pre-blocking segment record, and the sorting position of the currently read candidate pre-blocking segment record in the conductance sorting sequence is not later than that of the subsequent candidate pre-blocking segment record, then the subsequent candidate pre-blocking segment record will be written into the masking segment set. The remote education platform counts the number of downstream occlusion segments corresponding to candidate preceding blocking segments that are not written into the occlusion segment set, and forms the path coverage order. In the candidate preceding blocking segment records that have not been written into the occlusion segment set, the remote education platform determines the candidate preceding blocking segment record with the highest path coverage order as the target preceding blocking segment record. When multiple candidate anterior blockage fragment records have the same path coverage order, the candidate anterior blockage fragment record with the highest order in the conductance sorting sequence is determined as the target anterior blockage fragment record.

[0011] A further technical improvement of the present invention is that the interaction state is updated in the following way: After the remote education platform generates the corresponding interactive execution data corresponding to the visual interactive record on the teacher's end, it reads the current knowledge point answer record after the interactive execution, and re-forms and transmits the deviated parsing node record according to the standard parsing node sequence. If the updated conduction deviation resolution node record still contains the conduction deviation resolution node corresponding to the teacher's visual interaction record, the teacher's visual interaction record will be written to the continuous blockage state. When the updated transmission deviation analysis node record does not contain the transmission deviation analysis node corresponding to the teacher's visual interaction record, but contains a newly added transmission deviation analysis node located downstream of the transmission deviation analysis node, the teacher's visual interaction record is written to the newly added downstream deviation state, and the newly added transmission deviation analysis node is written to the new transmission deviation analysis node record, so that the new transmission deviation analysis node record re-enters the process of forming the candidate preceding transmission path set. When the updated transmission deviation resolution node record does not contain the transmission deviation resolution node corresponding to the teacher's visual interaction record, and does not contain any newly added transmission deviation resolution node located downstream of that transmission deviation resolution node, the teacher's visual interaction record will be written to the completely deactivated state.

[0012] Compared with the prior art, the present invention has the following beneficial effects: This invention compares the input, processing, and output objects during a student's answer process with standard parsing nodes, and combines this with the verification results of subsequent parsing nodes to form a record of deviation parsing nodes. This allows the remote education platform to obtain the correspondence between deviation nodes and failed connection nodes in the current knowledge point. As a result, the data received by the teacher is no longer limited to whether the answer is correct, but can be mapped to the parsing node order, knowledge call fields, and subsequent connection status, facilitating subsequent pre-path retrieval and interaction record generation. Furthermore, by introducing historical remedial verification results, learning behavior continuity, learning behavior overlap, and upstream and downstream dependencies into the candidate pre-transmission path set, this invention enables the determination process of pre-blocking segments to simultaneously combine the knowledge path structure, historical handling results, and students' actual learning behaviors. As a result, the target pre-blocking segment record can form a more stable association with the current transmission deviation parsing node record, reducing the offset caused when the blockage position is determined solely based on a single answer error or a single learning behavior. On the other hand, this invention rereads the current knowledge point answer record after the interaction is executed after the teacher's visual interaction record is generated, and writes the updated transmission deviation analysis node record into the continuous blockage state, the newly added downstream deviation state, or the completely released state, so that the teacher's interaction result can continue to enter the subsequent data processing process; thus, the remote education platform can form a closed data chain from deviation identification, pre-blockage location, teacher's visual interaction to interaction state update, and improve the correspondence between remote teaching intervention records and changes in student answers. Attached Figure Description

[0013] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0014] Figure 1 This is a schematic diagram of the method logic of the present invention. Detailed Implementation

[0015] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided.

[0016] Please see Figure 1 As shown, this invention provides a visual data interaction method based on distance education, comprising: Step 1: The remote education platform reads the standard parsing node sequence and student answer records of the current knowledge point, parses the student answer records into multiple answer action records with temporal relationship, compares the input object, processing object and output object of the answer action record with the standard parsing node node by node, and substitutes the output object corresponding to the candidate deviation parsing node into the subsequent parsing node to perform the inheritance verification, forming the transmission deviation parsing node record; Specifically, in one implementation, before performing deviation identification for the current knowledge point, the distance education platform first establishes a standard parsing node sequence for the current knowledge point. The platform reads the current knowledge point identifier, question identifier, and knowledge call field from the course content configuration data, and then reads the parsing step field, standard input field, standard processing field, standard output field, and parsing step sequence field from the question parsing template. The platform reads the parsing step sequence field item by item according to its order, and generates a standard parsing node for each field. For each generated standard parsing node, the platform writes a node identifier, node sequence, node input conditions, node processing relationship, node output result, and knowledge call field. The node sequence is written incrementally according to the parsing step sequence field; the node input conditions are written jointly by the standard input field and the node output result of the preceding standard parsing node; the node processing relationship is written by the standard processing field; the node output result is written by the standard output field; and the knowledge call field is written jointly by the knowledge call field in the course content configuration data and the knowledge call field of the corresponding parsing step field in the question parsing template. After all standard parsing nodes are generated, the remote education platform arranges the standard parsing nodes in ascending order of node position to form a standard parsing node sequence, and binds and stores the standard parsing node sequence with the current knowledge point identifier, question identifier, and question parsing template identifier.

[0017] After reading the student's answer record, the distance education platform segments the record according to the submission time and modification records. The answer content between adjacent submission times is written into an action record. When modification records exist for the same answer content, the platform retains the last modified answer and writes the original answer content into the action change field. The platform writes an action identifier, action sequence, input object, processing object, and output object into each action record. The input object is formed by the question stem conditions referenced in the action record, the output object of the previous action record, and the intermediate quantities already written by the student. The processing object is extracted from the action record by the platform according to preset action parsing rules, which are formed by the standard processing field, operator field, derivation relationship field, condition judgment field, variable substitution field, and unit conversion field in the question parsing template. The output object is formed by the numerical result, symbolic result, variable result, unit result, and conclusion field at the end of the action record. The distance education platform arranges the response action records in ascending order of the action occurrence sequence, so that the response action records can be compared node by node with the standard parsing node sequence.

[0018] When the remote education platform performs node-by-node comparison, it first reads the response action record with the action occurrence sequence number one and the standard parsing node with the node sequence number one. The platform then breaks down the input objects of the response action record into input field name, input field value, input field unit, and input field source. It further breaks down the node input conditions of the standard parsing node into node input field name, node input field value range, node input field unit, and node input field source. When the input field names match, the input field values ​​fall within the node input field value range, the input field units match, and the input field sources match, the corresponding result for the input object is written. The platform then breaks down the processing objects of the response action record into processing relationship type, processing relationship parameters, processing relationship order, and processing relationship belonging field. Finally, it breaks down the node processing relationships of the standard parsing node into node processing relationship type, node processing relationship parameters, node processing relationship order, and node processing relationship belonging field. When the processing relationship type, processing relationship parameters, processing relationship order, and processing relationship belonging field match, the corresponding result for the processing object is written. The distance education platform breaks down the output objects of the answer action record into output field names, output field values, output field units, and output field sign direction. It then breaks down the node output results of the standard parsing node into node output field names, node output field values, node output field units, and node output field sign direction. The determination of consistent output field values ​​includes matching numerical values, consistent expression structure after processing by preset expression simplification rules, matching interval endpoint values, and matching values ​​of the same ordinal position in a vector. When the output field names, values, units, and sign directions are consistent, the corresponding result is written to the output object.

[0019] When the input object, processing object, and output object all correspond to a valid result, the remote education platform writes the action identifier of the answer action record into the passed record of the corresponding standard parsing node, and reads the next answer action record and the next standard parsing node to continue performing node-by-node comparison. When the input object corresponds to a valid result but the processing object corresponds to a valid result, the remote education platform writes the current standard parsing node into the candidate deviation parsing node. When the input object corresponds to a valid result, the processing object corresponds to a valid result, but the output object corresponds to a valid result, the remote education platform writes the current standard parsing node into the candidate deviation parsing node. When the input object corresponds to a valid result but not a valid result, the remote education platform determines whether the current answer action record has a previous answer action record and whether the current standard parsing node has a previous standard parsing node. If at least one of the following conditions is met: the current answer action record does not have a previous answer action record, or the current standard parsing node does not have a previous standard parsing node, the remote education platform writes the current standard parsing node into the candidate deviation parsing node. If both the current answer action record and the current standard parsing node have a previous answer action record, the remote education platform performs a pre-order comparison between the output object of the previous answer action record and the node input conditions of the current standard parsing node. If the pre-order comparison fails, the previous standard parsing node is written into the candidate deviation parsing node. If the pre-order comparison succeeds, the current standard parsing node is written into the candidate deviation parsing node. The remote education platform generates a temporary candidate record for the candidate deviation parsing node. The temporary candidate record contains the node identifier, node order, action identifier of the corresponding answer action record, output object, knowledge call field, input object corresponding result, processing object corresponding result, and output object corresponding result of the candidate deviation parsing node.

[0020] Before performing acceptance verification, the remote education platform first reads the output object corresponding to the candidate deviation parsing node and then reads the subsequent parsing nodes after the candidate deviation parsing node. When there are fewer than two remaining subsequent parsing nodes after the candidate deviation parsing node, the remote education platform does not perform computation replay verification of consecutive subsequent parsing nodes, but writes the candidate deviation parsing node into the propagation deviation parsing node record and writes the end node mark into the propagation deviation parsing node record. The propagation deviation parsing node record with the end node mark includes the node identifier, node order, corresponding output object, knowledge call field, end node mark, and end verification endpoint order of the candidate deviation parsing node. The end verification endpoint order is written into the node order of the candidate deviation parsing node. When there are at least two remaining subsequent parsing nodes after the candidate deviation parsing node, the remote education platform reads at least two consecutive subsequent parsing nodes from the candidate deviation parsing node according to the node order of the standard parsing node sequence, and writes the read subsequent parsing nodes into the consecutive subsequent parsing node set in ascending order of node order. Each consecutive subsequent parsing node in the consecutive subsequent parsing node set retains its node input conditions, node processing relationship, node output result, and node order for subsequent computation replay calls.

[0021] To ensure a predefined node input format, the distance education platform generates a node input format table simultaneously when establishing the standard parsing node sequence. This table is generated from the node input conditions of the standard parsing nodes, with each condition corresponding to a node input format item. Each node input format item includes the field name, data type, unit, sign direction, value structure, and source. The data type can be selected from numeric, symbolic, text, relational, or structured types; the value structure can be selected from single value, range value, vector value, or expression value. The distance education platform binds and stores the node input format table with the node identifiers of the standard parsing nodes. During succession verification, it reads the corresponding node input format table based on the node identifiers of subsequent parsing nodes and uses that table as the predefined node input format.

[0022] The distance education platform converts the output objects corresponding to candidate deviation parsing nodes into input parameters readable by subsequent parsing nodes according to a preset node input format. The platform first reads the field names, data types, units, sign directions, and value structures of the output object, and then matches each item with the node input format items of subsequent parsing nodes. If the field names, data types, units, sign directions, and value structures are consistent, the platform directly writes the output object into the input parameter table of the subsequent parsing nodes. If the field data types are inconsistent but a corresponding field conversion relationship exists in the node input format table, the platform converts the output object to the data type required by the node input format table according to the field conversion relationship before writing it into the input parameter table. If the field units are inconsistent but a corresponding unit conversion relationship exists in the node input format table, the platform converts the output object to the unit required by the node input format table according to the unit conversion relationship before writing it into the input parameter table. If at least one of the following conditions is met: inconsistent field names, inconsistent field symbol direction, inconsistent field value structure, inconsistent field data types and no field conversion relationship, or inconsistent field units and no unit conversion relationship, the remote education platform will not write the output object into the input parameter table and will write an input parameter mismatch marker into the temporary candidate record.

[0023] When the input parameter mismatch flag is written into the temporary candidate record, the remote education platform reads the consecutive subsequent parsing node with the smallest node order in the consecutive subsequent parsing node set and identifies it as the failed succession node. The remote education platform generates a propagation deviation parsing node record and writes the node identifier, node order, output object, knowledge call field, input parameter mismatch flag, node identifier, and node order of the failed succession node into the same propagation deviation parsing node record. The remote education platform no longer performs calculation replay on the consecutive subsequent parsing node set.

[0024] If a temporary candidate record is not marked as not corresponding to the input parameters, the remote education platform substitutes the input parameters into the subsequent parsing nodes and performs replay calculations according to the node order of the subsequent parsing nodes. The platform first reads the subsequent parsing node with the smallest node order, reads its input parameter table and node processing relationships, and sequentially calls the operation relationships, derivation relationships, transformation relationships, and judgment relationships according to the processing relationship order recorded in the node processing relationships, forming the replay calculation result for that subsequent parsing node. The platform writes this replay calculation result into the input parameter table of the next subsequent parsing node and performs replay calculations in the same way, calling the node processing relationships of the next subsequent parsing node until all subsequent parsing nodes in the set have completed their replay calculations. During the replay calculation of each subsequent parsing node, the platform records the actual called processing relationship identifier and call order, forming the replay processing relationship order according to the call order.

[0025] The remote education platform compares the replay calculation result with the node output results of the corresponding subsequent parsing nodes. The platform reads the output field name, value, unit, and sign direction from the replay calculation result, and also reads the node output field name, value, unit, and sign direction from the node output result. The determination of consistent output field values ​​includes the following comparison results: equal numerical values, consistent expression structure after processing by preset expression simplification rules, consistent interval endpoint values, and consistent values ​​of the same position in the vector. When the output field name, value, unit, and sign direction are consistent, a replay calculation result consistency flag is written; when at least one of the following is true: inconsistent output field name, value, unit, or sign direction, a replay calculation result inconsistency flag is written.

[0026] The distance education platform compares the replay processing order with the node processing order in the standard parsing node sequence. The platform reads the processing relationship identifier and call order from the replay processing order, and also reads the node processing relationship identifier and standard call order of the corresponding subsequent parsing nodes in the standard parsing node sequence. When both the processing relationship identifier and call order are consistent, a replay processing order consistency flag is written. When at least one of the following conditions is met—inconsistent processing relationship identifier or inconsistent call order—a replay processing order inconsistency flag is written.

[0027] When each consecutive subsequent parsing node in the set of consecutive subsequent parsing nodes has a replay calculation result consistency flag and a replay processing relationship order consistency flag, the remote education platform will not write the candidate deviation parsing node into the propagation deviation parsing node record, but will write the acceptance pass flag into the temporary candidate record. The acceptance pass flag is only used as a temporary filtering result for this acceptance verification and does not participate in the formation of the candidate preceding propagation path set.

[0028] When at least one of the following flags—inconsistent replay calculation results or inconsistent replay processing order—is written into the set of consecutive subsequent parsing nodes, the distance education platform identifies the consecutive subsequent parsing nodes with the corresponding flags as failed succession nodes. The distance education platform generates a propagation deviation parsing node record and writes the node identifier of the candidate deviation parsing node, the node order of the candidate deviation parsing node, the output object corresponding to the candidate deviation parsing node, the knowledge call field corresponding to the candidate deviation parsing node, the node identifier of the failed succession node, the node order of the failed succession node, the replay calculation result, the node output result, the replay processing order, and the standard processing order into the same propagation deviation parsing node record.

[0029] After the transmission deviation resolution node record is formed, the distance education platform binds and stores the transmission deviation resolution node record with the current knowledge point, student answer record, standard resolution node sequence, and candidate deviation resolution nodes. When forming the candidate preceding transmission path set, the distance education platform reads the knowledge call field in the transmission deviation resolution node record and retrieves the preceding knowledge point connection segment in the course knowledge dependency table. When there is a node sequence position of a failed node in the transmission deviation resolution node record, the distance education platform reads the node sequence position of the failed node and uses it as the endpoint position of dependency continuity verification. When there is an end node mark in the transmission deviation resolution node record, the distance education platform reads the end verification endpoint sequence position and uses it as the endpoint position of dependency continuity verification. Thus, a continuous technical action chain is formed between the standard resolution node sequence generation, student answer record parsing, candidate deviation resolution node formation, acceptance verification execution, transmission deviation resolution node record writing, and candidate preceding transmission path set formation.

[0030] Step 2: Based on the knowledge call field in the transmission deviation resolution node record, the remote education platform retrieves the prerequisite knowledge point connection segment in the course knowledge dependency table, and prunes it through dependency continuity verification and historical remediation verification results to form a candidate prerequisite transmission path set. The candidate prerequisite transmission path set is written into the upstream and downstream dependency relationship between the prerequisite knowledge point connection segments. Specifically, in one implementation, the remote education platform reads the knowledge call field and the current knowledge point identifier from the propagation deviation resolution node record; when the propagation deviation resolution node record contains a node sequence number of a failed successor node, it reads the node sequence number of the failed successor node; when the propagation deviation resolution node record contains an end node marker, it reads the end verification endpoint sequence number and retrieves the preceding knowledge point connection segment corresponding to the knowledge call field from the course knowledge dependency relationship table. The course knowledge dependency relationship table is generated from course content configuration data and knowledge point relationship configuration data. Each table record includes a preceding knowledge point connection segment identifier, an upstream knowledge point identifier, a downstream knowledge point identifier, a dependency connection field, a connection segment start knowledge fragment identifier, a connection segment end knowledge fragment identifier, and a connection segment direction field. The remote education platform writes the preceding knowledge point connection segments corresponding to the dependency connection field and the knowledge call field in the propagation deviation resolution node record into an initial preceding knowledge point connection segment set.

[0031] The remote education platform performs dependency continuity verification on the initial set of prerequisite knowledge point connection segments. The platform reads each prerequisite knowledge point connection segment from the initial set, recursively reads adjacent prerequisite knowledge point connection segments according to the connection segment direction field, and compares the termination knowledge fragment identifier of the current prerequisite knowledge point connection segment with the starting knowledge fragment identifier of the adjacent prerequisite knowledge point connection segment. If the termination knowledge fragment identifier of the current prerequisite knowledge point connection segment matches the starting knowledge fragment identifier of the adjacent prerequisite knowledge point connection segment, and the dependency connection field of the adjacent prerequisite knowledge point connection segment can continue to point to the knowledge call field in the propagation deviation parsing node record, the remote education platform writes the adjacent prerequisite knowledge point connection segment into the dependency continuity connection segment record. The platform continues to recursively read the connection segment direction field until the adjacent prerequisite knowledge point connection segment can no longer point to the knowledge call field, and then arranges the continuously recursively propagated prerequisite knowledge point connection segments into candidate prerequisite propagation paths according to the reading order.

[0032] After the candidate preceding transmission paths are formed, the remote education platform reads the historical transmission deviation analysis node records corresponding to each preceding knowledge point connection segment in the candidate preceding transmission path set. The historical transmission deviation analysis node records include the historical current knowledge point identifier, the historical transmission deviation analysis node identifier, the historical knowledge call field, the historical failed connection node identifier, the historical failed connection node sequence, the preceding knowledge point connection segment identifier, and the obstruction handling task identifier. Using the preceding knowledge point connection segment identifier as the search key, the remote education platform reads historical transmission deviation analysis node records with the same preceding knowledge point connection segment identifier and writes the read historical transmission deviation analysis node records into the connection segment history record set.

[0033] The remote education platform reads the occurrence time of the obstruction handling task corresponding to each historical conduction deviation parsing node record in the connection segment history record set, and sorts them according to the occurrence time of the obstruction handling task to form a historical obstruction handling sequence. Each historical obstruction handling record in the historical obstruction handling sequence retains the obstruction handling task identifier, the obstruction handling task occurrence time, the conduction deviation parsing node record before handling, and the conduction deviation parsing node record re-identified after handling. The conduction deviation parsing node record before handling is the conduction deviation parsing node record corresponding to the obstruction handling task before it is generated; the conduction deviation parsing node record re-identified after handling is formed after the obstruction handling task is executed, and the remote education platform performs a connection verification on the new response action record according to the standard parsing node sequence.

[0034] The remote education platform sequentially reads the pre-treatment conduction deviation analysis node record and the post-treatment re-identified conduction deviation analysis node record for each historical obstruction treatment record in the historical obstruction treatment sequence, and reads the pre-treatment conduction deviation analysis node identifier in the pre-treatment conduction deviation analysis node record. If the post-treatment re-identified conduction deviation analysis node record does not contain a pre-treatment conduction deviation analysis node, the remote education platform includes the corresponding historical obstruction treatment record in the historical release record; if the post-treatment re-identified conduction deviation analysis node record still contains a pre-treatment conduction deviation analysis node, the remote education platform includes the corresponding historical obstruction treatment record in the historical continuity record; if the post-treatment re-identified conduction deviation analysis node record contains a newly added conduction deviation analysis node downstream of the pre-treatment conduction deviation analysis node, the remote education platform includes the corresponding historical obstruction treatment record in the historical downstream migration record. Whether a newly added conduction deviation analysis node is downstream of the pre-treatment conduction deviation analysis node is determined by the node position in the standard analysis node sequence; when the node position of the newly added conduction deviation analysis node is greater than the node position of the pre-treatment conduction deviation analysis node, it is determined that the newly added conduction deviation analysis node is downstream of the pre-treatment conduction deviation analysis node.

[0035] The distance education platform reads the occurrence time and current processing time of each historical obstruction handling record, calculates the time interval between the current processing time and the occurrence time of the obstruction handling task, and reads the course task cycle duration. The course task cycle duration is formed by the course start time, course end time, and the number of task issuances, specifically by dividing the duration between the course end time and course start time by the number of task issuances. The distance education platform generates a time decay coefficient, which is equal to the course task cycle duration divided by the sum of the course task cycle duration and the time interval, and binds each historical obstruction handling record to the corresponding time decay coefficient.

[0036] The distance education platform reads the time decay coefficient corresponding to each historical release record and sums the time decay coefficients corresponding to all historical release records to form the decay release amount; it reads the time decay coefficient corresponding to each historical persistence record and sums the time decay coefficients corresponding to all historical persistence records to form the decay persistence amount; it reads the time decay coefficient corresponding to each historical downstream migration record and sums the time decay coefficients corresponding to all historical downstream migration records to form the decay migration amount. The distance education platform sums the decay release amount, decay persistence amount, and decay migration amount to form the total decay.

[0037] The distance education platform calculates the historical verification value of a connection segment based on the attenuation removal amount, attenuation duration, attenuation migration amount, and total attenuation. When the total attenuation is greater than zero, the platform divides the attenuation removal amount by the total attenuation to obtain the historical verification value of the connection segment; when the total attenuation is equal to zero, the platform marks the connection segment for that prerequisite knowledge point as unverified. The platform then writes the historical verification value of the connection segment into the historical remedial verification results. The historical remedial verification results are stored as an index using the prerequisite knowledge point connection segment identifier. Each index entry includes the prerequisite knowledge point connection segment identifier, the number of historical removal records, the number of historical duration records, the number of historical downstream migration records, the attenuation removal amount, the attenuation duration, the attenuation migration amount, the total attenuation, the historical verification value of the connection segment, and the unverified flag.

[0038] When the remote education platform performs historical remedial verification result pruning on the candidate preceding transmission path set, it reads the historical remedial verification result of each preceding knowledge point connection segment in the candidate preceding transmission path. If a preceding knowledge point connection segment corresponds to an unverified flag, the remote education platform retains the preceding knowledge point connection segment for subsequent path verification. If a preceding knowledge point connection segment has a historical verification value, and an adjacent downstream preceding knowledge point connection segment also has a historical verification value, the remote education platform reads the historical verification value of the preceding knowledge point connection segment and the historical verification value of the adjacent downstream preceding knowledge point connection segment, and performs a comparison. If the historical verification value of the preceding knowledge point connection segment is lower than that of the adjacent downstream preceding knowledge point connection segment, the remote education platform deletes the preceding knowledge point connection segment from the candidate preceding transmission path set; if the historical verification value of the preceding knowledge point connection segment is not lower than that of the adjacent downstream preceding knowledge point connection segment, the remote education platform retains the preceding knowledge point connection segment. If at least one of the following conditions is met: the preceding knowledge point connection segment does not have a historical verification value, or the adjacent downstream preceding knowledge point connection segment does not have a historical verification value, the remote education platform will not perform a comparison of the historical verification values ​​of the preceding knowledge point connection segment with the adjacent downstream preceding knowledge point connection segment, and will retain the preceding knowledge point connection segment.

[0039] After completing dependency continuity verification and historical remediation verification pruning, the remote education platform writes the retained prerequisite knowledge point connection segments into the candidate prerequisite propagation path set according to upstream and downstream dependencies. Each path record in the candidate prerequisite propagation path set includes a path identifier, current knowledge point identifier, propagation deviation resolution node record identifier, prerequisite knowledge point connection segment identifier, historical verification value of the connection segment, upstream and downstream dependencies, and path order.

[0040] Step 3: The remote education platform reads the learning behavior data of the candidate prerequisite knowledge points along the candidate prerequisite transmission path set, maps the learning behavior data to the corresponding knowledge segment identifier, and forms a candidate prerequisite obstruction segment record based on the continuity and overlap of the learning behavior data on the knowledge segment identifier. Specifically, in one implementation, after forming a set of candidate prerequisite transmission paths, the remote education platform sequentially reads the learning behavior data of candidate prerequisite knowledge points along the set of candidate prerequisite transmission paths. Each piece of learning behavior data includes a student identifier, a candidate prerequisite knowledge point identifier, a behavior occurrence sequence, a behavior type field, a behavior start time, a behavior end time, a behavior resource location, a behavior-related knowledge call field, and a behavior result field. The behavior occurrence sequence is written in ascending order according to the behavior start time of the same student within the same candidate prerequisite knowledge point; the behavior type field includes resource access type, practice answer type, interactive question type, and page dwell type; the behavior result field is used to record the resource access completion result, practice answer result, interactive question attribution result, and page dwell completion result.

[0041] The distance education platform reads the knowledge fragment configuration table corresponding to the candidate prerequisite knowledge points. The knowledge fragment configuration table is generated from course content configuration data. Each knowledge fragment configuration record includes a knowledge fragment identifier, a candidate prerequisite knowledge point identifier, a knowledge fragment start boundary, a knowledge fragment end boundary, a knowledge fragment sequence, a fragment knowledge call field, and a fragment resource location range. The knowledge fragment start boundary and knowledge fragment end boundary are jointly determined by the course resource annotation location, the parsing step boundary in the question parsing template, and the knowledge call field. The course resource annotation location determines the resource access range, the parsing step boundary determines the exercise answer range, and the knowledge call field determines the knowledge relationship range corresponding to the knowledge fragment. The distance education platform compares the behavior resource location in the learning behavior data with the fragment resource location range in the knowledge fragment configuration record. When the behavior resource location falls within the fragment resource location range of a certain knowledge fragment configuration record, the learning behavior data is mapped to the knowledge fragment identifier corresponding to that knowledge fragment configuration record.

[0042] The distance education platform reads the knowledge fragment identifier, behavior type field, and behavior occurrence sequence corresponding to the learning behavior data under the same candidate prerequisite knowledge point, and forms a standard knowledge traversal path within that candidate prerequisite knowledge point based on the course knowledge dependency table. The course knowledge dependency table records the upstream and downstream dependencies between knowledge fragment identifiers within the same candidate prerequisite knowledge point. The distance education platform uses a knowledge fragment identifier without an upstream knowledge fragment identifier as the path starting point, sequentially reads downstream knowledge fragment identifiers according to the upstream and downstream dependencies, and writes the read knowledge fragment identifiers into the standard knowledge traversal path in the reading order. Each item in the standard knowledge traversal path includes a knowledge fragment identifier, path sequence, upstream knowledge fragment identifier, and downstream knowledge fragment identifier.

[0043] The distance education platform reads the knowledge fragment identifiers mapped from the learning behavior data under the same candidate prerequisite knowledge point according to the behavior occurrence sequence, and forms an actual knowledge fragment access sequence in ascending order of behavior occurrence sequence. Each item in the actual knowledge fragment access sequence includes the behavior occurrence sequence, knowledge fragment identifier, behavior type field, behavior start time, and behavior end time. The distance education platform performs continuous repeated access merging processing on the actual knowledge fragment access sequence. The distance education platform reads two adjacent actual knowledge fragment access records according to the behavior occurrence sequence; when the knowledge fragment identifiers corresponding to the two adjacent actual knowledge fragment access records are the same, and the behavior type field corresponding to the two adjacent actual knowledge fragment access records both have valid field values, the distance education platform writes the learning behavior data identifier of the latter actual knowledge fragment access record into the same fragment behavior set of the former actual knowledge fragment access record, and removes the latter actual knowledge fragment access record from the path comparison items; when the knowledge fragment identifiers corresponding to the two adjacent actual knowledge fragment access records are different, the distance education platform retains the two adjacent actual knowledge fragment access records as path comparison items. After completing the continuous repeated access merging process, the distance education platform will form a path comparison sequence by reserving the path comparison items in ascending order of the occurrence sequence of the behavior.

[0044] The distance education platform compares the knowledge fragment identifiers in the path comparison sequence with the knowledge fragment identifiers at the same position in the standard knowledge traversal path, position by position. The first occurrence where the knowledge fragment identifier in the path comparison sequence does not match the knowledge fragment identifier at the same position in the standard knowledge traversal path is written as a path branch point. The path branch point includes the order in which the branch point behavior occurs, the actual knowledge fragment identifier, the standard knowledge fragment identifier, and the corresponding candidate preceding knowledge point identifier.

[0045] In one implementation, the preset path deviation length threshold is formed as follows: When generating a question analysis template, the remote education platform reads the standard knowledge traversal path length corresponding to the current candidate prerequisite knowledge point. The standard knowledge traversal path length is the number of knowledge segment identifiers in the standard knowledge traversal path. The remote education platform reads the path deviation detection ratio field from the course configuration data. This field is written by the course maintenance end when configuring candidate prerequisite knowledge points and is bound to the candidate prerequisite knowledge point identifier for storage. The remote education platform multiplies the standard knowledge traversal path length by the path deviation detection ratio field to obtain a path deviation reference value. The path deviation reference value is then rounded up to form a path deviation reference quantity. When the path deviation reference quantity is less than two, the remote education platform writes the preset path deviation length threshold to two. When the path deviation reference quantity is not less than two, the remote education platform writes the preset path deviation length threshold to the path deviation reference quantity. The remote education platform binds and stores the preset path deviation length threshold with the candidate prerequisite knowledge point identifier, the standard knowledge traversal path identifier, and the question analysis template identifier. After determining the path branching point, the remote education platform reads the preset path deviation length threshold bound to the candidate preceding knowledge point identifier, and then sequentially reads the knowledge fragment identifiers in the actual knowledge fragment access sequence starting from the path branching point. When the read knowledge fragment identifier does not fall into the standard knowledge traversal path, the remote education platform increments the consecutive deviation count by one. When the read knowledge fragment identifier falls into the standard knowledge traversal path, the remote education platform clears the consecutive deviation count to zero and continues to read the next knowledge fragment identifier. When the consecutive deviation count reaches the preset path deviation length threshold, the remote education platform excludes the corresponding consecutive knowledge fragment identifier from the writing range of the consecutive behavior record, and writes the path branching point identifier, the consecutive deviation start sequence, the consecutive deviation end sequence, the consecutive deviation count, and the corresponding knowledge fragment identifier into the path deviation exclusion record.

[0046] When a knowledge fragment identifier after a path fork point falls back into the standard knowledge traversal path, the distance education platform reads the path order of that knowledge fragment identifier in the standard knowledge traversal path, and also reads the path order of the standard knowledge fragment identifier corresponding to the path fork point in the standard knowledge traversal path. When the path order of the knowledge fragment identifier that falls back into the path is not earlier than the path order of the standard knowledge fragment identifier corresponding to the path fork point, the distance education platform re-establishes the writing range of continuous behavior records starting from the knowledge fragment identifier that falls back into the standard knowledge traversal path, and makes the knowledge fragment identifier that falls back into the path participate in the comparison of the adjacency relationship between the behavior type field and the knowledge fragment identifier.

[0047] In this embodiment, the preset behavior succession relationship is formed in the following way. When generating the course behavior configuration table, the remote education platform reads the set of behavior type fields in the course content configuration data and reads the learning behavior order field corresponding to each knowledge segment in the question analysis template. The remote education platform writes the previous behavior type field, the next behavior type field, and the knowledge segment type to which the succession belongs into the allowed succession order record according to the learning behavior order field. Each allowed succession order record is bound and stored with the candidate prerequisite knowledge point identifier and the knowledge segment identifier. When forming continuous behavior records, the remote education platform reads the previous behavior type field and the next behavior type field of adjacent learning behavior data and searches the course behavior configuration table to see if there is a corresponding allowed succession order record. When a corresponding allowed succession order record is found, the remote education platform determines that the behavior type field satisfies the preset behavior succession relationship. The remote education platform also reads the knowledge segment identifier corresponding to adjacent learning behavior data and compares the path order of the previous knowledge segment identifier and the next knowledge segment identifier in the standard knowledge traversal path. When the path order of the next knowledge segment identifier is equal to the path order of the previous knowledge segment identifier plus one, the remote education platform determines that the knowledge segment identifiers are adjacent. When the behavior type field satisfies the preset behavior succession relationship and the knowledge fragment identifiers are adjacent, the remote education platform will write the corresponding learning behavior data into the continuous behavior record. The continuous behavior record includes student identifier, candidate prerequisite knowledge point identifier, previous knowledge fragment identifier, next knowledge fragment identifier, previous behavior type field, next behavior type field, allowed succession order record identifier, behavior occurrence order, and path order.

[0048] When writing continuous behavior records, the distance education platform simultaneously reads the learning behavior data identifier sets corresponding to the previous knowledge segment identifier and the next knowledge segment identifier, and writes these sets into the continuous behavior record. The learning behavior data identifier set corresponding to the same knowledge segment identifier is formed by the actual knowledge segment access sequence and the behavior set of the same segment. When the distance education platform subsequently generates multiple overlapping learning behavior results, it reads back multiple learning behavior data corresponding to the same knowledge segment identifier based on the learning behavior data identifier sets in the continuous behavior record.

[0049] The distance education platform reads multiple learning behavior data corresponding to the same knowledge fragment identifier from continuous behavior records, and also reads the knowledge call field from the propagation deviation parsing node record. The platform reads the behavior start time, behavior end time, behavior resource location, and behavior-related knowledge call fields from each learning behavior data entry, and reads the behavior's belonging fragment boundary based on the behavior resource location corresponding to the learning behavior data. The behavior's belonging fragment boundary is determined by the knowledge fragment configuration record where the behavior resource location falls, including the behavior's belonging fragment start boundary and behavior's belonging fragment end boundary. The distance education platform simultaneously reads the knowledge fragment start boundary and knowledge fragment end boundary corresponding to the same knowledge fragment identifier.

[0050] The distance education platform compares the boundaries of behavior-assigned segments with the boundaries of knowledge segments. Specifically, when the start boundary of a behavior-assigned segment is no earlier than the start boundary of a knowledge segment, and the end boundary of a behavior-assigned segment is no later than the end boundary of a knowledge segment, the boundary of the learning behavior data is determined to fall within the same knowledge segment boundary. The distance education platform compares the overlap of the start and end times of at least two types of learning behavior data in terms of time intervals; when the start time of the first type of learning behavior data is no later than the end time of the second type of learning behavior data, and the start time of the second type of learning behavior data is no later than the end time of the first type of learning behavior data, the time intervals of the two types of learning behavior data are determined to overlap. The distance education platform compares the consistency of the behavior-related knowledge call field of each learning behavior data with the knowledge call field in the propagation deviation parsing node record; when the field codes of two knowledge call fields are consistent, the fields are determined to be consistent; when two knowledge call fields correspond to the same preceding knowledge point connection segment through the field mapping relationship in the course knowledge dependency relationship table, the fields are determined to be consistent.

[0051] When the boundaries of at least two types of learning behavior data fall within the same knowledge segment boundary, their time intervals overlap, and their fields are consistent, the distance education platform will write the corresponding knowledge segment identifier into the multi-type learning behavior overlap matching result. The multi-type learning behavior overlap matching result includes the knowledge segment identifier, the set of matching behavior type fields, the matching behavior data identifier, the overlapping time interval of the behavior, the boundary of the behavior's affiliated segment, and the corresponding knowledge retrieval field. Based on the multi-type learning behavior overlap matching result, the distance education platform will determine the corresponding knowledge segment identifier as the knowledge segment identifier that meets the overlap determination, and write the knowledge segment identifier that meets the overlap determination into the overlap determination record.

[0052] The distance education platform reads the results of overlapping learning behaviors from each knowledge fragment identifier in the continuous behavior record. For the current knowledge fragment identifier, the platform reads the previous and next knowledge fragment identifiers. When the current knowledge fragment identifier has overlapping results from multiple learning behaviors, the previous knowledge fragment identifier is included in the continuous behavior record, and the next knowledge fragment identifier is not included in the continuous behavior record, the platform writes the current knowledge fragment identifier into the candidate pre-blocking fragment record. The candidate pre-blocking fragment record includes the candidate pre-knowledge point identifier, knowledge fragment identifier, continuous behavior record identifier, overlapping result identifier for multiple learning behaviors, previous knowledge fragment identifier, next knowledge fragment identifier, path branching point identifier, and transmission deviation resolution node record identifier. The platform binds and stores the candidate pre-blocking fragment record with the corresponding pre-knowledge point connection segment in the candidate pre-transmission path set, enabling subsequent path carrying components, behavior overlapping components, and historical verification components to read the candidate pre-blocking fragment record for calculation.

[0053] Step 4: The remote education platform calculates the path acceptance component, behavior overlap component, and historical verification component based on the candidate pre-blocking segment record and the candidate pre-transmission path set, and generates the pre-blocking transmission quantity by weighting. The target pre-blocking segment record is determined based on the ranking of the pre-blocking transmission quantity and the upstream and downstream dependencies. Specifically, in one implementation, the distance education platform reads candidate pre-obstacle fragment records, a set of candidate pre-obstacle transmission paths, historical remedial verification results, and overlapping hit results of multiple learning behaviors. The distance education platform uses the knowledge fragment identifier in the candidate pre-obstacle fragment records as an index to read the pre-obstacle knowledge point connection segment containing that knowledge fragment identifier and the path order of that pre-obstacle knowledge point connection segment in the set of candidate pre-obstacle transmission paths.

[0054] The distance education platform calculates the path continuity component. It reads the number of connection segments between the preceding knowledge point connection segment of the candidate preceding obstruction fragment record and the current knowledge point of the conduction deviation resolution node record. It also reads the path order of the corresponding knowledge fragment identifier in the standard knowledge traversal path and the path order of the path bifurcation point. The platform determines the absolute value of the difference between the path order of the corresponding knowledge fragment identifier and the path order of the path bifurcation point as the path order difference. The platform sums the number of connection segments and the path order difference to form the path interval, and then forms the path continuity base value by dividing the path interval by one. When the preceding knowledge point connection segment is located in the candidate preceding transmission path set, and the upstream and downstream dependencies of the preceding knowledge point connection segment can continuously point to the knowledge call field in the transmission deviation parsing node record, the remote education platform writes the path bearing base value into the path bearing component; when the preceding knowledge point connection segment cannot continuously point to the knowledge call field in the transmission deviation parsing node record along the upstream and downstream dependencies, the remote education platform writes the path bearing component to zero.

[0055] The distance education platform calculates the overlap component of behaviors. It reads the overlap results of multiple learning behaviors corresponding to candidate pre-existing obstacle segments, counts the number of behavior type fields in the set of hit behavior type fields to form the number of hit types, counts the number of hit behavior data identifiers to form the number of hit behaviors, and calculates the behavior time overlap duration based on the start and end times of the behavior time overlap interval. The platform reads the number of hit types, the number of hit behaviors, and the behavior time overlap duration corresponding to each candidate pre-existing obstacle segment record under the same candidate pre-existing knowledge point, and determines the maximum value of the number of hit types, the maximum value of the number of hit behaviors, and the maximum value of the behavior time overlap duration, respectively. When the maximum value of the number of hit types is greater than zero, the platform divides the number of hit types by the maximum value of the number of hit types to form a normalized number of hit types; when the maximum value of the number of hit types is zero, the normalized number of hit types is written to zero. When the maximum number of hit behaviors is greater than zero, the remote education platform divides the total number of hit behaviors by the maximum number of hit behaviors to form a normalized number of hit behaviors; when the maximum number of hit behaviors is zero, the normalized number of hit behaviors is written to zero. When the maximum duration of overlapping behavior times is greater than zero, the remote education platform divides the duration of overlapping behavior times by the maximum duration of overlapping behavior times to form a normalized duration of overlapping behavior times; when the maximum duration of overlapping behavior times is zero, the normalized duration of overlapping behavior times is written to zero. The remote education platform sums the normalized number of hit types, the normalized number of hit behaviors, and the normalized duration of overlapping behavior times, then divides the sum by three to form the behavior overlap component.

[0056] The distance education platform calculates the historical verification component. It reads the historical remedial verification results corresponding to the connection segments of the preceding knowledge points where the candidate pre-blockage fragment records. If a historical verification value for a connection segment exists in the historical remedial verification results, the platform writes this value into the historical verification component. If an unverified marker exists in the historical remedial verification results, the platform reads the historical verification values ​​of the connection segments of other preceding knowledge points within the same candidate pre-conduction path set, averages the readable historical verification values, and writes them into the historical verification component. If no readable historical verification value exists within the same candidate pre-conduction path set, the platform writes the historical verification component into a neutral verification value, which is one divided by two.

[0057] The distance education platform reads the conductance calculation parameter table. This table is generated as follows: When generating a question analysis template, the platform reads the path continuity weight configuration field, behavior overlap weight configuration field, and historical verification weight configuration field. When the sum of these three weight configuration fields is greater than zero, the platform divides each field by its sum to form the path continuity component weight, behavior overlap component weight, and historical verification component weight. When the sum is zero, the platform writes one-third of each weight into the table. Finally, the platform binds and stores these three weights with the current knowledge point identifier and the question analysis template identifier to form the conductance calculation parameter table. The remote education platform generates the pre-blocking transmission quantity according to the following method: the pre-blocking transmission quantity is equal to the sum of the products of the path receiving component multiplied by the path receiving component weight, the behavior overlapping component multiplied by the behavior overlapping component weight, and the historical verification component multiplied by the historical verification component weight. The pre-blocking transmission quantity is then written into the candidate pre-blocking segment record.

[0058] The distance education platform sorts candidate pre-blocking segment records from largest to smallest based on the pre-blocking conductance, forming an initial conductance sorting sequence. When the initial conductance sorting sequence contains only one candidate pre-blocking segment record, the distance education platform directly identifies this candidate pre-blocking segment record as the target pre-blocking segment record and binds and stores it with the corresponding conduction deviation resolution node record. In the case where the initial conductance sorting sequence contains only one candidate pre-blocking segment record, the distance education platform does not perform adjacent conductance difference calculation, weight adjustment, or the formation of the occluded segment set.

[0059] When the initial conductance sorting sequence contains at least two candidate anterior blockage segment records, the remote education platform reads the adjacent candidate anterior blockage segment records and calculates the difference between the two adjacent anterior blockage conductances to form the adjacent conductance difference. The remote education platform reads all adjacent conductance differences in the same initial conductance sorting sequence and averages all adjacent conductance differences to form the mean conductance difference.

[0060] The "Ranking Proximity Ratio" field is written by the remote education platform when generating the question parsing template. The platform reads the "Number of Candidate Preceding Paths" and "Number of Parsing Nodes" fields from the template. When the number of candidate preceding paths is greater than the number of parsing nodes, the platform writes half of the "Ranking Proximity Ratio" field; when the number of candidate preceding paths is not greater than the number of parsing nodes, it writes one-third. The platform then binds and stores the "Ranking Proximity Ratio" field with the current knowledge point identifier and the question parsing template identifier. After generating the average conductance difference, the platform reads this "Ranking Proximity Ratio" field, multiplies it by the average conductance difference, and forms a preset conductance difference threshold, which is then bound and stored with the initial conductance ranking sequence identifier. When the difference in pre-block conductance between adjacent candidate pre-block segments in the pre-block conductance sorting is zero, the remote education platform reads the conductance calculation precision bits from the conductance calculation parameter table. The conductance calculation precision bits are written by the question analysis template. The remote education platform divides the value one by the power of ten conductance calculation precision bits to form the minimum conductance difference judgment value, and writes the minimum conductance difference judgment value into the preset conductance difference threshold, so that adjacent sorting records with a pre-block conductance difference of zero can enter the blocking stability and relative position reading process.

[0061] When the difference in pre-block conduction between adjacent candidate pre-block conduction segments is lower than a preset conduction difference threshold, the remote education platform reads the block stability and relative position of the corresponding candidate pre-block conduction segment record.

[0062] The stability of the obstruction is formed as follows: The distance education platform reads the number of times the same candidate preceding obstruction fragment record appears repeatedly in multiple current knowledge point conduction deviation analysis node records, and reads the total number of current knowledge point conduction deviation analysis node records participating in the comparison; in this embodiment, multiple current knowledge point conduction deviation analysis node records are conduction deviation analysis node records formed by multiple students for the same current knowledge point. The distance education platform uses the current knowledge point identifier, the target knowledge call field, and the knowledge fragment identifier in the candidate preceding obstruction fragment record as a common search key to read the conduction deviation analysis node records corresponding to multiple students from the same teaching class and the same course task batch, and uses the read conduction deviation analysis node records as the current knowledge point conduction deviation analysis node records participating in the comparison. The number of recurrences is the number of times the knowledge fragment identifier in the candidate preceding obstruction fragment record is associated and hit in the current knowledge point conduction deviation analysis node records participating in the comparison.

[0063] When the total number of records of the current knowledge point transmission deviation from the parsing node is greater than zero, the remote education platform will divide the number of repetitions by the total number of records of the current knowledge point transmission deviation from the parsing node to form the stagnation stability; when the total number of records of the current knowledge point transmission deviation from the parsing node is zero, the remote education platform will write the stagnation stability to zero; the remote education platform will write the stagnation stability to the candidate preceding stagnation segment record.

[0064] The remote education platform reads the path order of the knowledge fragment identifier corresponding to the candidate preceding obstruction fragment record in the standard knowledge traversal path, and reads the path order of the path branch point in the standard knowledge traversal path. When the path order of the knowledge fragment identifier corresponding to the candidate preceding obstruction fragment record is less than the path order of the path branch point, the candidate preceding obstruction fragment record is determined to be upstream of the path branch point. When the path order of the knowledge fragment identifier corresponding to the candidate preceding obstruction fragment record is equal to the path order of the path branch point, the candidate preceding obstruction fragment record is determined to be at the path branch point and processed as being upstream of the path branch point. When the path order of the knowledge fragment identifier corresponding to the candidate preceding obstruction fragment record is greater than the path order of the path branch point, the candidate preceding obstruction fragment record is determined to be downstream of the path branch point.

[0065] The preset stagnation stability threshold is formed as follows: The remote education platform reads the stagnation stability of all candidate preceding stagnation segment records in the same initial conductance sorting sequence, averages all stagnation stability values ​​to form the stagnation stability mean, and writes the stagnation stability mean into the preset stagnation stability threshold. When there is no readable stagnation stability in the initial conductance sorting sequence, the remote education platform does not perform weight adjustment based on stagnation stability, and determines the target preceding stagnation segment record according to the preceding stagnation conductance sorting and upstream and downstream dependencies.

[0066] The weight adjustment step size field is formed as follows: The remote education platform reads the number of adjacent records in the initial conductance sorting sequence whose adjacent conductance difference is lower than a preset conductance difference threshold, and reads the total number of adjacent records in the initial conductance sorting sequence; when the total number of adjacent records is greater than zero, the remote education platform divides the number of adjacent records by the total number of adjacent records to form a sorting proximity ratio; when the total number of adjacent records is zero, the remote education platform writes the sorting proximity ratio to zero; when the sorting proximity ratio is less than one-tenth, the weight adjustment step size field is written to one-tenth; when the sorting proximity ratio is not less than one-tenth and not greater than one-half, the weight adjustment step size field is written to the sorting proximity ratio; when the sorting proximity ratio is greater than one-half, the weight adjustment step size field is written to one-half. The remote education platform binds and stores the weight adjustment step size field with the initial conductance sorting sequence identifier. The remote education platform adds 1 to the weight adjustment step size field to form a preset gain coefficient; the remote education platform subtracts the weight adjustment step size field from 1 to form a preset weight reduction coefficient, and binds and stores the preset gain coefficient and preset weight reduction coefficient with the current knowledge point identifier, the question analysis template identifier and the initial conductance sorting sequence identifier.

[0067] When a candidate preceding obstruction segment record is located upstream of a path bifurcation point and its obstruction stability reaches a preset obstruction stability threshold, the remote education platform multiplies the path acceptance component weight and historical verification component weight of the corresponding candidate preceding obstruction segment record by a preset gain coefficient to form adjusted path acceptance component weights and adjusted historical verification component weights. The overlapping behavior component weight remains unchanged and is then written into the adjusted overlapping behavior component weights. Conversely, when either the candidate preceding obstruction segment record is located downstream of a path bifurcation point or its obstruction stability is lower than a preset obstruction stability threshold, the remote education platform multiplies the path acceptance component weight of the corresponding candidate preceding obstruction segment record by a preset weight reduction coefficient to form adjusted path acceptance component weights. The overlapping behavior component weight and historical verification component weight remain unchanged and are then written into the adjusted overlapping behavior component weights and adjusted historical verification component weights, respectively. The distance education platform sums the adjusted path carrying component weights, the adjusted behavior overlap component weights, and the adjusted historical verification component weights to form a total adjusted weight. Then, it divides each adjusted weight by the total adjusted weight to form the normalized path carrying component weights, the normalized behavior overlap component weights, and the normalized historical verification component weights.

[0068] The distance education platform recalculates the pre-blockage transmission rate based on the normalized path acceptance component weight, the normalized behavior overlap component weight, and the normalized historical verification component weight. Specifically, the platform multiplies the path acceptance component by the normalized path acceptance component weight, the behavior overlap component by the normalized behavior overlap component weight, and the historical verification component by the normalized historical verification component weight, then sums these three products to form the recalculated pre-blockage transmission rate. The platform updates the corresponding candidate pre-blockage segment records with the recalculated pre-blockage transmission rate and re-forms the transmission rate ranking sequence.

[0069] The remote education platform selects target pre-blocking segment records based on the re-formed conduction sorting sequence and the upstream and downstream dependencies in the candidate pre-conduction path set. The platform sequentially reads candidate pre-blocking segment records according to the conduction sorting sequence and reads the path order of the pre-knowledge point connection segment containing the current candidate pre-blocking segment record within the candidate pre-conduction path set. The platform then continues reading the path order of the pre-knowledge point connection segments containing subsequent candidate pre-blocking segment records. When the path order of the pre-knowledge point connection segment containing the current candidate pre-blocking segment record is less than the path order of the pre-knowledge point connection segment containing the subsequent candidate pre-blocking segment record, and the current candidate pre-blocking segment record's sorting position in the conduction sorting sequence is not later than the subsequent candidate pre-blocking segment record, the platform writes the subsequent candidate pre-blocking segment record into the masking segment set and writes the current candidate pre-blocking segment record identifier, the subsequent candidate pre-blocking segment record identifier, and the corresponding path order into the same masking record.

[0070] The distance education platform counts the number of downstream occluded segments corresponding to candidate preceding blocking segment records that are not written to the occluded segment set. The platform uses each candidate preceding blocking segment record not written to the occluded segment set as an upstream candidate preceding blocking segment record identifier, and counts the number of corresponding downstream candidate preceding blocking segment record identifiers in the occluded segment set to form the downstream occluded segment count. The platform sorts the candidate preceding blocking segment records not written to the occluded segment set according to the downstream occluded segment count from largest to smallest, forming a path coverage order; the larger the downstream occluded segment count, the higher the path coverage order. When multiple candidate preceding blocking segment records not written to the occluded segment set have the same number of downstream occluded segments, the platform assigns the same path coverage order to these records.

[0071] In the candidate pre-blocking segment records that are not written into the occlusion segment set, the distance education platform identifies the candidate pre-blocking segment record with the highest path coverage order as the target pre-blocking segment record. When multiple candidate pre-blocking segment records have the same path coverage order, the distance education platform reads the sorting position of each candidate pre-blocking segment record in the conductance sorting sequence and identifies the candidate pre-blocking segment record with the highest sorting position in the conductance sorting sequence as the target pre-blocking segment record. The target pre-blocking segment record includes the target candidate pre-knowledge point identifier, the target knowledge segment identifier, the target pre-blocking conductance, the path connection component, the behavior overlap component, the historical verification component, the path coverage order, and the conductance sorting position, and is bound and stored with the corresponding conductance deviation parsing node record for the teacher's visual interactive record generation process to read.

[0072] Step 5: The remote education platform generates a visual interaction record for the teacher based on the records of the target pre-blockage segments, the transmission deviation analysis nodes, and the candidate pre-transmission paths where the target pre-blockage segments are located, and updates the corresponding interaction status based on the visual interaction record for the teacher. Specifically, in one implementation, the remote education platform reads the target pre-obstruction fragment record, the conduction deviation resolution node record, and the candidate pre-obstruction conduction path where the target pre-obstruction fragment record is located. The target pre-obstruction fragment record includes the target candidate pre-knowledge point identifier, the target knowledge fragment identifier, the target pre-obstruction conduction quantity, the path acceptance component, the behavior overlap component, the historical verification component, the path coverage order, and the conduction quantity sorting position; the conduction deviation resolution node record includes the current knowledge point identifier, the conduction deviation resolution node identifier, the acceptance failure node identifier, the knowledge call field, the node order, and the corresponding answer action record identifier; the candidate pre-obstruction conduction path includes the path identifier, the pre-knowledge point connection segment identifier, the upstream and downstream dependencies, and the path order. The remote education platform binds the target knowledge fragment identifier in the target pre-obstruction fragment record with the pre-knowledge point connection segment identifier in the candidate pre-obstruction conduction path, and writes the knowledge call field in the conduction deviation resolution node record into this binding relationship, forming the basic record for interactive generation.

[0073] The remote education platform generates a visual interaction record for the teacher's end based on the basic interaction record. The platform reads the knowledge fragment configuration record corresponding to the target candidate prerequisite knowledge point identifier, and reads the knowledge fragment start boundary, knowledge fragment end boundary, fragment knowledge call field, and fragment resource location range corresponding to the target knowledge fragment identifier. The platform also reads the transmission deviation resolution node identifier, failure node identifier, and node order from the transmission deviation resolution node record, and reads the path order and upstream / downstream dependencies from the candidate prerequisite transmission path. The platform writes the target knowledge fragment identifier, transmission deviation resolution node identifier, failure node identifier, knowledge call field, path order, target prerequisite obstruction transmission quantity, path acceptance component, behavior overlap component, and historical verification component into the teacher's visual interaction record. The teacher's visual interaction record also includes an interaction record identifier, student identifier, current knowledge point identifier, target candidate prerequisite knowledge point identifier, target knowledge fragment resource location, interaction generation time, and interaction status field. The interaction status field is written as pending execution status when the teacher's visual interaction record is generated.

[0074] The distance education platform generates interaction execution data based on the teacher's visual interaction records. This data includes an interaction record identifier, a teacher confirmation field, a resource access field, a student execution completion field, an execution start time, an execution completion time, and a post-execution response record identifier. The teacher confirmation field is written after the teacher reads the visual interaction record and confirms the corresponding target knowledge segment. The resource access field is read and written by the platform based on the location of the target knowledge segment resource. The student execution completion field is written after the student completes at least one of the following: resource access, practice answering, or interaction feedback. The post-execution response record identifier is written when the student submits their answer to the current knowledge point after the interaction execution. The distance education platform binds and stores the interaction execution data with the teacher's visual interaction records, ensuring that subsequent interaction status updates can read the execution results under the same interaction record identifier.

[0075] After the remote education platform generates the interaction execution data corresponding to the visual interaction record on the teacher's end, it reads the current knowledge point answer record after the interaction execution and reconstructs the updated propagation deviation analysis node record according to the standard analysis node sequence. When reconstructing the updated propagation deviation analysis node record, the remote education platform reads the current knowledge point answer record after the interaction execution and reads the standard analysis node sequence corresponding to the current knowledge point according to the aforementioned standard analysis node sequence generation rules. The remote education platform parses the current knowledge point answer record after the interaction execution into multiple updated answer action records with temporal relationships. It compares the input object, processing object, and output object of the updated answer action record with the standard analysis nodes node by node, and substitutes the output object corresponding to the candidate deviation analysis node into the subsequent analysis nodes for acceptance verification. If the acceptance verification fails, the remote education platform writes the candidate deviation analysis node and the failed acceptance node into the updated propagation deviation analysis node record. The updated propagation deviation analysis node record includes the updated record identifier, interaction record identifier, updated propagation deviation analysis node identifier, updated failed acceptance node identifier, updated node sequence, and updated knowledge call field.

[0076] The remote education platform reads the conduction deviation resolution node identifier corresponding to the teacher's visual interaction record and searches for the same conduction deviation resolution node identifier in the updated conduction deviation resolution node record. When the updated conduction deviation resolution node record still contains the conduction deviation resolution node corresponding to the teacher's visual interaction record, the remote education platform writes the teacher's visual interaction record to a persistent blocking state. When writing to the persistent blocking state, the remote education platform writes the interaction record identifier, the original conduction deviation resolution node identifier, the updated conduction deviation resolution node identifier, the original node sequence, the updated node sequence, and the execution completion time into the persistent blocking state record, and updates the interaction status field in the teacher's visual interaction record to the persistent blocking state.

[0077] When the updated conduction deviation resolution node record does not contain the conduction deviation resolution node corresponding to the teacher-side visual interaction record, the remote education platform continues to read the updated conduction deviation resolution node identifier and updated node sequence from the updated conduction deviation resolution node record, and reads the original node sequence of the conduction deviation resolution node corresponding to the teacher-side visual interaction record. The remote education platform compares the updated node sequence with the original node sequence; when the updated node sequence is greater than the original node sequence, the remote education platform determines that the updated conduction deviation resolution node record contains a new conduction deviation resolution node downstream of the conduction deviation resolution node corresponding to the teacher-side visual interaction record, writes the new downstream deviation status to the teacher-side visual interaction record, and updates the interaction status field in the teacher-side visual interaction record to the new downstream deviation status.

[0078] The remote education platform writes the teacher's visual interaction record to the newly added downstream deviation status and updates the interaction status field in the teacher's visual interaction record to the newly added downstream deviation status.

[0079] The distance education platform will write the newly added conduction deviation resolution node into a new conduction deviation resolution node record. The new conduction deviation resolution node record includes a new record identifier, an interaction record identifier, a newly added conduction deviation resolution node identifier, a new node sequence number, a new knowledge call field, the original conduction deviation resolution node identifier, and the corresponding current knowledge point identifier. The distance education platform will write the new conduction deviation resolution node record into the pending record of the candidate preceding conduction path set, and during the next execution of the preceding knowledge point connection segment retrieval, it will read the newly added knowledge call field from the new conduction deviation resolution node record, causing the new conduction deviation resolution node record to re-enter the formation process of the candidate preceding conduction path set.

[0080] When the updated conduction deviation resolution node record does not contain a conduction deviation resolution node corresponding to the teacher-side visual interaction record, and there are no newly added conduction deviation resolution nodes with a node order greater than the original node order in the updated conduction deviation resolution node record, the remote education platform writes the teacher-side visual interaction record to a completely deactivated state. The completely deactivated state is used to record that the original conduction deviation resolution node corresponding to the teacher-side visual interaction record and any newly added conduction deviation resolution nodes downstream of the original conduction deviation resolution node no longer appear; it is not used to exclude other conduction deviation resolution nodes with a node order less than the original node order. Other conduction deviation resolution nodes with a node order less than the original node order remain in the updated conduction deviation resolution node record and enter the corresponding candidate preceding conduction path set formation process according to their corresponding knowledge call fields.

[0081] When writing the fully released state, the remote education platform writes the interaction record identifier, the original conduction deviation analysis node identifier, the target pre-blockage segment record identifier, the execution completion time, the updated record identifier, and the fully released state boundary field into the fully released state record, and updates the interaction state field in the teacher's visual interaction record to the fully released state. The fully released state boundary field is used to record the judgment range corresponding to the fully released state, which is the original conduction deviation analysis node and the newly added conduction deviation analysis node downstream of the original conduction deviation analysis node. The remote education platform binds and stores the continuous blockage state record, the newly added downstream deviation state record, and the fully released state record to the teacher's visual interaction record, so that the same teacher's visual interaction record has a unique interaction state result after the interaction is executed.

[0082] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A visual data interaction method based on distance education, characterized in that, include: Step 1: The remote education platform reads the standard parsing node sequence and student answer records of the current knowledge point, parses the student answer records into multiple answer action records with temporal relationship, compares the input object, processing object and output object of the answer action record with the standard parsing node node by node, and substitutes the output object corresponding to the candidate deviation parsing node into the subsequent parsing node to perform the inheritance verification, forming the transmission deviation parsing node record; Step 2: Based on the knowledge call field in the transmission deviation resolution node record, the remote education platform retrieves the prerequisite knowledge point connection segment in the course knowledge dependency table, and prunes it through dependency continuity verification and historical remediation verification results to form a candidate prerequisite transmission path set. The candidate prerequisite transmission path set is written into the upstream and downstream dependency relationship between the prerequisite knowledge point connection segments. Step 3: The remote education platform reads the learning behavior data of the candidate prerequisite knowledge points along the candidate prerequisite transmission path set, maps the learning behavior data to the corresponding knowledge segment identifier, and forms a candidate prerequisite obstruction segment record based on the continuity and overlap of the learning behavior data on the knowledge segment identifier. Step 4: The remote education platform calculates the path acceptance component, behavior overlap component, and historical verification component based on the candidate pre-blocking segment record and the candidate pre-transmission path set, and generates the pre-blocking transmission quantity by weighting. The target pre-blocking segment record is determined based on the ranking of the pre-blocking transmission quantity and the upstream and downstream dependencies. Step 5: The remote education platform generates a visual interaction record for the teacher based on the records of the target pre-blockage segments, the transmission deviation analysis nodes, and the candidate pre-transmission paths where the target pre-blockage segments are located, and updates the corresponding interaction status based on the visual interaction record for the teacher.

2. The visual data interaction method based on distance education according to claim 1, characterized in that, The acceptance verification in step one includes: The remote education platform reads the output object corresponding to the candidate deviation parsing node and converts the output object into input parameters that can be read by subsequent parsing nodes according to the preset node input format; The remote education platform selects at least two consecutive subsequent parsing nodes according to the arrangement order of the standard parsing node sequence, substitutes the input parameters into the selected consecutive subsequent parsing nodes in sequence, and calls the node processing relationship of each consecutive subsequent parsing node to perform calculation replay, thereby obtaining the replay calculation result and replay processing relationship order corresponding to each consecutive subsequent parsing node. The remote education platform compares the replay calculation results with the node output results of the corresponding consecutive subsequent parsing nodes, and compares the order of replay processing relationships with the order of node processing relationships in the standard parsing node sequence. When the replay calculation results of each consecutive subsequent parsing node are consistent with the output results of the corresponding node, and the replay processing order is consistent with the node processing order in the standard parsing node sequence, the candidate deviating parsing node is not written into the propagated deviating parsing node record. When at least one of the following conditions is met: the replay calculation result is inconsistent with the output result of the corresponding node, or the replay processing order is inconsistent with the node processing order in the standard parsing node sequence, the corresponding consecutive subsequent parsing node is determined as the failure to accept the task, and the candidate deviation parsing node and the failure to accept the task are written into the propagation deviation parsing node record.

3. The visual data interaction method based on distance education according to claim 1, characterized in that, Methods for obtaining historical remediation verification results include: The remote education platform reads the historical transmission deviation analysis node records corresponding to the connection segments of the preceding knowledge points in the candidate preceding transmission path set, and sorts them according to the occurrence time of the obstruction handling task to form a historical obstruction handling sequence. The remote education platform sequentially reads the pre-treatment conduction deviation analysis node record and the post-treatment re-identified conduction deviation analysis node record for each historical obstruction treatment record in the historical obstruction treatment sequence. Historical obstruction treatment records that do not contain the pre-treatment conduction deviation analysis node in the post-treatment re-identified conduction deviation analysis node record are included in the historical release record. Historical obstruction treatment records that still contain the pre-treatment conduction deviation analysis node in the post-treatment re-identified conduction deviation analysis node record are included in the historical continuity record. Historical obstruction treatment records that contain newly added conduction deviation analysis nodes located downstream of the pre-treatment conduction deviation analysis node in the post-treatment re-identified conduction deviation analysis node record are included in the historical downstream migration record. The remote education platform generates a time decay coefficient based on the time interval between the occurrence time of the obstruction handling task and the current processing time for each historical obstruction handling record. The number of historical unblocking records, historical continuous records, and historical downstream migration records are multiplied by the corresponding time decay coefficient to form the decayed unblocking amount, decayed continuous amount, and decayed migration amount. The remote education platform calculates the historical verification value of the connection segment based on the attenuation removal amount, attenuation duration amount, and attenuation migration amount, and writes the historical verification value of the connection segment into the historical remediation verification result.

4. The visual data interaction method based on distance education according to claim 1, characterized in that, The formation methods of candidate pre-arrest fragment records include: The remote education platform reads the knowledge fragment identifier, behavior type field and behavior occurrence sequence corresponding to the learning behavior data under the same candidate prerequisite knowledge point, and forms a standard knowledge traversal path within the candidate prerequisite knowledge point based on the course knowledge dependency table. The distance education platform forms an actual knowledge fragment access sequence according to the order of behavior occurrence, and compares the actual knowledge fragment access sequence with the standard knowledge traversal path position by position to determine the path fork point; When none of the consecutive knowledge fragment identifiers after the path fork point fall into the standard knowledge traversal path, and the number of consecutive deviations reaches the preset path deviation length threshold, the corresponding knowledge fragment identifiers will be excluded from the continuous behavior record. When a knowledge fragment identifier after a path fork point falls back into the standard knowledge traversal path, the knowledge fragment identifier that falls back into the path will be included in the continuous behavior record. The remote education platform compares the behavior type field of adjacent learning behavior data with the preset behavior succession relationship. When the behavior type field meets the preset behavior succession relationship and the knowledge fragment identifiers are adjacent, a continuous behavior record is formed. The remote education platform reads the overlapping hit results of multiple learning behaviors of each knowledge segment identifier in the continuous behavior record. When multiple learning behaviors overlap and the previous knowledge segment identifier is included in the continuous behavior record but the next knowledge segment identifier is not included in the continuous behavior record, the current knowledge segment identifier is written into the candidate pre-blocking segment record.

5. The visual data interaction method based on distance education according to claim 4, characterized in that, The result of overlapping learning behaviors is formed in the following way: The remote education platform reads multiple learning behavior data corresponding to the same knowledge segment identifier in the continuous behavior record and the knowledge call field in the transmission deviation parsing node record, and reads the behavior start time, end time, behavior belonging segment boundary and behavior associated knowledge call field of each learning behavior data; The remote education platform compares the boundaries of the behavior-related fragments with the boundaries of the knowledge fragments, compares the overlap of the start and end times of the behavior in terms of time intervals, and compares the consistency of the behavior-related knowledge call fields with the knowledge call fields in the transmission deviation parsing node records. When the boundaries of at least two types of learning behavior data fall into the same knowledge segment boundary, the behavior time intervals overlap and the fields are consistent, the corresponding knowledge segment identifier is written into the multi-type learning behavior overlap hit result; the remote education platform determines the corresponding knowledge segment identifier as the knowledge segment identifier that meets the overlap judgment based on the multi-type learning behavior overlap hit result, and uses it to form a candidate pre-blocking segment record.

6. The visual data interaction method based on distance education according to claim 4, characterized in that, The target pre-blocking segment record is determined in the following way: The remote education platform reads the number of times the candidate pre-blocking fragment records are repeated in the records of multiple current knowledge point transmission deviation parsing nodes, and forms the blockage stability based on the number of repetitions; The distance education platform determines the relative position between the candidate preceding blocking fragment record and the path bifurcation point based on the order of the knowledge fragment identifier corresponding to the candidate preceding blocking fragment record in the standard knowledge traversal path. When the difference in pre-block conduction between adjacent candidate pre-block segments is lower than a preset conduction difference threshold, the remote education platform reads the block stability and relative position of the corresponding candidate pre-block segment record. When the candidate pre-blocking segment record is located upstream of the path bifurcation point and the blocking stability reaches the preset blocking stability threshold, the path carrying component weight and the historical verification component weight of the corresponding candidate pre-blocking segment record are multiplied by the preset gain coefficient respectively. If any of the following conditions are met: the candidate preceding blocking segment record is located downstream of the path bifurcation point, or the blocking stability is lower than the preset blocking stability threshold, the weight of the path receiving component of the corresponding candidate preceding blocking segment record will be multiplied by the preset weight reduction coefficient. The remote education platform recalculates the pre-blocking transmission volume based on the adjusted path carrying component weight, behavior overlap component weight, and historical verification component weight, and determines the target pre-blocking segment record based on the recalculated pre-blocking transmission volume.

7. The visual data interaction method based on distance education according to claim 1, characterized in that, Step four determines the target pre-blocking segment record based on the pre-blocking conduction volume ranking and upstream-downstream dependencies, including: The distance education platform sorts candidate pre-blocking segment records based on pre-blocking conductance, forming a conductance sorting sequence; The remote education platform reads candidate pre-blocking segment records sequentially according to the conduction volume sorting sequence, and determines whether the currently read candidate pre-blocking segment record is located in the upstream path segment of subsequent candidate pre-blocking segment records based on the upstream and downstream dependencies in the candidate pre-conduction path set. If the currently read candidate pre-blocking segment record is located in the upstream path segment of the subsequent candidate pre-blocking segment record, and the sorting position of the currently read candidate pre-blocking segment record in the conductance sorting sequence is not later than that of the subsequent candidate pre-blocking segment record, then the subsequent candidate pre-blocking segment record will be written into the masking segment set. The remote education platform counts the number of downstream occlusion segments corresponding to candidate preceding blocking segments that are not written into the occlusion segment set, and forms the path coverage order. In the candidate preceding blocking segment records that have not been written into the occlusion segment set, the remote education platform determines the candidate preceding blocking segment record with the highest path coverage order as the target preceding blocking segment record. When multiple candidate anterior blockage fragment records have the same path coverage order, the candidate anterior blockage fragment record with the highest order in the conductance sorting sequence is determined as the target anterior blockage fragment record.

8. The visual data interaction method based on distance education according to claim 1, characterized in that, The interaction state is updated in the following ways: After the remote education platform generates the corresponding interactive execution data corresponding to the visual interactive record on the teacher's end, it reads the current knowledge point answer record after the interactive execution, and re-forms and updates the deviated parsing node record according to the standard parsing node sequence. If the updated conduction deviation resolution node record still contains the conduction deviation resolution node corresponding to the teacher's visual interaction record, the teacher's visual interaction record will be written to the continuous blockage state. When the updated transmission deviation analysis node record does not contain the transmission deviation analysis node corresponding to the teacher's visual interaction record, but contains a newly added transmission deviation analysis node located downstream of the transmission deviation analysis node, the teacher's visual interaction record is written to the newly added downstream deviation state, and the newly added transmission deviation analysis node is written to the new transmission deviation analysis node record, so that the new transmission deviation analysis node record re-enters the process of forming the candidate preceding transmission path set. When the updated transmission deviation resolution node record does not contain the transmission deviation resolution node corresponding to the teacher's visual interaction record, and does not contain any newly added transmission deviation resolution node located downstream of that transmission deviation resolution node, the teacher's visual interaction record will be written to the completely deactivated state.

Images