Cognitive interaction training system and method for group collaboration

Abstract

The application discloses a cognitive interaction training system and method for group cooperation, and relates to the technical field of human-computer cognitive interaction, and specifically comprises the following steps: continuously recording cognitive interaction operations generated by multiple users around the same training target, generating operation characterization marks for each cognitive interaction operation, and forming a cognitive interaction sequence set according to user identification, operation content and occurrence sequence; successively comparing adjacent cognitive interaction operations in the cognitive interaction sequence set, identifying covering features in the state change process through the operation characterization marks, judging whether the situation that multiple user continuous input operations and training states are continuously covered occurs, and forming a covering section set. The application solves the problem that the training state is covered and the evolution process is not traceable due to multiple user continuous input, realizes dynamic regulation and control of complete reconstruction and state updating of the cognitive interaction process, and thus improves the explainability and collaborative control ability of the training process.

Description

Technical Field

[0001] This invention relates to the field of human-computer cognitive interaction technology, specifically to a cognitive interaction training system and method for group collaboration. Background Technology

[0002] Group-based collaborative cognitive interactive training refers to a technical solution that combines cognitive training tasks with group collaboration and competition mechanisms by constructing a human-computer interaction environment that supports simultaneous participation of multiple users. This allows multiple users to complete cognitive training within a unified task framework. In existing technologies, this type of training is typically implemented based on a ground-based projection interface or a spatial interactive platform. Cameras or sensors collect multi-user action data, and a computer processing module identifies and parses the actions, mapping different users' operations to corresponding task instructions. Based on this, the system constructs task rules that include collaborative and competitive relationships, enabling multiple users to interact and participate around the training objective. Simultaneously, real-time feedback on user operations is provided through projection or a graphical interface, and the accuracy of user responses and group collaborative performance are recorded and analyzed. A quantitative evaluation model is used to characterize the training state, and training task parameters are adaptively adjusted based on the analysis results, thus forming a complete training process encompassing data acquisition, behavior recognition, task construction, interactive feedback, and dynamic adjustment.

[0003] The existing technology has the following shortcomings: In group-based collaborative cognitive interaction training, when multiple users continuously input different operations around the same training objective, the training state may be frequently updated within a short period of time. Because the user inputs are concurrent and lack sequential records, the system directly overwrites the previous state when processing subsequent inputs, resulting in the inability to retain and restore the state change path. Furthermore, since the state update only reflects the current operation result and does not retain historical evolution information, the evolution of the training state becomes untraceable. Existing technologies cannot adjust the state update strategy for group-based collaborative cognitive interaction training based on the cognitive interaction information under conditions of continuous input from multiple users and ongoing overwriting of the training state. This leads to a lack of interpretability in the training process, consequently affecting subsequent training analysis and dynamic control.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a cognitive interaction training system and method for group collaboration, so as to solve the problems in the background art mentioned above.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a cognitive interaction training method for group collaboration, specifically including the following steps: S1. Continuously record the cognitive interaction operations generated by multiple users around the same training objective, and generate operation characterization marks for each cognitive interaction operation, forming a set of cognitive interaction sequences according to user identification, operation content and occurrence order. S2. Compare adjacent cognitive interaction operations in the cognitive interaction sequence set one by one, identify the coverage features in the state change process by operation characterization markers, and determine whether there are multiple users continuously inputting operations and the training state is continuously covered, and form a set of coverage segments. S3. Perform reverse expansion processing on the cognitive interaction operations corresponding to the covered segment set, use the evolution index identifier to connect the covered training states, form cognitive interaction information during the connection process, and generate a continuously changing trajectory. S4. Analyze the continuously changing trajectory, perform calculations according to the order of the evolution index identifiers, and generate an adjustment identifier sequence; S5. Match and control the continuously changing trajectory based on the adjustment label sequence, and dynamically regulate the state update process of group collaborative cognitive interaction training by combining cognitive interaction information.

[0007] Preferably, S1 is as follows: The cognitive interaction operations generated by multiple users around the same training objective are continuously recorded. Each cognitive interaction operation is sequentially assigned a user identifier, operation content, and occurrence time identifier. The cognitive interaction operations are then arranged in order according to the occurrence time identifier to form a continuously recorded sequence of cognitive interaction operations. For each cognitive interaction operation in the cognitive interaction operation sequence, an operation characterization mark generation process is performed. The user identifier, operation content and occurrence time identifier are combined and encoded to generate the corresponding operation characterization mark, and the operation characterization mark is kept in a one-to-one correspondence with the cognitive interaction operation. The operation markers are sorted according to the occurrence time, categorized and integrated according to the user identifier, and then sequentially linked with the operation content to form a set of cognitive interaction sequences that include user identifier, operation content, and occurrence order.

[0008] Preferably, S2 specifically includes the following steps: S201. Traverse the cognitive interaction operations in the cognitive interaction sequence set in the order of occurrence, pair adjacent cognitive interaction operations in turn to form adjacent cognitive interaction operation pairs, and extract the operation characterization mark corresponding to each adjacent cognitive interaction operation pair for successive comparison. S202. In the successive comparison process, the operation characterization markers in adjacent cognitive interaction operation pairs are processed by content parsing. The user identifier, operation content and occurrence order in the operation characterization markers are compared to identify the change in the state corresponding to the previous cognitive interaction operation being replaced by the next cognitive interaction operation. Adjacent cognitive interaction operation pairs that meet the replacement conditions are marked as coverage features in the state change process to determine whether there are multiple users continuously inputting operations and the training state is continuously covered. S203. For adjacent cognitive interaction pairs marked as covering features, perform continuity detection according to the order of occurrence, divide the continuously occurring covering features into intervals, and sequentially merge the covering features within the same interval to form a corresponding set of covering segments.

[0009] Preferably, S202 specifically refers to: During the successive comparison process, the operation characterization markers in adjacent cognitive interaction operation pairs are split into fields, and the operation characterization markers are parsed into user identifier field, operation content field and occurrence sequence field, and each field is read independently. Based on field splitting, the user identifier field, operation content field and occurrence order field in adjacent cognitive interaction operation pairs are compared item by item. The relationship between the preceding and following operations is determined according to the occurrence order field. When the operation content field changes and the operation content corresponding to the subsequent cognitive interaction operation replaces the operation content corresponding to the previous cognitive interaction operation, the change in the state corresponding to the previous cognitive interaction operation being replaced by the subsequent cognitive interaction operation is identified. After identifying the substitution change, adjacent cognitive interaction pairs that meet the substitution conditions are marked. The corresponding operation characterization is recorded as the coverage feature in the state change process. The coverage feature is continuously detected in the order of occurrence. When the user identifier field corresponding to the consecutively occurring coverage feature contains different user identifiers and the occurrence order field is continuously increasing, it is determined that multiple users have continuously input operations and the training state is continuously covered.

[0010] Preferably, S3 specifically includes the following steps: S301. The cognitive interaction operations corresponding to the covered segment set are arranged in reverse order according to the order of occurrence. Starting from the end of the covered segment set, the cognitive interaction operations are extracted sequentially and formed in reverse order to form a reverse unfolding sequence, so as to realize the reverse unfolding process of the covered process. S302. Based on the reverse expansion sequence, construct an evolution index identifier for each cognitive interaction operation, take the occurrence order as the index main order, and embed the user identifier and operation content into the evolution index identifier so that the evolution index identifier corresponds one-to-one with the cognitive interaction operation. S303. Based on the evolution index identifier, sequentially connect the training states corresponding to the cognitive interaction operations in the reverse unfolding sequence, connect the training states corresponding to adjacent cognitive interaction operations in the order of occurrence, and extract the user identifier, operation content and training state change information during the connection process to form cognitive interaction information. S304. Based on the cognitive interaction information, the training state change information is continuously spliced ​​according to the order of occurrence, and adjacent training state change information is sequentially connected to generate the corresponding continuous change trajectory.

[0011] Preferably, S303 is as follows: Based on the evolution index identifier, the cognitive interaction operations in the reverse unfolding sequence are sorted and mapped. The training states corresponding to each cognitive interaction operation are mapped according to the order of occurrence, and the sequential connection relationship between adjacent cognitive interaction operations is established. Based on the sorting mapping process, the training states corresponding to adjacent cognitive interaction operations are connected one by one. Adjacent training states are read in sequence according to the order of occurrence, and the state identifiers and state contents in the training states are connected and combined to form a continuous training state sequence. During the concatenation of the training state sequence, the corresponding user identifier, operation content, and training state change information are extracted for each connection node and recorded and integrated in the order of occurrence. The continuously extracted user identifier, operation content, and training state change information are sequentially associated to form corresponding cognitive interaction information.

[0012] Preferably, S4 is as follows: The continuously changing trajectory is parsed and processed. The training state change information in the continuously changing trajectory is extracted in the order of occurrence. The state identifier, state content and evolution index identifier corresponding to each training state change information are separated and read. At the same time, the parsing results are sorted and rearranged according to the occurrence order information in the evolution index identifier. The sorted data are written into the storage unit in the order of occurrence to form a parsing data structure with sequential pointing relationship. Based on the parsed data structure, the operation is performed in the order of the evolution index identifier. By sequentially traversing the adjacent training state change information in the parsed data structure, the state identifier and state content in the adjacent training state change information are read in turn. The state identifier of the previous training state change information is matched with the state identifier of the next training state change information. At the same time, the differences between the state content of the previous training state change information and the state content of the next training state change information are extracted and combined to generate the corresponding state change operation data. For state change operation data, an identifier construction process is performed. The state change operation data is associated with the corresponding evolution index identifier and encoded. The data is then arranged in the order of the evolution index identifier to generate an adjustment identifier sequence.

[0013] Preferably, S5 is as follows: The continuous trajectory is matched and controlled based on the adjustment identifier sequence. The adjustment identifiers in the adjustment identifier sequence are read sequentially according to the order of the evolution index identifiers. Each adjustment identifier is matched with the training state change information in the continuous trajectory. By comparing the identifier content in the adjustment identifier with the state identifier and state content in the training state change information, the corresponding matching position is determined. At the corresponding matching position, the training state change information is rearranged and the content is replaced to complete the matching control of the continuous trajectory. Based on the continuous change trajectory of the matching control, the state update process of the group collaborative cognitive interaction training is dynamically regulated by combining cognitive interaction information. The user identifier, operation content and training state change information in the cognitive interaction information are read in sequence according to the occurrence order. The cognitive interaction information and the matched training state change information are correlated and processed accordingly. Based on the correlation results, the update order and update content of the training state change information are adjusted to form an update driving sequence for use in the state update process.

[0014] The preferred cognitive interaction training system for group collaboration includes a cognitive interaction modeling module, a coverage feature recognition module, a state evolution reconstruction module, a regulation label generation module, and a dynamic regulation execution module. The cognitive interaction modeling module continuously records the cognitive interaction operations generated by multiple users around the same training objective, and generates operation characterization tags for each cognitive interaction operation, forming a set of cognitive interaction sequences according to user identifier, operation content and occurrence order; The coverage feature recognition module compares adjacent cognitive interaction operations in the cognitive interaction sequence set one by one. It identifies the coverage features in the state change process by using operation characterization markers to determine whether there are multiple users continuously inputting operations and the training state is continuously covered, and forms a set of coverage segments. The state evolution reconstruction module performs reverse expansion processing on the cognitive interaction operations corresponding to the covered segment set, uses the evolution index identifier to connect the covered training states, forms cognitive interaction information during the connection process, and generates a continuous change trajectory. The adjustment identifier generation module analyzes and processes the continuously changing trajectory, performs calculations according to the order of the evolution index identifiers, and generates an adjustment identifier sequence. The dynamic control execution module matches and controls the continuously changing trajectory based on the adjustment identifier sequence, and dynamically controls the state update process of the group collaborative cognitive interaction training by combining cognitive interaction information.

[0015] The technical effects and advantages provided by the present invention in the above technical solution are as follows: This invention constructs a complete processing chain—"cognitive interaction operation recording—coverage feature identification—state evolution reconstruction—adjustment identifier generation—dynamic control execution"—to achieve structured recovery and reconstruction of the state evolution process in scenarios with continuous input from multiple users and continuously covered training states. Based on a set of cognitive interaction sequences, it identifies coverage features by comparing adjacent cognitive interaction operations one after another, and further expands the covered segment set in reverse, introducing an evolution index identifier to sequentially concatenate the covered training states. This forms cognitive interaction information containing user identifiers, operation content, and training state change information, generating a continuous change trajectory. This process re-emphasizes the intermediate evolution process that was originally lost due to state coverage, transforming the training state from "only retaining the current result" to "fully recording the evolution path," thereby achieving a traceable expression of the training state change process and providing complete data support for subsequent analysis.

[0016] This invention analyzes and processes continuously changing trajectories to generate adjustment identifier sequences, further transforming state evolution information into structured identifier data that can be used for control. During the state update phase, matching control and dynamic adjustment strategies are introduced, matching the adjustment identifier sequences with the continuously changing trajectories. Through rearrangement and content replacement processing, fine-grained control of the training state change path is achieved. Simultaneously, cognitive interaction information is combined to dynamically adjust the state update order and content, enabling the state update process to respond to changes in multi-user cognitive interaction behavior. This technical solution not only achieves controllable adjustment of the state update logic in complex group collaborative interaction processes but also improves the interpretability, consistency, and adaptability of the training process. It transforms group collaborative cognitive interaction training from a passive response to an active adjustment process driven by cognitive interaction information, thereby improving the overall training effect and interactive experience in multi-user collaborative participation scenarios. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0018] Figure 1 This is a schematic diagram of the process of the present invention.

[0019] Figure 2This is a schematic diagram of the modules of the present invention. Detailed Implementation

[0020] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0021] This invention provides, for example Figure 1 The cognitive interaction training method for group collaboration shown includes the following steps: S1. Continuously record the cognitive interaction operations generated by multiple users around the same training objective, and generate operation characterization marks for each cognitive interaction operation, forming a set of cognitive interaction sequences according to user identification, operation content and occurrence order. In this embodiment, S1 specifically refers to: The cognitive interaction operations generated by multiple users around the same training objective are continuously recorded. Each cognitive interaction operation is sequentially assigned a user identifier, operation content, and occurrence time identifier. The cognitive interaction operations are then arranged in order according to the occurrence time identifier to form a continuously recorded sequence of cognitive interaction operations. For each cognitive interaction operation in the cognitive interaction operation sequence, an operation characterization mark generation process is performed. The user identifier, operation content and occurrence time identifier are combined and encoded to generate the corresponding operation characterization mark, and the operation characterization mark is kept in a one-to-one correspondence with the cognitive interaction operation. The operation markers are sorted according to the occurrence time, categorized and integrated according to the user identifier, and then sequentially linked with the operation content to form a set of cognitive interaction sequences that include user identifier, operation content, and occurrence order.

[0022] In the cognitive interaction training process oriented towards group collaboration, image acquisition devices or interactive sensing devices are deployed to continuously record the cognitive interaction operations generated by multiple users around the same training objective. During the acquisition process, a unified clock source is used to synchronize the time of all inputs, transforming each triggered behavior into a identifiable data event. In the acquisition phase, each cognitive interaction operation is directly attached with a user identifier, operation content, and occurrence time identifier. The user identifier is obtained through target tracking or device binding, the operation content is determined through the mapping relationship between the interaction area and the task semantics, and the occurrence time identifier is generated by the system's time sequence recording. All data is written to a cache queue according to the order of occurrence and output in sequence, thus forming a continuously recorded cognitive interaction operation sequence, making the inputs of different users distinguishable and comparable on the same timeline.

[0023] After the formation of a continuously recorded sequence of cognitive interaction operations, operation characterization marker generation is performed on each cognitive interaction operation in the sequence. The user identifier, operation content, and occurrence time identifier are combined and encoded in a unified format. The combined encoding adopts a combination of field concatenation and positional encoding to embed information of different dimensions into data units with fixed structure. An index pointer is used to establish a one-to-one correspondence between operation characterization markers and cognitive interaction operations. In actual implementation, for different user operations under the same training objective, the user identifier, operation content, and time information can be arranged by preset field positions, so that each operation characterization marker has uniqueness and parsability, thereby providing a unified data foundation for subsequent sorting and concatenation.

[0024] After generating the operation characterization tags, they are globally sorted according to their occurrence time, ensuring a sequential arrangement on a unified timeline. Simultaneously, the operation characterization tags are categorized and integrated based on user identifiers, grouping tags belonging to the same user into the same set. Within each set, the tags are sequentially linked according to their operation content, forming a continuous chain structure. In the specific implementation, the sorted tag sequence is traversed, grouping containers are created based on user identifiers, and tags are inserted sequentially into the corresponding containers. The chain structure is then connected with the operation content, forming a cognitive interaction sequence set containing user identifiers, operation content, and the order of occurrence. This transforms concurrent input from multiple users into structured, traceable data.

[0025] The same training objective refers to a unified task object preset by the system during the training process. This object remains consistent throughout the training cycle, enabling multiple users to perform cognitive interaction operations around the same objective. Cognitive interaction operations refer to the user's response behavior to the training objective through body movements or touch behavior. User identifiers are used to distinguish different user sources, operation content is used to characterize the specific type of execution behavior, and occurrence time identifiers are used to record the chronological order of operations. Continuously recorded cognitive interaction operation sequences are used to describe the arrangement of all user inputs on the timeline. Operation characterization mark generation processing is used to transform scattered operation data into data units of a unified format, and combined encoding is used to achieve the fusion expression of multi-dimensional information, so that each operation characterization mark contains both source information and behavioral semantics.

[0026] Categorization and integration are used to group operation characterization markers according to user identifiers, so that the behavioral trajectories of different users can be distinguished; operation content is used to describe the specific task semantics of each cognitive interaction; sequential concatenation is used to connect operation characterization markers into a continuous chain based on time ordering, so that the behavior process is traceable; the set of cognitive interaction sequences containing user identifiers, operation content and occurrence order is used to fully represent the evolution process of multi-user behavior during group collaborative cognitive interaction training. Through this set of sequences, the operation state change path at any time can be recovered, providing basic data support for subsequent state recognition, coverage judgment and dynamic control.

[0027] S2. Compare adjacent cognitive interaction operations in the cognitive interaction sequence set one by one, identify the coverage features in the state change process by operation characterization markers, and determine whether there are multiple users continuously inputting operations and the training state is continuously covered, and form a set of coverage segments. In this embodiment, S2 specifically includes the following steps: S201. Traverse the cognitive interaction operations in the cognitive interaction sequence set in the order of occurrence, pair adjacent cognitive interaction operations in turn to form adjacent cognitive interaction operation pairs, and extract the operation characterization mark corresponding to each adjacent cognitive interaction operation pair for successive comparison. The cognitive interaction sequence set organizes cognitive interaction operations generated by multiple users around the same training objective in chronological order. Each cognitive interaction operation is accompanied by a user identifier, operation content, and occurrence time identifier during the acquisition phase, and is sorted and stored according to the occurrence time identifier. Based on this, by sequentially traversing the cognitive interaction sequence set, cognitive interaction operations at adjacent positions in the sequence are extracted and paired to form adjacent cognitive interaction operation pairs. Specifically, this can be implemented through a sliding index, that is, starting from the current cognitive interaction operation, it forms a pairing unit with the next cognitive interaction operation, and gradually moves along the sequence to form a continuous pairing structure. For each adjacent cognitive interaction operation pair, the corresponding operation characterization marker is extracted and compared successively by comparing the user identifier, operation content, and occurrence time identifier. The system reads and compares each item sequentially to obtain differences between adjacent cognitive interaction operations. For example, in a set of actual collected data, the continuously recorded cognitive interaction operation sequence is "User A performs a selection operation and triggers the target area", "User B performs a correction operation to change the target selection", and "User C performs a confirmation operation to fix the current result". By traversing the data, adjacent cognitive interaction operation pairs of "User A and User B" and "User B and User C" are formed. Then, by comparing each field through operation characterization tags, the continuous input behaviors of different users on the same training target and their changes can be clearly identified. This successive pairing and comparison method breaks down the overall sequence into multiple continuous analysis units, so that the identification of state changes is based on local continuous data, thereby improving the stability and consistency of the processing.

[0028] The cognitive interaction sequence set represents a set of multi-user cognitive interaction operations arranged in chronological order, where each cognitive interaction operation corresponds to a specific input behavior of a user during training. Cognitive interaction operations characterize the user's action response to the training objective. Pairing processing combines adjacent cognitive interaction operations into analysis units, enabling temporally continuous input behaviors to form a directly comparable data structure. Adjacent cognitive interaction operation pairs represent the relationship between two consecutive cognitive interaction operations. Operation characterization markers provide a structured description of cognitive interaction operations, including user identification, operation content, and occurrence time. Successive comparisons perform comparison processing on each adjacent cognitive interaction operation pair in sequence, gradually extracting change information from the entire cognitive interaction sequence set. Through the above data organization and processing, the original cognitive interaction sequence set can be transformed into a continuous comparison structure composed of multiple adjacent cognitive interaction operation pairs, providing a clear data foundation and processing path for subsequent analysis of state change processes.

[0029] S202. In the successive comparison process, the operation characterization markers in adjacent cognitive interaction operation pairs are processed by content parsing. The user identifier, operation content and occurrence order in the operation characterization markers are compared to identify the change in the state corresponding to the previous cognitive interaction operation being replaced by the next cognitive interaction operation. Adjacent cognitive interaction operation pairs that meet the replacement conditions are marked as coverage features in the state change process to determine whether there are multiple users continuously inputting operations and the training state is continuously covered. S203. For adjacent cognitive interaction pairs marked as covering features, perform continuity detection according to the order of occurrence, divide the continuously occurring covering features into intervals, and sequentially merge the covering features within the same interval to form a corresponding set of covering segments.

[0030] When processing adjacent cognitive interaction pairs marked as covering features, all covering features can be organized into a sequential list according to their occurrence order, and continuity detection can be performed through sequential traversal. Specifically, a sliding window scan or adjacent element comparison method can be used to sequentially associate the current covering feature with the previous covering feature. When the occurrence order field maintains a progressive relationship, it is considered to occur consecutively. Based on the continuity detection, consecutively occurring covering features are divided into consecutive segments within the same time interval, and the interval division is achieved by marking the start and end positions of the interval. Subsequently, the covering features within each interval are sequentially merged according to their original order, which can be done through chaining or sequential container splicing. Coverage features within an interval are integrated into a continuous set of units, thus forming a set of coverage segments. For example, in a set of experimental data, if "User A's operation is replaced by User B," "User B's operation is replaced by User C," and "User C's operation is replaced by User D" are detected consecutively, and the corresponding occurrences occur in a progressive order, then the three coverage features are divided into the same interval and sequentially merged to form a set of coverage segments. If subsequent intervals or interruptions occur, they are re-divided as new intervals. Through this continuous detection and interval division process, discrete coverage features can be transformed into continuous segment data, enabling subsequent analysis to be based on complete state change fragments, thereby improving the stability and traceability of data organization.

[0031] Adjacent cognitive interaction pairs, labeled as coverage features, represent combinations of operations that occur consecutively in time and have a substitution relationship. They form the basic data units for subsequent continuity analysis. Continuity detection is used to determine the continuous occurrence of coverage features in a time series. Its core lies in comparing adjacent coverage features one by one based on the occurrence order field. Continuously occurring coverage features refer to a set of coverage records that progress without interruption in time. Interval partitioning is used to divide continuously occurring coverage features into multiple independent segments according to temporal continuity, so that each segment corresponds to a complete state change process. Sequential merging is used to connect coverage features within the same interval according to the occurrence order to form a continuous chain structure. Coverage segment set is used to represent complete interval data formed by the combination of multiple coverage features. It contains multiple coverage features arranged in order. This set can reflect the continuous coverage changes of the training state over a period of time, thus providing a structured data foundation for subsequent state evolution analysis.

[0032] In this embodiment, S202 specifically refers to: During the successive comparison process, the operation characterization markers in adjacent cognitive interaction operation pairs are split into fields, and the operation characterization markers are parsed into user identifier field, operation content field and occurrence sequence field, and each field is read independently. Based on field splitting, the user identifier field, operation content field and occurrence order field in adjacent cognitive interaction operation pairs are compared item by item. The relationship between the preceding and following operations is determined according to the occurrence order field. When the operation content field changes and the operation content corresponding to the subsequent cognitive interaction operation replaces the operation content corresponding to the previous cognitive interaction operation, the change in the state corresponding to the previous cognitive interaction operation being replaced by the subsequent cognitive interaction operation is identified. After identifying the substitution change, adjacent cognitive interaction pairs that meet the substitution conditions are marked. The corresponding operation characterization is recorded as the coverage feature in the state change process. The coverage feature is continuously detected in the order of occurrence. When the user identifier field corresponding to the consecutively occurring coverage feature contains different user identifiers and the occurrence order field is continuously increasing, it is determined that multiple users have continuously input operations and the training state is continuously covered.

[0033] During processing, operation characterization markers are typically organized with a fixed structure, including user identification fields, operation content fields, and occurrence sequence fields. Each field has a clear location or identifier in data storage. During successive comparisons, the operation characterization markers can be split into fields using a structured parsing program. The original data is segmented according to field boundaries and loaded into independent cache units, thereby enabling independent reading of the user identification field, operation content field, and occurrence sequence field. In specific implementation, a sequential scanning method can be used to read the data buffer, and the content of each field can be extracted through field separators or fixed-length slices. For example, in a set of experimental data, there is a continuous operation characterization marker sequence "User A - Select Target - Sequence 1" and "User B - Correct Target - Sequence 2". After field splitting, the user A and user B, the selection and correction operations, and the corresponding sequence information can be obtained respectively, thus establishing subsequent comparisons on a unified field dimension.

[0034] After the fields are split, the user identifier field, operation content field, and occurrence order field of adjacent cognitive interaction operation pairs are compared item by item. First, the occurrence order field is used to determine the sequential relationship between the previous and subsequent cognitive interaction operations. Then, the operation content field is matched and changes are detected. When the operation content corresponding to the subsequent cognitive interaction operation covers the operation content corresponding to the previous cognitive interaction operation, it is identified as a substitution change. In specific implementation, the operation content field can be mapped to the task semantic category after reading each field. Then, it is determined whether the subsequent semantic category substitutes for the previous semantic category. For example, "correct target" covers "select target", and "reselect" substitutes for "original selection". In the experimental data, when "user A performs a selection operation" and "user B performs a correction operation to change the selection result" appear consecutively, the substitution relationship can be identified by comparing the operation content field.

[0035] After identifying substitution changes, adjacent cognitive interaction pairs that meet the substitution conditions are marked. The corresponding operation characterization is recorded as coverage features in the state change process and continuously detected in the time series. In specific implementation, a sliding window method can be used to scan the coverage features, comparing the current coverage feature with the previous coverage feature. When adjacent coverage features maintain a continuous progression in the occurrence order field, they are determined to appear consecutively. At the same time, a set judgment is performed on the user identifier field. When different user identifiers appear in the consecutive coverage features, it is considered that multiple users are involved. For example, if the experimental data shows consecutive records of "User A is replaced by User B" and "User B is replaced by User C" in a continuous and progressive order, it can be determined that multiple users have continuously input operations and the training state is continuously covered.

[0036] The successive comparison process refers to the process of analyzing adjacent cognitive interaction pairs sequentially according to the arrangement order of the cognitive interaction sequence set. This process is based on time order and achieves overall change recognition through comparison of local adjacent data. Field splitting is used to parse the operation characterization mark into user identification field, operation content field, and occurrence sequence field, so that different types of information can be processed separately. The user identification field is used to distinguish different user sources, the operation content field is used to represent specific interaction behavior, and the occurrence sequence field is used to characterize the time sequence of the operation. Independent reading refers to accessing and processing each field separately, thereby achieving item-by-item comparison.

[0037] When the content field of an operation changes and the content of a subsequent cognitive interaction replaces the content of the previous cognitive interaction, it indicates that the subsequent operation semantically alters the effect of the previous operation. Adjacent cognitive interaction pairs that satisfy the substitution condition refer to combinations of operations that form a covering relationship when they are continuous in time and the content of the operation changes. Coverage features during state changes are used to identify this substitution relationship. Continuous detection is used to identify the continuous occurrence of coverage features in the time series. User identifier fields corresponding to continuously occurring coverage features contain different user identifiers, indicating that the operations involved in the substitution come from different users. The occurrence order field is continuously increasing to characterize the continuity of operations in time. Through the combination of these conditions, the process of continuous input from multiple users leading to continuous coverage of the training state can be identified from continuous data, thus providing a stable data foundation for subsequent processing.

[0038] S3. Perform reverse expansion processing on the cognitive interaction operations corresponding to the covered segment set, use the evolution index identifier to connect the covered training states, form cognitive interaction information during the connection process, and generate a continuously changing trajectory. In this embodiment, S3 specifically includes the following steps: S301. The cognitive interaction operations corresponding to the covered segment set are arranged in reverse order according to the order of occurrence. Starting from the end of the covered segment set, the cognitive interaction operations are extracted sequentially and formed in reverse order to form a reverse unfolding sequence, so as to realize the reverse unfolding process of the covered process. The cognitive interaction operations corresponding to the covered segment set are usually stored in a continuous data structure in chronological order. Each data segment represents a sequence of cognitive interaction operations that have an overlapping relationship within the same time interval. When processing this sequence, the original order is reversed by reversing the order. This can be achieved by using array index reverse access or chain structure tail node backtracking. Cognitive interaction operations are extracted sequentially starting from the end of the covered segment set, and the extracted results are written into a new storage area in the order of reading, thus forming a reverse unfolded sequence. For example, in a set of experimental data collection, the continuously recorded cognitive interaction operations are "User A performs a selection operation", "User B performs a correction operation", and "User C performs a further correction operation". Since the subsequent operations cover the previous states, the early states in the original sequence are difficult to directly participate in the subsequent analysis. After reversing the order, a reverse unfolded sequence of "User C operation", "User B operation", and "User A operation" is obtained, so that the covered cognitive interaction operations re-enter the processing flow in reverse order. In implementation, a tail pointer can be set to read data step by step from the end position and push the reading results into the new sequence structure in sequence, thus completing the reverse unfolding process. This process can explicitly express the covered state change path again.

[0039] The cognitive interaction operations corresponding to the coverage segment set represent a set of cognitive interaction operations that have a continuous substitution relationship within the same coverage interval, arranged in the order of occurrence. Reverse order processing refers to rearranging the cognitive interaction operations in this set in reverse order of occurrence, so that cognitive interaction operations that occur later in time are processed first. The end of the coverage segment set represents the position of the last cognitive interaction operation in the order of occurrence and is the starting position of the reverse order processing. The reverse unfolding sequence is a sequence of cognitive interaction operations obtained through reverse order processing. The order of this sequence is the reverse of the original time order and is used to recover the historical state change process covered by subsequent cognitive interaction operations. Through this data rearrangement and recombination, cognitive interaction operations that were originally covered in the forward sequence can be reintroduced into the processing chain, so that the subsequent training state concatenation based on the evolution index has a complete data source.

[0040] S302. Based on the reverse expansion sequence, construct an evolution index identifier for each cognitive interaction operation, take the occurrence order as the index main order, and embed the user identifier and operation content into the evolution index identifier so that the evolution index identifier corresponds one-to-one with the cognitive interaction operation. After the reverse unfolding sequence is formed, each cognitive interaction operation corresponds to a clear order of occurrence and specific interaction content. These cognitive interactions can be uniformly identified and located by constructing an evolutionary index identifier. In implementation, the order of occurrence can be used as the primary index order, and each cognitive interaction operation can be sorted and located by sequence number or sequence label. Based on this primary order, the user identifier and operation content are embedded into the same data structure to form a composite identifier, so that each evolutionary index identifier simultaneously possesses temporal sequence information and behavioral semantic information. Specifically, field concatenation or structured data encapsulation can be used, with the order of occurrence field as the primary sorting criterion, and then the user identifier and operation content are added to form a complete index item. Furthermore, a one-to-one correspondence between evolution index identifiers and cognitive interaction operations is established through a mapping table. For example, in a set of experimental data, the reverse unfolding sequence is "User C performs a correction operation again", "User B performs a correction operation", and "User A performs a selection operation". Evolution index identifiers can be constructed for each of the three cognitive interaction operations, marking the order of occurrence as the reverse position, and embedding the corresponding user identifier and operation content, so that each cognitive interaction operation can be uniquely located and accessed through the evolution index identifier. Through this construction method, cognitive interaction operations can be directly retrieved and associated based on the evolution index identifier in subsequent processing, providing a stable data index foundation for the sequential concatenation of training states.

[0041] The evolution index identifier is used to describe the unique identification information of cognitive interaction operations in the overall evolution process. It contains key fields such as occurrence order, user identifier, and operation content. The occurrence order describes the chronological relationship of cognitive interaction operations and is the core basis of the index master order. The index master order refers to the sorting rules established based on the occurrence order, so that all cognitive interaction operations can be arranged and accessed in a unified order. The user identifier is used to distinguish the different sources of participating users, and the operation content is used to represent the semantics of specific interaction behaviors. Embedding the user identifier and operation content into the evolution index identifier enables the index to not only have positioning functions but also behavior description capabilities. Through the one-to-one correspondence between the evolution index identifier and the cognitive interaction operation, a direct mapping from the index to the specific operation can be achieved. This allows for the accurate acquisition of the corresponding operation and its state change information based on the index order during subsequent training state concatenation, ensuring the continuity and consistency of the data processing process.

[0042] S303. Based on the evolution index identifier, sequentially connect the training states corresponding to the cognitive interaction operations in the reverse unfolding sequence, connect the training states corresponding to adjacent cognitive interaction operations in the order of occurrence, and extract the user identifier, operation content and training state change information during the connection process to form cognitive interaction information. S304. Based on the cognitive interaction information, the training state change information is continuously spliced ​​according to the order of occurrence, and adjacent training state change information is sequentially connected to generate the corresponding continuous change trajectory.

[0043] After the cognitive interaction information is formed, it already contains training state change information arranged in chronological order. Each training state change corresponds to a change record between adjacent training states. Based on this, continuous concatenation processing can be performed on the training state change information. Specifically, this can be achieved by sequentially traversing the structured dataset, reading adjacent training state change information one by one, and concatenating them in chronological order. The ending state of the previous training state change information is connected to the starting state of the next training state change information, thus constructing a continuous change chain. In the implementation process, a chain structure or sequential container can be used, treating each training state change information as a node unit. By establishing forward-backward relationships to achieve continuous connections, and merging or connecting state change descriptions during the splicing process, the change process remains semantically coherent. For example, in a set of experimental data, training state change information is continuously recorded as "from initial state to target selection state", "from target selection state to correction state", and "from correction state to further correction state". Through continuous splicing processing, the three segments of change information can be sequentially connected to form a complete continuous change trajectory, so that the change path of the training state from the initial to the final can be fully expressed. This processing can integrate discrete change information into a continuous structure, thereby supporting subsequent trajectory analysis and state control.

[0044] Training state change information describes the changes between adjacent training states, including information about the state's starting and ending points. Continuous splicing refers to connecting multiple training state change information entries in sequence to form a continuous chain of changes. Adjacent training state change information represents two consecutive state change records occurring in time, with their connection established by the ending point of the previous change and the starting point of the next. A continuous change trajectory is the overall change path formed by sequentially connecting training state change information entries, representing the entire process of training state evolution over time. Through continuous splicing, consistency in both the temporal and semantic dimensions of each training state change information entry can be ensured, enabling the continuous change trajectory to accurately reflect the complete evolution of states during group collaborative cognitive interaction training, thus providing a reliable data foundation for subsequent analysis.

[0045] In this embodiment, S303 specifically refers to: Based on the evolution index identifier, the cognitive interaction operations in the reverse unfolding sequence are sorted and mapped. The training states corresponding to each cognitive interaction operation are mapped according to the order of occurrence, and the sequential connection relationship between adjacent cognitive interaction operations is established. Based on the sorting mapping process, the training states corresponding to adjacent cognitive interaction operations are connected one by one. Adjacent training states are read in sequence according to the order of occurrence, and the state identifiers and state contents in the training states are connected and combined to form a continuous training state sequence. During the concatenation of the training state sequence, the corresponding user identifier, operation content, and training state change information are extracted for each connection node and recorded and integrated in the order of occurrence. The continuously extracted user identifier, operation content, and training state change information are sequentially associated to form corresponding cognitive interaction information.

[0046] After the cognitive interaction operations in the reverse unfolded sequence are constructed, they can be sorted and mapped using evolution index identifiers, mapping the training states corresponding to each cognitive interaction operation to a unified sequence space. In specific implementation, an index table based on evolution index identifiers can be established, using the order of occurrence as the sorting criterion to reorder the cognitive interaction operations in the reverse unfolded sequence, and assigning continuous position identifiers to the training states corresponding to each cognitive interaction operation. At the same time, connection pointers are established between adjacent positions to describe the sequential connection relationship between adjacent cognitive interaction operations. For example, in the experimental data, the reverse unfolded sequence contains "User C performs a correction operation", "User B performs a correction operation", and "User A performs a selection operation". By parsing the order of occurrence information in the evolution index identifiers, the sequence structure of "User A operation", "User B operation", and "User C operation" can be restored, and continuous positions can be assigned to each training state. At the same time, connection relationships between adjacent states are established, so that the training states form a logically ordered arrangement structure.

[0047] After completing the sorting and mapping process, pairwise connection processing can be performed on the training states corresponding to adjacent cognitive interaction operations. By sequentially traversing the position mapping results, adjacent training states are read in turn, and the state identifiers and state contents in the training states are connected and combined. In specific implementation, each training state can be represented as a data unit containing a state identifier and state content. By establishing a forward-backward pointing relationship or a sequential container structure, the current training state is connected to the next training state, and the state identifier and state content are concatenated or associated and stored to form a continuous training state sequence. In the experimental data, when there are continuous states "initial selection state", "correction state" and "further correction state", the three states can be connected sequentially through pairwise connection processing to form a continuous state chain, so that the state change process is fully presented.

[0048] During the formation of the training state sequence, information extraction processing can be performed on each connection node to extract training state change information between the current training state and adjacent training states, and integrate it with the user identifier and operation content of the corresponding cognitive interaction operation. In specific implementation, the state change part can be identified by comparing the state identifier and state content of adjacent training states, and the change content can be bound with the user identifier and operation content of the corresponding cognitive interaction operation to form a structured record unit. Then, these record units are written into the sequence structure in the order of occurrence to form continuous cognitive interaction information. For example, in the experimental data, from "User A selects a target" to "User B corrects the target", and then to "User C corrects the target again", the state change content of each node can be extracted and combined with the corresponding user identifier and operation content to form a complete information sequence.

[0049] The sorting mapping process rearranges the cognitive interaction operations in the reverse unfolding sequence according to the evolution index identifier, so that the training state corresponding to each cognitive interaction operation has a clear sequential position; the training state corresponding to each cognitive interaction operation represents the system state data formed after the occurrence of the cognitive interaction operation; the position mapping process assigns sequential position identifiers to the training states and establishes sequential connection relationships between adjacent cognitive interaction operations, so that the data forms a logically continuous structure; the pairwise connection process connects adjacent training states one by one, so that the state change process can be continuously expressed; the state identifier is used to distinguish different training states, the state content is used to describe specific state information, and the continuously connected training state sequence represents the state chain structure formed by connecting multiple training states in sequence.

[0050] User identifiers are used to identify the source of users participating in cognitive interaction; operation content describes the behavioral semantics of cognitive interaction operations; training state change information is used to characterize the differences between adjacent training states; sequential association processing is used to connect the extracted information according to the order of occurrence, so that the information maintains a temporal continuity; cognitive interaction information represents a structured information set formed by the combination of user identifiers, operation content, and training state change information, which is used to describe the behavioral and state evolution path in the process of group collaborative cognitive interaction training; through the above processing, discrete cognitive interaction operations and training state changes can be transformed into continuous structured data, so that subsequent trajectory generation and dynamic control have a complete and traceable data foundation.

[0051] S4. Analyze the continuously changing trajectory, perform calculations according to the order of the evolution index identifiers, and generate an adjustment identifier sequence; In this embodiment, S4 specifically refers to: The continuously changing trajectory is parsed and processed. The training state change information in the continuously changing trajectory is extracted in the order of occurrence. The state identifier, state content and evolution index identifier corresponding to each training state change information are separated and read. At the same time, the parsing results are sorted and rearranged according to the occurrence order information in the evolution index identifier. The sorted data are written into the storage unit in the order of occurrence to form a parsing data structure with sequential pointing relationship. Based on the parsed data structure, the operation is performed in the order of the evolution index identifier. By sequentially traversing the adjacent training state change information in the parsed data structure, the state identifier and state content in the adjacent training state change information are read in turn. The state identifier of the previous training state change information is matched with the state identifier of the next training state change information. At the same time, the differences between the state content of the previous training state change information and the state content of the next training state change information are extracted and combined to generate the corresponding state change operation data. For state change operation data, an identifier construction process is performed. The state change operation data is associated with the corresponding evolution index identifier and encoded. The data is then arranged in the order of the evolution index identifier to generate an adjustment identifier sequence.

[0052] After the continuous change trajectory is formed, the continuous change process of the training state is recorded in a sequential storage structure. Each training state change information contains a state identifier, state content, and evolution index identifier. During the parsing process, the continuous change trajectory can be read one by one through sequential scanning and field splitting. The state identifier, state content, and evolution index identifier of each training state change information are extracted and written to a temporary cache area. At the same time, the extraction results are sorted and rearranged according to the occurrence order information in the evolution index identifier. Specifically, an index mapping table is constructed, and the evolution index identifier is used as the sorting key to reorder all parsing results. The data is then written to the storage unit in sequence according to the sorting result. During the writing process, a forward and backward pointing relationship is established for adjacent data, so that each data contains an association identifier pointing to the previous and next data, thereby forming a parsing data structure with a sequential pointing relationship. For example, in the experimental acquisition, the continuous change trajectory records "state one changes to state two", "state two changes to state three", and "state three changes to state four". After parsing and sorting, it is restored to a unified order and a data structure with a forward and backward pointing relationship is formed in the storage unit, so that each state change information can be accessed and located sequentially.

[0053] After the data structure is established, the computation is performed sequentially, selecting adjacent training state change information for calculation. During processing, the state identifiers of the previous and subsequent training state change information are first matched to confirm their continuity in time, using the state identifiers as the basis for state connection. Then, the differences between the state content of the previous and subsequent training state change information are extracted. Specifically, this is done by comparing each field to identify the changed content and separating it from the state content. The change result of the previous state is then combined with the change content of the subsequent state to form a complete state change description. For example, in the experimental data, the state changes from "target not selected" to "target selected" and then to "target corrected." The continuity of the state is confirmed by matching the state identifiers, and the changes from "not selected" to "selected" and "corrected" are extracted by comparing the state content. These changes are then combined to form structured state change computation data, ensuring that each data point simultaneously contains the relationship between the preceding and following states and the changes themselves.

[0054] After obtaining the state change operation data, the identifier construction process generates an identifier by encoding each operation data. The state change operation data is associated with the corresponding evolution index identifier. Specifically, the evolution index identifier can be used as the main identifier field, and the changes in the state change operation data can be used as additional fields. A unified identifier unit is formed by field concatenation or structured encapsulation, and written into the identifier storage area in the order of the evolution index identifiers to form a continuously arranged adjustment identifier sequence. During the experiment, for the state change caused by the selection operation, the state change caused by the correction operation, and the state change caused by the correction operation again, corresponding identifiers are generated and arranged in the order of occurrence, so that the adjustment identifier sequence can completely reflect the state evolution process in the continuous change trajectory.

[0055] The continuous change trajectory describes the continuous change path of the training state in the time dimension. The parsing process is used to transform the continuous change trajectory into structured data and extract the training state change information. The training state change information includes state identifier, state content, and evolution index identifier, which are the basic units used to describe the state change. Separate reading is used to extract different field information independently for subsequent processing. The sorted data represents a data set rearranged according to the evolution index identifier. The storage unit is used to hold the parsed data and establishes a parsed data structure with sequential pointing relationship through the writing process, so that the data have a clear sequential association.

[0056] The computational processing is used to process adjacent training state change information, which represents two consecutive state change records in time. State identifiers are used to determine state positions and connections, while state content describes specific state information. Matching the state identifiers of the previous and subsequent training state change information confirms continuity. Difference extraction and combination processing between the state content of the previous and subsequent training state change information extracts the state change content. State change computational data represents the processed change result. Identifier construction processing generates adjustment identifiers, and association encoding combines the state change computational data with the evolution index identifier. The adjustment identifier sequence represents a set of sequentially arranged control identifiers, thus fully reflecting the training state change process and providing data support for subsequent control.

[0057] S5. Match and control the continuously changing trajectory based on the adjustment label sequence, and dynamically regulate the state update process of group collaborative cognitive interaction training by combining cognitive interaction information.

[0058] In this embodiment, S5 specifically refers to: The continuous trajectory is matched and controlled based on the adjustment identifier sequence. The adjustment identifiers in the adjustment identifier sequence are read sequentially according to the order of the evolution index identifiers. Each adjustment identifier is matched with the training state change information in the continuous trajectory. By comparing the identifier content in the adjustment identifier with the state identifier and state content in the training state change information, the corresponding matching position is determined. At the corresponding matching position, the training state change information is rearranged and the content is replaced to complete the matching control of the continuous trajectory. Based on the continuous change trajectory of the matching control, the state update process of the group collaborative cognitive interaction training is dynamically regulated by combining cognitive interaction information. The user identifier, operation content and training state change information in the cognitive interaction information are read in sequence according to the occurrence order. The cognitive interaction information and the matched training state change information are correlated and processed accordingly. Based on the correlation results, the update order and update content of the training state change information are adjusted to form an update driving sequence for use in the state update process.

[0059] Before processing, a mapping relationship based on evolution index identifiers is constructed between the adjustment identifier sequence and the continuous change trajectory. The adjustment identifier sequence is sequentially traversed, and the identifier content is extracted line by line. The evolution index identifier is then used to locate the corresponding training state change information in the continuous change trajectory. The location process involves matching the state identifiers in the adjustment identifiers with the state identifiers in the training state change information, while also performing secondary verification based on the state content to ensure the uniqueness of the matching position. After determining the corresponding matching position, when performing rearrangement processing for that position, the forward and backward pointing relationships of the training state change information in the storage structure can be adjusted to change the position of the target training state change information in the continuous change trajectory. Simultaneously, during content replacement processing, the state content in the adjustment identifier is written into the training state change information corresponding to the matching position, overwriting the original state content. For example, in experimental data containing "selecting a state change record," "correcting a state change record," and "correcting a state change record again," when the adjustment identifier points to the corrected state change record, the order can be shifted forward by changing the pointing relationship between that record and its adjacent records. Simultaneously, the corrected content in the adjustment identifier is written into the corresponding record, thus achieving precise matching control of the continuous change trajectory.

[0060] After matching control is completed, the continuously changing trajectories have formed a new arrangement structure. At this point, cognitive interaction information is introduced into the update chain. By reading the user identifier, operation content, and training state change information in the cognitive interaction information according to the order of occurrence, a corresponding association relationship is established between the cognitive interaction information and the matched training state change information. The association process can be achieved by constructing a mapping relationship based on the user identifier, binding the operation content corresponding to the same user with the training state change information, and simultaneously using the operation content to correct the state content in the training state change information. Based on this, when adjusting the update order of the training state change information, the training state change information can be reordered according to the order of occurrence in the cognitive interaction information. When adjusting the update content, the content is updated by mapping the operation content to the corresponding field in the training state change information. For example, in the experimental scenario, when the cognitive interaction information reflects that a certain user's operation has priority, the position of the corresponding training state change information of that user in the update chain can be adjusted so that the state change participates in the update first, and the state content is modified synchronously according to the operation content, thus forming an update-driven sequence.

[0061] The adjustment identifier sequence describes a set of identifiers used to control training state change information. Each adjustment identifier contains a state identifier and state content information, used to locate and modify training state change information. The continuous change trajectory describes the sequential structure of training state change information, with internal connections maintained by sequential pointing relationships. Matching control includes position-correspondence matching, rearrangement processing, and content replacement processing. Position-correspondence matching determines the position of the adjustment identifier, rearrangement processing adjusts the order by changing the pointing relationship, and content replacement processing updates the state content by writing data. Through these processes, the structure and content of the continuous change trajectory can be synchronously controlled.

[0062] Cognitive interaction information is used to describe behavioral data in group collaborative cognitive interaction training, including user identification, operation content, and training state change information. The state update process is formed by the sequential execution of training state change information. Dynamic control is achieved through the correspondence between cognitive interaction information and the matched training state change information. The update order adjustment is used to change the execution order of training state change information, and the update content adjustment is used to modify the state content in the training state change information. The update driving sequence represents the adjusted set of training state change information, which is executed sequentially in a predetermined order during the update process, so that the training state update process can respond to changes in cognitive interaction information and realize the dynamic control of group collaborative cognitive interaction training.

[0063] like Figure 2 The cognitive interaction training system for group collaboration shown includes a cognitive interaction modeling module, a coverage feature recognition module, a state evolution reconstruction module, a regulation label generation module, and a dynamic regulation execution module. The cognitive interaction modeling module continuously records the cognitive interaction operations generated by multiple users around the same training objective, and generates operation characterization tags for each cognitive interaction operation, forming a set of cognitive interaction sequences according to user identifier, operation content and occurrence order; The coverage feature recognition module compares adjacent cognitive interaction operations in the cognitive interaction sequence set one by one. It identifies the coverage features in the state change process by using operation characterization markers to determine whether there are multiple users continuously inputting operations and the training state is continuously covered, and forms a set of coverage segments. The state evolution reconstruction module performs reverse expansion processing on the cognitive interaction operations corresponding to the covered segment set, uses the evolution index identifier to connect the covered training states, forms cognitive interaction information during the connection process, and generates a continuous change trajectory. The adjustment identifier generation module analyzes and processes the continuously changing trajectory, performs calculations according to the order of the evolution index identifiers, and generates an adjustment identifier sequence. The dynamic control execution module matches and controls the continuously changing trajectory based on the adjustment identifier sequence, and dynamically controls the state update process of the group collaborative cognitive interaction training by combining cognitive interaction information.

[0064] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0065] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0066] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0067] In the several embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0068] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0069] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0070] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A cognitive interaction training method for group collaboration, characterized in that, Specifically, the following steps are included: S1. Continuously record the cognitive interaction operations generated by multiple users around the same training objective, and generate operation characterization marks for each cognitive interaction operation, forming a set of cognitive interaction sequences according to user identification, operation content and occurrence order. S2. Compare adjacent cognitive interaction operations in the cognitive interaction sequence set one by one, identify the coverage features in the state change process by operation characterization markers, and determine whether there are multiple users continuously inputting operations and the training state is continuously covered, and form a set of coverage segments. S3. Perform reverse expansion processing on the cognitive interaction operations corresponding to the covered segment set, use the evolution index identifier to connect the covered training states, form cognitive interaction information during the connection process, and generate a continuous change trajectory. S4. Analyze the continuously changing trajectory, perform calculations according to the order of the evolution index identifiers, and generate an adjustment identifier sequence; S5. Match and control the continuously changing trajectory based on the adjustment label sequence, and dynamically regulate the state update process of group collaborative cognitive interaction training by combining cognitive interaction information.

2. The cognitive interaction training method for group collaboration according to claim 1, characterized in that, S1 specifically refers to: The cognitive interaction operations generated by multiple users around the same training objective are continuously recorded. Each cognitive interaction operation is sequentially assigned a user identifier, operation content, and occurrence time identifier. The cognitive interaction operations are then arranged in order according to the occurrence time identifier to form a continuously recorded sequence of cognitive interaction operations. For each cognitive interaction operation in the cognitive interaction operation sequence, an operation characterization mark generation process is performed. The user identifier, operation content and occurrence time identifier are combined and encoded to generate the corresponding operation characterization mark, and the operation characterization mark is kept in a one-to-one correspondence with the cognitive interaction operation. The operation markers are sorted according to the occurrence time, categorized and integrated according to the user identifier, and then sequentially linked with the operation content to form a set of cognitive interaction sequences that include user identifier, operation content, and occurrence order.

3. The cognitive interaction training method for group collaboration according to claim 1, characterized in that, S2 specifically includes the following steps: S201. Traverse the cognitive interaction operations in the cognitive interaction sequence set in the order of occurrence, pair adjacent cognitive interaction operations in turn to form adjacent cognitive interaction operation pairs, and extract the operation characterization mark corresponding to each adjacent cognitive interaction operation pair for successive comparison. S202. In the successive comparison process, the operation characterization markers in adjacent cognitive interaction operation pairs are processed by content parsing. The user identifier, operation content and occurrence order in the operation characterization markers are compared to identify the change in the state corresponding to the previous cognitive interaction operation being replaced by the next cognitive interaction operation. Adjacent cognitive interaction operation pairs that meet the replacement conditions are marked as coverage features in the state change process to determine whether there are multiple users continuously inputting operations and the training state is continuously covered. S203. For adjacent cognitive interaction pairs marked as covering features, perform continuity detection according to the order of occurrence, divide the continuously occurring covering features into intervals, and sequentially merge the covering features within the same interval to form a corresponding set of covering segments.

4. The cognitive interaction training method for group collaboration according to claim 3, characterized in that, S202 specifically refers to: During the successive comparison process, the operation characterization markers in adjacent cognitive interaction operation pairs are split into fields, and the operation characterization markers are parsed into user identifier field, operation content field and occurrence sequence field, and each field is read independently. Based on field splitting, the user identifier field, operation content field and occurrence order field in adjacent cognitive interaction operation pairs are compared item by item. The relationship between the preceding and following operations is determined according to the occurrence order field. When the operation content field changes and the operation content corresponding to the subsequent cognitive interaction operation replaces the operation content corresponding to the previous cognitive interaction operation, the change in the state corresponding to the previous cognitive interaction operation being replaced by the subsequent cognitive interaction operation is identified. After identifying the substitution change, adjacent cognitive interaction pairs that meet the substitution conditions are marked. The corresponding operation characterization is recorded as the coverage feature in the state change process. The coverage feature is continuously detected in the order of occurrence. When the user identifier field corresponding to the consecutively occurring coverage feature contains different user identifiers and the occurrence order field is continuously increasing, it is determined that multiple users have continuously input operations and the training state is continuously covered.

5. The cognitive interaction training method for group collaboration according to claim 1, characterized in that, S3 specifically includes the following steps: S301. The cognitive interaction operations corresponding to the covered segment set are arranged in reverse order according to the order of occurrence. Starting from the end of the covered segment set, the cognitive interaction operations are extracted sequentially and formed in reverse order to form a reverse unfolding sequence, so as to realize the reverse unfolding process of the covered process. S302. Based on the reverse expansion sequence, construct an evolution index identifier for each cognitive interaction operation, take the occurrence order as the index main order, and embed the user identifier and operation content into the evolution index identifier so that the evolution index identifier corresponds one-to-one with the cognitive interaction operation. S303. Based on the evolution index identifier, sequentially connect the training states corresponding to the cognitive interaction operations in the reverse unfolding sequence, connect the training states corresponding to adjacent cognitive interaction operations in the order of occurrence, and extract the user identifier, operation content and training state change information during the connection process to form cognitive interaction information. S304. Based on the cognitive interaction information, the training state change information is continuously spliced ​​according to the order of occurrence, and adjacent training state change information is sequentially connected to generate the corresponding continuous change trajectory.

6. The cognitive interaction training method for group collaboration according to claim 5, characterized in that, S303 specifically refers to: Based on the evolution index identifier, the cognitive interaction operations in the reverse unfolding sequence are sorted and mapped. The training states corresponding to each cognitive interaction operation are mapped according to the order of occurrence, and the sequential connection relationship between adjacent cognitive interaction operations is established. Based on the sorting mapping process, the training states corresponding to adjacent cognitive interaction operations are connected one by one. Adjacent training states are read in sequence according to the order of occurrence, and the state identifiers and state contents in the training states are connected and combined to form a continuous training state sequence. During the concatenation of the training state sequence, the corresponding user identifier, operation content, and training state change information are extracted for each connection node and recorded and integrated in the order of occurrence. The continuously extracted user identifier, operation content, and training state change information are sequentially associated to form corresponding cognitive interaction information.

7. The cognitive interaction training method for group collaboration according to claim 1, characterized in that, S4 specifically refers to: The continuously changing trajectory is parsed and processed. The training state change information in the continuously changing trajectory is extracted in the order of occurrence. The state identifier, state content and evolution index identifier corresponding to each training state change information are separated and read. At the same time, the parsing results are sorted and rearranged according to the occurrence order information in the evolution index identifier. The sorted data are written into the storage unit in the order of occurrence to form a parsing data structure with sequential pointing relationship. Based on the parsed data structure, the operation is performed in the order of the evolution index identifier. By sequentially traversing the adjacent training state change information in the parsed data structure, the state identifier and state content in the adjacent training state change information are read in turn. The state identifier of the previous training state change information is matched with the state identifier of the next training state change information. At the same time, the differences between the state content of the previous training state change information and the state content of the next training state change information are extracted and combined to generate the corresponding state change operation data. For state change operation data, an identifier construction process is performed, which associates and encodes the state change operation data with the corresponding evolution index identifier, and arranges them according to the order of the evolution index identifier to generate an adjustment identifier sequence.

8. The cognitive interaction training method for group collaboration according to claim 1, characterized in that, S5 specifically refers to: The continuous trajectory is matched and controlled based on the adjustment identifier sequence. The adjustment identifiers in the adjustment identifier sequence are read sequentially according to the order of the evolution index identifiers. Each adjustment identifier is matched with the training state change information in the continuous trajectory. By comparing the identifier content in the adjustment identifier with the state identifier and state content in the training state change information, the corresponding matching position is determined. At the corresponding matching position, the training state change information is rearranged and the content is replaced to complete the matching control of the continuous trajectory. Based on the continuous change trajectory of the matching control, the state update process of the group collaborative cognitive interaction training is dynamically regulated by combining cognitive interaction information. The user identifier, operation content and training state change information in the cognitive interaction information are read in sequence according to the occurrence order. The cognitive interaction information and the matched training state change information are correlated and processed accordingly. Based on the correlation results, the update order and update content of the training state change information are adjusted to form an update driving sequence for use in the state update process.

9. A cognitive interaction training system for group collaboration, used to implement the cognitive interaction training method for group collaboration as described in any one of claims 1-8, characterized in that, It includes a cognitive interaction modeling module, a coverage feature recognition module, a state evolution reconstruction module, a regulation identifier generation module, and a dynamic regulation execution module; The cognitive interaction modeling module continuously records the cognitive interaction operations generated by multiple users around the same training objective, and generates operation characterization tags for each cognitive interaction operation, forming a set of cognitive interaction sequences according to user identifier, operation content and occurrence order; The coverage feature recognition module compares adjacent cognitive interaction operations in the cognitive interaction sequence set one by one. It identifies the coverage features in the state change process by using operation characterization markers to determine whether there are multiple users continuously inputting operations and the training state is continuously covered, and forms a set of coverage segments. The state evolution reconstruction module performs reverse expansion processing on the cognitive interaction operations corresponding to the covered segment set, uses the evolution index identifier to connect the covered training states, forms cognitive interaction information during the connection process, and generates a continuous change trajectory. The adjustment identifier generation module analyzes and processes the continuously changing trajectory, performs calculations according to the order of the evolution index identifiers, and generates an adjustment identifier sequence. The dynamic control execution module matches and controls the continuously changing trajectory based on the adjustment identifier sequence, and dynamically controls the state update process of the group collaborative cognitive interaction training by combining cognitive interaction information.

Images