An adaptive cognitive learning network simulation system based on a large language model

Abstract

The application relates to the technical field of cognitive learning, in particular to a self-adaptive cognitive learning network simulation system based on a large language model. The system comprises a user cognitive learning feature vector conversion module, a hierarchical cognitive reasoning and learning network construction module, a cognitive learning interaction bottleneck simulation analysis module and a cognitive learning network feedback optimization module, can acquire user historical cognitive learning data, user cognitive interaction record data and user cognitive learning feedback information data and perform cognitive feature vector fusion conversion to obtain a user cognitive learning feature vector; performs hierarchical cognitive reasoning analysis and self-adaptive cognitive learning network construction based on a large language model to generate an initial architecture of a user self-adaptive cognitive learning network, and performs cognitive learning interaction bottleneck simulation analysis and cognitive learning network feedback optimization adjustment to generate an optimized adjustment network of the user self-adaptive cognitive learning. The application can improve the learning efficiency and effect of the user cognitive learning process.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cognitive learning, and particularly relates to an adaptive cognitive learning network simulation system based on a large language model. BACKGROUND

[0002] With the rapid development of artificial intelligence technology, especially the application of large language models, the research on brain cognitive learning has gradually attracted attention. The simulation of cognitive learning network aims to simulate the cognitive learning process of human brain to improve learning efficiency and knowledge acquisition ability. By taking the large language model as the core component, its powerful natural language understanding and generation ability is utilized to provide rich semantic and knowledge support for the learning network. Through the analysis of user input, the model can understand the needs and knowledge background of learners. By introducing adaptive learning strategies, the learning path and content are dynamically adjusted according to the cognitive characteristics, learning progress and feedback information of learners. Through real-time analysis of user performance, the learning materials and difficulty are automatically optimized to ensure the personalization of the learning process. At the same time, by establishing a real-time feedback mechanism, feedback is provided in real time according to the answers and performance of learners to help them identify knowledge gaps and improve understanding. This mechanism can promote active learning and self-reflection. However, the traditional cognitive learning network simulation system cannot be individually adjusted according to the cognitive learning characteristics and needs of individual learners, and cannot effectively handle diverse knowledge structures and learning tasks, resulting in a lack of flexibility and effectiveness in the cognitive learning process. SUMMARY

[0003] Therefore, it is necessary to provide an adaptive cognitive learning network simulation system based on a large language model to solve at least one of the above technical problems.

[0004] To achieve the above-mentioned purpose, an adaptive cognitive learning network simulation system based on a large language model comprises the following modules:

[0005] A user cognitive learning feature vector conversion module is configured to obtain user historical cognitive learning data, user cognitive interaction record data and user cognitive learning feedback information data, and perform cognitive feature vector fusion conversion based on the user historical cognitive learning data, the user cognitive interaction record data and the user cognitive learning feedback information data to obtain a user cognitive learning feature vector.

[0006] A hierarchical cognitive reasoning and learning network construction module is configured to perform hierarchical cognitive reasoning analysis on the user cognitive learning feature vector based on the large language model to obtain a user cognitive knowledge low-level learning task and a user cognitive knowledge high-level learning reasoning relationship, and perform adaptive cognitive learning network construction based on the user cognitive knowledge low-level learning task and the user cognitive knowledge high-level learning reasoning relationship to generate an initial architecture of a user adaptive cognitive learning network.

[0007] The cognitive learning interaction bottleneck simulation analysis module is used to obtain real-time cognitive learning behavior data of users, and perform cognitive learning interaction bottleneck simulation analysis on the initial architecture of the user adaptive cognitive learning network based on the real-time cognitive learning behavior data of users, to obtain the bottleneck point of the user cognitive learning interaction simulation process;

[0008] The cognitive learning network feedback optimization module is used to perform cognitive learning network feedback optimization adjustment on the initial architecture of the user adaptive cognitive learning network based on the bottleneck points of the user cognitive learning interaction simulation process to generate a user adaptive cognitive learning optimization adjustment network.

[0009] Furthermore, the user cognitive learning feature vector analysis module includes the following functions:

[0010] Obtaining historical cognitive learning data records from the user cognitive learning platform database to obtain user historical cognitive learning data;

[0011] Conduct cognitive learning trajectory analysis on the user's historical cognitive learning data to obtain the user's historical cognitive learning activity trajectory;

[0012] Based on the user's historical cognitive learning activity trajectory, the user's historical cognitive learning data is analyzed and recorded to obtain user cognitive interaction record data;

[0013] Build a user cognitive interaction quality evaluation system based on user cognitive interaction record data, and conduct cognitive learning feedback analysis on user historical cognitive learning data based on the user cognitive interaction quality evaluation system to obtain user cognitive learning feedback information data;

[0014] The cognitive feature vector is fused and transformed according to the user's historical cognitive learning data, the user's cognitive interaction record data and the user's cognitive learning feedback information data to obtain the user's cognitive learning feature vector.

[0015] Furthermore, the cognitive learning trajectory analysis of the user's historical cognitive learning data includes:

[0016] Perform cognitive learning time and content extraction processing on the user's historical cognitive learning data to obtain the user's historical cognitive learning timestamp and the user's historical cognitive learning content;

[0017] Conduct cognitive learning theme mining and analysis on the user's historical cognitive learning content to obtain the theme of each user's historical cognitive learning activity;

[0018] Conduct correlation analysis on the themes of each user's historical cognitive learning activities to obtain the correlation relationship between the learning activity behaviors of each user's cognitive learning activity themes;

[0019] The user historical cognitive learning content is analyzed based on the learning activity behavior correlation relationship between the learning activity themes of each user cognitive learning activity, to generate a user historical cognitive learning behavior correlation graph;

[0020] The user historical cognitive learning activity trajectory is obtained by performing cognitive learning trajectory analysis on the user historical cognitive learning behavior correlation graph based on the user historical cognitive learning timestamp.

[0021] Further, the cognitive feature vector fusion conversion according to the user historical cognitive learning data, the user cognitive interaction record data and the user cognitive learning feedback information data includes:

[0022] The cognitive learning content features of the user historical cognitive learning data are analyzed to obtain user cognitive learning content features, wherein the user cognitive learning content features include user cognitive learning content themes, user cognitive learning difficulty levels and user cognitive learning time lengths;

[0023] The user interaction frequency features of the user cognitive interaction record data are analyzed to obtain user cognitive learning interaction frequency features, wherein the user cognitive learning interaction frequency features include user cognitive learning click rates, user cognitive learning viewing time lengths and user cognitive learning discussion participation degrees;

[0024] The cognitive learning feedback features of the user cognitive learning feedback information data are analyzed to obtain user cognitive learning feedback emotion features, wherein the user cognitive learning feedback emotion features include user cognitive learning positive feedback emotion indexes, user cognitive learning neutral feedback emotion indexes and user cognitive learning negative feedback emotion indexes;

[0025] The cognitive feature vector fusion conversion is performed on the user cognitive learning content features, the user cognitive learning interaction frequency features and the user cognitive learning feedback emotion features to obtain a user cognitive learning feature vector.

[0026] Further, the hierarchical cognitive reasoning and learning network construction module includes the following functions:

[0027] The user cognitive learning feature vector is analyzed based on the low-level cognitive module in the large language model to obtain a user cognitive knowledge low-level learning task;

[0028] The task content classification processing is performed on the user cognitive knowledge low-level learning task to obtain a user cognitive low-level learning task content classification matrix;

[0029] A high-level cognitive reasoning framework capable of processing low-level learning tasks is constructed by a high-level cognitive module in a large language model, and a high-level cognitive reasoning path analysis is performed on corresponding low-level learning tasks in a user cognitive low-level learning task content classification matrix based on the high-level cognitive reasoning framework, to obtain high-level cognitive reasoning paths between the low-level learning tasks;

[0030] Based on the high-level cognitive reasoning paths between the low-level learning tasks, high-level learning reasoning adjustment analysis is performed on corresponding user cognitive knowledge low-level learning tasks, to obtain user cognitive knowledge high-level learning reasoning relationships;

[0031] An adaptive cognitive learning network is constructed according to the user cognitive knowledge low-level learning tasks and the user cognitive knowledge high-level learning reasoning relationships, to generate an initial architecture of a user adaptive cognitive learning network.

[0032] Further, the high-level learning reasoning adjustment analysis of the corresponding user cognitive knowledge low-level learning tasks based on the high-level cognitive reasoning paths between the low-level learning tasks comprises:

[0033] Based on the high-level cognitive reasoning paths between the low-level learning tasks, cognitive reasoning path relationship connection processing is performed on corresponding user cognitive knowledge low-level learning tasks, to generate a high-level cognitive reasoning path relationship graph between the low-level learning tasks;

[0034] The reasoning relationship strength of the high-level cognitive reasoning path relationship graph between the low-level learning tasks is evaluated and calculated by using a learning task reasoning relationship strength calculation formula, to obtain the cognitive reasoning relationship strength between the low-level learning tasks;

[0035] Based on the cognitive reasoning relationship strength between the low-level learning tasks, high-level learning reasoning adjustment analysis is performed on the high-level cognitive reasoning path relationship graph between the low-level learning tasks, to obtain user cognitive knowledge high-level learning reasoning relationships.

[0036] Further, the learning task reasoning relationship strength calculation formula is specifically:

[0037]

[0038] In the formula, R ij is the cognitive reasoning relationship strength between the i th low-level learning task and the j th low-level learning task, i and j are item degree measurement parameters of the low-level learning tasks, T is an upper limit of a time range for reasoning relationship strength calculation, t is a time variable parameter, w ij is the reasoning relationship influence weight between the i th low-level learning task and the j th low-level learning task, s ij(t) is the number of inference paths between the i-th low-level learning task and the j-th low-level learning task at time t, a is the inference path number influence decay coefficient, c ij (t) is the task content similarity between the i-th low-level learning task and the j-th low-level learning task at time t, β is the task content similarity influence decay coefficient, p ij (t) is the context correlation parameter between the i-th low-level learning task and the j-th low-level learning task at time t, γ is the context correlation influence decay coefficient, η is the correction coefficient of cognitive inference relationship strength.

[0039] Further, the adaptive cognitive learning network construction according to the user cognitive knowledge low-level learning task and the user cognitive knowledge high-level learning inference relationship comprises:

[0040] The user cognitive knowledge low-level learning task is configured as a base node to obtain a user cognitive low-level learning task base node;

[0041] The user cognitive low-level learning task base node is converted into an adaptive learning node to obtain a user cognitive low-level adaptive learning node;

[0042] Based on the user cognitive knowledge high-level learning inference relationship, the corresponding user cognitive low-level adaptive learning node is constructed into an adaptive cognitive learning network to generate a user adaptive cognitive learning network initial architecture.

[0043] Further, the cognitive learning interaction bottleneck simulation analysis module comprises the following functions:

[0044] Obtain user real-time cognitive learning behavior data;

[0045] The user real-time cognitive learning behavior data is processed to extract learning behavior features to obtain user cognitive learning behavior features; based on the user cognitive learning behavior features, the user real-time cognitive learning behavior data is analyzed to recognize the cognitive learning interaction mode to obtain the user cognitive learning interaction behavior mode;

[0046] Based on the user cognitive learning interaction behavior mode, the user cognitive learning behavior features are divided and fused to obtain a learning feature fusion vector corresponding to each user cognitive learning interaction behavior;

[0047] By inputting the learning feature fusion vector corresponding to each user cognitive learning interaction behavior into the user adaptive cognitive learning network initial architecture, the cognitive learning interaction simulation analysis is performed to obtain a cognitive learning interaction simulation process corresponding to each user cognitive learning interaction behavior;

[0048] The cognitive learning interaction simulation process corresponding to the learning interaction behavior of each user is subjected to bottleneck identification analysis, and a bottleneck point of the cognitive learning interaction simulation process of the user is obtained.

[0049] Further, the cognitive learning network feedback optimization module comprises the following functions:

[0050] The cognitive learning interaction bottleneck feature of the cognitive learning interaction simulation process bottleneck point of the user is analyzed, and cognitive learning interaction process bottleneck feature description data of the user is obtained.

[0051] Based on the cognitive learning interaction process bottleneck feature description data of the user, learning network optimization target analysis is performed on the initial architecture of the user adaptive cognitive learning network, and a cognitive learning network optimization target list of the user is obtained.

[0052] The learning network optimization strategy analysis is performed on the cognitive learning network optimization target list of the user, and a cognitive learning network optimization adjustment strategy scheme of the user is obtained.

[0053] According to the cognitive learning network optimization adjustment strategy scheme of the user, the cognitive learning network feedback optimization adjustment is performed on the initial architecture of the user adaptive cognitive learning network, so as to generate a user adaptive cognitive learning optimization adjustment network.

[0054] The beneficial effects of the present application are as follows:

[0055] The adaptive cognitive learning network simulation system based on a large language model comprises a user cognitive learning feature vector conversion module, a hierarchical cognitive reasoning and learning network construction module, a cognitive learning interaction bottleneck simulation analysis module, and a cognitive learning network feedback optimization module. Compared with the prior art, the application has the beneficial effects that historical cognitive learning data records are obtained from the user cognitive learning platform database to provide basic data support for subsequent analysis. Historical data is the key to understanding user cognitive learning behavior, preferences, and effects. By collecting cognitive learning records of users at different time periods, including cognitive learning content, duration, frequency, and performance information, the cognitive learning process and state of users can be fully depicted, which not only helps to design personalized learning paths, but also provides quantitative data basis for subsequent cognitive interaction processes. By analyzing and recording the user historical cognitive learning data, the interactive mode of users in the cognitive learning process can be deeply mined. This analysis not only focuses on the relationship between users and learning content, but also emphasizes the interaction between users and the platform and other learners. By recording the interactive behavior of users in the learning process, detailed cognitive interaction record data can be constructed, which provides an important perspective for understanding the learning habits and social behavior of users. For example, users repeatedly review, ask questions, or participate in discussions on specific content, indicating their interest or confusion level about the content. By constructing a corresponding quality evaluation system based on cognitive interaction record data, feedback analysis of user historical cognitive learning data is performed. This system is established to evaluate the interaction quality of users in the learning process, including the frequency, depth, and diversity of interaction. Through these indicators, the interaction of users with learning content and the community in the learning process can be better understood, and the impact of these interactions on their learning effectiveness can be evaluated, thereby providing basic data support for subsequent processing. Meanwhile, by fusing and converting the cognitive feature vectors according to the user historical cognitive learning data, user cognitive interaction record data, and user cognitive learning feedback information data, a comprehensive understanding of user cognitive features can be achieved. This fusion process integrates the learning history, interaction behavior, and feedback information of users, forming a multi-dimensional cognitive feature vector. These feature vectors can not only reflect the knowledge level, learning preferences, and interaction quality of users, but also provide strong data support for subsequent adaptive cognitive learning network simulation construction.Further, by analyzing the reasoning of high-level learning tasks for user cognitive knowledge, it can be identified which learning tasks have a greater impact on the overall learning process of the user by analyzing the reasoning path between learning tasks. This analysis enables the user to better understand the dependency between learning tasks and provides a basis for corresponding cognitive learning strategy adjustment. More importantly, the user can optimize their cognitive learning plan according to the high-level reasoning relationship to ensure that those tasks that are crucial to subsequent learning are completed first in the learning process. Further, by constructing an adaptive cognitive learning network based on user cognitive knowledge low-level learning tasks and user cognitive knowledge high-level learning reasoning relationships, a corresponding adaptive cognitive learning network is constructed. This network is personalized and optimized for the specific cognitive learning characteristics and needs of the user, and it can dynamically adapt to the user's learning progress and changes. Through the establishment of an adaptive learning network, it is possible to adjust the learning content and strategy in real time, thereby providing personalized learning experiences. Through this network, users can obtain suitable learning resources and recommendations at different learning stages, and can effectively handle diverse knowledge structures and learning tasks, further improving the flexibility and effectiveness of the cognitive learning process. This networked learning structure also encourages users to explore and practice more deeply in the cognitive learning process, enabling them to improve their cognitive development comprehensively, thereby providing sustainable data support for long-term cognitive development of users. Then, by obtaining real-time cognitive learning behavior data of the user, this step collects the user's learning behavior in real time through various channels (such as learning platforms, mobile applications, sensors, etc.), including learning time, participation, problem solving, and interaction frequency. These data provide real and dynamic information for subsequent analysis, which helps to understand the user's learning habits and preferences. Further, by simulating cognitive learning interaction bottlenecks based on the initial architecture of the user's adaptive cognitive learning network, it can help the system identify the bottlenecks and challenges in the user's learning process. The results of this step not only reveal which links cause poor cognitive learning results, but also provide data support for educators to develop effective intervention strategies. By identifying and solving these bottlenecks, the cognitive learning process of the user can be more effectively supported to ensure timely assistance when encountering difficulties. The ultimate goal of this analysis is to optimize the learning experience and improve learning efficiency, enabling users to achieve better results in the cognitive learning process. Finally, by adjusting the cognitive learning network feedback optimization based on the initial architecture of the user's adaptive cognitive learning network, it can help the system better understand the bottlenecks and obstacles in the user's cognitive learning process to develop specific optimization strategy solutions to effectively solve the previously identified learning bottlenecks.By dynamically optimizing the learning network architecture and adjusting the feedback based on previous analysis results, a more effective cognitive learning environment is formed. Through the implementation of these optimizations, the cognitive learning network can respond to the learning needs of users in real time, providing personalized learning resources and support. This feedback optimization not only improves the user's learning experience, but also promotes the improvement of learning effectiveness, making the learning process more efficient and fruitful. This provides users with a smoother and more effective learning experience, achieving higher levels of cognitive learning achievement. BRIEF DESCRIPTION OF DRAWINGS

[0056] Other features, objects, and advantages of the present application will become more apparent from the following detailed description of non-limiting embodiments thereof, read in conjunction with the accompanying drawings:

[0057] Figure 1 Module schematic diagram of the adaptive cognitive learning network simulation system based on large language model of the present application;

[0058] Figure 2 For Figure 1 Functional flowchart of the user cognitive learning feature vector conversion module in the present application;

[0059] Figure 3 For Figure 1 Functional flowchart of the hierarchical cognitive reasoning and learning network construction module in the present application. DETAILED DESCRIPTION

[0060] The technical system of the present application will be described in detail below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0061] In addition, the accompanying drawings are only schematic illustrations of the present application, and are not necessarily drawn to scale. The same reference numerals in the drawings represent the same or similar parts, and thus repeated descriptions thereof will be omitted. Some of the block diagrams shown in the drawings are functional entities, which do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different network and / or processor systems and / or microcontroller systems.

[0062] It should be understood that, although the terms "first", "second" and the like can be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element without departing from the scope of the example embodiments. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0063] To achieve the above object, please refer to Figures 1 to 3 The application provides a large language model-based adaptive cognitive learning network simulation system, which comprises the following modules:

[0064] A user cognitive learning feature vector conversion module is configured to acquire user historical cognitive learning data, user cognitive interaction record data and user cognitive learning feedback information data, and perform cognitive feature vector fusion conversion according to the user historical cognitive learning data, the user cognitive interaction record data and the user cognitive learning feedback information data to obtain a user cognitive learning feature vector.

[0065] A hierarchical cognitive reasoning and learning network construction module is configured to perform hierarchical cognitive reasoning analysis on the user cognitive learning feature vector based on a large language model to obtain a user cognitive knowledge low-level learning task and a user cognitive knowledge high-level learning reasoning relationship, and perform adaptive cognitive learning network construction according to the user cognitive knowledge low-level learning task and the user cognitive knowledge high-level learning reasoning relationship to generate an initial architecture of a user adaptive cognitive learning network.

[0066] A cognitive learning interaction bottleneck simulation analysis module is configured to acquire user real-time cognitive learning behavior data, and perform cognitive learning interaction bottleneck simulation analysis on the initial architecture of the user adaptive cognitive learning network based on the user real-time cognitive learning behavior data to obtain a user cognitive learning interaction simulation process bottleneck point.

[0067] A cognitive learning network feedback optimization module is configured to perform cognitive learning network feedback optimization adjustment on the initial architecture of the user adaptive cognitive learning network based on the user cognitive learning interaction simulation process bottleneck point to generate an optimized and adjusted user adaptive cognitive learning network.

[0068] In the embodiments of the application, please refer to Figure 1 The figure shows a large language model-based adaptive cognitive learning network simulation system, which comprises the following modules in the example:

[0069] S1: a user cognitive learning feature vector conversion module, configured to obtain user historical cognitive learning data, user cognitive interaction record data, and user cognitive learning feedback information data, and perform cognitive feature vector fusion conversion according to the user historical cognitive learning data, the user cognitive interaction record data, and the user cognitive learning feedback information data to obtain a user cognitive learning feature vector;

[0070] In the embodiment of the present application, by connecting the database of the user cognitive learning platform, the SQL query statement is used to extract the historical cognitive learning record information related to the user from the learning record table, including the fields of user ID, learning topic, learning time, learning duration, and completion status, to obtain the user historical cognitive learning data. By identifying and analyzing the learning trajectory of the obtained user historical cognitive learning data, the time series analysis method is used to sort the learning activities according to the timestamp, create a learning trajectory graph, record the learning topic and the corresponding learning duration of each user in a specific time period, combine the graphical visualization tool Matplotlib, display the user learning trajectory in the form of a line chart, and analyze the cognitive learning interaction of the corresponding user historical cognitive learning data by combining the previously obtained cognitive learning activity trajectory, to model the interaction between the user and the content in the learning activity by using the graph analysis tool, record the interaction behavior in each learning activity, such as question answering, note recording, discussion participation, etc., to form cognitive interaction record data. At the same time, by establishing a cognitive interaction quality evaluation system according to the previously analyzed user cognitive interaction record, the corresponding evaluation indicators are defined, such as interaction frequency, interaction depth, and feedback effectiveness, and based on the previously established user cognitive interaction quality evaluation system, the weight distribution method is used to set the corresponding weight for each evaluation indicator, to comprehensively evaluate the corresponding learning content in the user historical cognitive learning data, calculate the interaction quality score, and generate the user's cognitive learning feedback evaluation data, to obtain the user cognitive learning feedback information data. Then, by combining the previously analyzed user historical cognitive learning data, user cognitive interaction record data, and user cognitive learning feedback information data, the cognitive feature vector fusion conversion is performed, to convert various data into feature data by using the feature engineering method, such as obtaining the features including learning content topic, learning difficulty level, and learning duration from the user's historical cognitive learning data, obtaining the features including click rate, viewing duration, and discussion participation from the user's cognitive interaction record data, and obtaining the features including positive, neutral, and negative feedback sentiment index from the user cognitive learning feedback information data. The feature superposition method is used to fuse the features to generate the final feature vector, such as [A, B, C], where A is the user cognitive learning content feature set, B is the user cognitive learning interaction frequency feature set, and C is the user cognitive learning feedback sentiment feature set, to finally obtain the user cognitive learning feature vector.

[0071] S2: a hierarchical cognitive reasoning and learning network construction module, configured to perform hierarchical cognitive reasoning analysis on the user cognitive learning feature vector based on the large language model, to obtain a user cognitive knowledge low-level learning task and a user cognitive knowledge high-level learning reasoning relationship; and perform self-adaptive cognitive learning network construction according to the user cognitive knowledge low-level learning task and the user cognitive knowledge high-level learning reasoning relationship, to generate an initial architecture of a user self-adaptive cognitive learning network;

[0072] In the embodiment of the present application, the user cognitive learning feature vector obtained through previous fusion and conversion is analyzed for learning tasks by using the low-level cognitive module in the large language model, wherein the input feature vector includes learning content features, learning interaction frequency features, and learning feedback emotion features, and the like, to analyze the input feature vectors through a specific machine learning algorithm (such as a decision tree or a support vector machine) in the low-level cognitive module, and identify corresponding learning tasks from them, including information memorization, concept understanding, and the like, to obtain the user cognitive knowledge low-level learning task. The previously identified user cognitive knowledge low-level learning task is classified by task content, to classify the low-level learning tasks according to themes, difficulties, and correlations by constructing a task content classification matrix, and to group the learning tasks by using a clustering algorithm (such as K-means or hierarchical clustering), which can be classified into memory tasks, understanding tasks, and application tasks, and the like, and a high-level cognitive reasoning framework capable of processing low-level learning tasks is constructed by using the high-level cognitive module in the large language model, to define the basic structure of high-level cognitive reasoning, including reasoning rules, logical relationships, and reasoning paths, and to analyze the high-level reasoning path of the user's low-level learning task content by using a reasoning engine (such as Prolog or a self-defined rule engine), the model derives the relationship between each low-level learning task through the established reasoning rules, to generate a high-level cognitive reasoning path. At the same time, the high-level cognitive reasoning path obtained through previous reasoning analysis is used to adjust the learning reasoning relationship of the corresponding user cognitive knowledge low-level learning task, to delete or reconstruct the low relationship strength path in the high-level reasoning path by using the adjustment method in graph theory, to optimize the learning reasoning relationship between learning tasks, to generate an optimized user cognitive knowledge high-level learning reasoning relationship. Then, the self-adaptive cognitive learning network is constructed according to the previously analyzed user cognitive knowledge low-level learning task and high-level learning reasoning relationship, the construction process includes defining the nodes and edges of the network, wherein the nodes represent learning tasks and the edges represent the reasoning relationship between tasks, and the network is optimized by using a graph neural network (GNN) or a reinforcement learning algorithm, so that the network can be self-adaptively adjusted according to the user's feedback, to finally generate an initial architecture of a user self-adaptive cognitive learning network.

[0073] S3: a cognitive learning interaction bottleneck simulation analysis module, configured to acquire real-time cognitive learning behavior data of a user, and perform cognitive learning interaction bottleneck simulation analysis on an initial adaptive cognitive learning network architecture of the user based on the real-time cognitive learning behavior data of the user, to obtain a bottleneck point in a simulation process of cognitive learning interaction of the user;

[0074] In the embodiments of the present application, the cognitive learning activities of the user are monitored in real time by using sensors, learning management systems (LMS) and data collection tools, including learning progress trackers, online assessment platforms and interactive learning software, and through these systems, the collected data includes the user's learning time, response speed, number of incorrect answers, frequency of participating in discussions and use of learning resources, etc., thereby integrating the user's real-time cognitive learning behavior data. Through data cleaning and preprocessing of the previously collected real-time cognitive learning behavior data of the user, noise and outliers are removed, and through the use of feature extraction algorithms such as principal component analysis (PCA) and clustering algorithms, the user's learning behavior is analyzed in depth, and key features are extracted, including the user's learning efficiency (such as the ratio of time required to complete a task to accuracy), learning mode (such as the ratio of active learning to passive learning) and emotional state (such as anxiety level during learning), etc. At the same time, by applying machine learning models (such as random forests or support vector machines) to identify and analyze the cognitive learning interaction patterns of the corresponding user cognitive learning behavior process within the user's real-time cognitive learning behavior data, the user's learning behavior interaction patterns are identified through the trained model, such as self-regulated learning, social learning or cooperative learning patterns, and by combining the previously identified and analyzed user cognitive learning interaction behavior patterns, the corresponding learning behavior features are classified into the same interaction behavior pattern, and by using clustering analysis, the behavior features with similar interaction behavior patterns are grouped, and a corresponding feature fusion vector is generated for each group of behavior features. These vectors will contain the common features of each group of user cognitive learning behavior, such as typical performance and interaction methods in a specific learning task, thereby obtaining the learning feature fusion vectors corresponding to each user cognitive learning interaction behavior. At the same time, by inputting the previously fused learning feature fusion vectors corresponding to each user cognitive learning interaction behavior into the previously constructed initial architecture of the user adaptive cognitive learning network for simulation analysis of the cognitive learning interaction process, the behavior performance and learning effect of each user in a specific learning environment are predicted, and the network is trained using the backpropagation algorithm, and the network weights are optimized through multiple iterations, thereby simulating an accurate cognitive learning interaction process. The simulation results show the expected path and learning achievements of the user in the learning process, and then, after completing the simulation of the cognitive learning interaction process corresponding to each user cognitive learning interaction behavior, bottleneck identification analysis of the interaction simulation process is performed to identify the bottleneck points in the cognitive learning interaction simulation process by applying data mining techniques such as outlier detection and time series analysis, including understanding barriers for specific knowledge points, frequent incorrect responses or significant lag in learning progress, etc., and finally obtaining the corresponding user cognitive learning interaction simulation process bottlenecks.

[0075] S4: a cognitive learning network feedback optimization module, configured to perform cognitive learning network feedback optimization adjustment on the initial architecture of the user-adaptive cognitive learning network based on the user cognitive learning interaction simulation process bottleneck point, to generate a user-adaptive cognitive learning optimization adjustment network.

[0076] In the embodiment of the present application, by performing statistical analysis on the cognitive learning bottleneck characteristics of the previously analyzed user cognitive learning interaction simulation process bottleneck point (such as understanding obstacles of specific knowledge points, frequent error reactions, or obvious lag in learning progress, etc.), the bottleneck performance of the user in a specific cognitive learning situation is statistically analyzed to reveal characteristic description factors that affect cognitive learning efficiency, such as user distraction, weak knowledge mastery, or low interaction participation, etc., and to help the system better understand the obstacles that occur in the cognitive learning process of the user, and by combining the previously analyzed user cognitive learning interaction process bottleneck characteristic description, the learning network optimization goal of the initial architecture of the user-adaptive cognitive learning network used in the cognitive learning interaction process is identified and analyzed, so as to map the bottleneck characteristic data into the structure of the learning network by using system thinking theory, and to clearly define the optimization goals, including reducing the cognitive load of the user, improving the depth of knowledge mastery, and improving the learning initiative, wherein each optimization goal corresponds to a specific quantitative indicator, such as a 20% reduction in error rate, a 15% improvement in learning initiative, etc., and by constructing an optimization goal list, and after determining the optimization goal list of the user cognitive learning network, learning network optimization strategy analysis is performed, to evaluate the performance of the current learning network by using model prediction technology, and to compare it with the optimization goal, then, a variety of optimization strategies are designed by using reinforcement learning algorithm, for example, for users with excessively high cognitive load, an optimization strategy based on hierarchical teaching method is implemented, which provides step-by-step guidance and timely feedback to relieve learning pressure, and simulation tools are used to evaluate the strategy to ensure its effectiveness and operability, thereby forming a comprehensive optimization adjustment strategy scheme, then, the initial architecture of the corresponding user-adaptive cognitive learning network is adjusted based on the previously analyzed learning network optimization adjustment strategy scheme, to analyze the new data of the user in the learning process in real time by using machine learning algorithm, and to respond to the learning needs of the user by adjusting the network parameters and learning path, for example, if a user shows extremely high frustration in a specific link, the learning content will be dynamically adjusted, the corresponding basic knowledge review materials will be provided, and interactive exercises will be increased to enhance the cognitive learning confidence, thereby optimizing and adjusting the corresponding adaptive cognitive learning optimization network, which will effectively improve the learning efficiency and effect of the user, achieve the goal of personalized cognitive learning, and finally optimize the generation of the user-adaptive cognitive learning optimization adjustment network.

[0077] Further, the user cognitive learning feature vector analysis module includes the following functions:

[0078] Obtaining historical cognitive learning data records from the user cognitive learning platform database to obtain user historical cognitive learning data;

[0079] Conduct cognitive learning trajectory analysis on the user's historical cognitive learning data to obtain the user's historical cognitive learning activity trajectory;

[0080] Based on the user's historical cognitive learning activity trajectory, the user's historical cognitive learning data is analyzed and recorded to obtain user cognitive interaction record data;

[0081] Build a user cognitive interaction quality evaluation system based on user cognitive interaction record data, and conduct cognitive learning feedback analysis on user historical cognitive learning data based on the user cognitive interaction quality evaluation system to obtain user cognitive learning feedback information data;

[0082] The cognitive feature vector is fused and transformed according to the user's historical cognitive learning data, the user's cognitive interaction record data and the user's cognitive learning feedback information data to obtain the user's cognitive learning feature vector.

[0083] As an embodiment of the present invention, refer to Figure 2 As shown, Figure 1 Schematic diagram of the functional flow of the user cognitive learning feature vector analysis module in the embodiment. In this embodiment, the functions of the user cognitive learning feature vector analysis module include:

[0084] S11: Obtaining historical cognitive learning data records from a user cognitive learning platform database to obtain user historical cognitive learning data;

[0085] In an embodiment of the present invention, by connecting to the database of the user cognitive learning platform, SQL query statements are used to extract historical cognitive learning record information related to the user from the learning record table, including fields such as user ID, learning topic, learning time, learning duration and completion status. After the data is extracted, it is formatted to ensure that the time format of each record is consistent, avoid data redundancy, and finally obtain the user's historical cognitive learning data.

[0086] S12: Analyze the user's historical cognitive learning data to obtain the user's historical cognitive learning activity trajectory;

[0087] In the embodiment of the present application, the learning activity is sorted by timestamp by using the time series analysis method through the identification and analysis of the learning trajectory of the obtained user historical cognitive learning data, and the learning trajectory graph is created by using the Pandas library in Python, the learning theme and the corresponding learning duration of each user in a specific time period are recorded, the user learning trajectory is displayed in the form of line chart by combining the graphical visualization tool Matplotlib, the learning behavior pattern is more intuitive, and finally the user historical cognitive learning activity trajectory is obtained.

[0088] S13: cognitive interaction analysis and recording of user historical cognitive learning data based on user historical cognitive learning activity trajectory, to obtain user cognitive interaction record data;

[0089] In the embodiment of the present application, the cognitive learning interaction analysis of the corresponding user historical cognitive learning data is performed by combining the user historical cognitive learning activity trajectory obtained by the previous analysis, the interaction between the user and the content in the learning activity is modeled by using the graph analysis tool, the interaction behavior in each learning activity is recorded, such as question answering, note recording, discussion participation, etc., and these behaviors are quantified, thereby forming cognitive interaction record data and stored in the database in a structured form, and finally the user cognitive interaction record data is obtained.

[0090] S14: constructing a user cognitive interaction quality evaluation system according to the user cognitive interaction record data, and performing cognitive learning feedback analysis on the user historical cognitive learning data based on the user cognitive interaction quality evaluation system, to obtain user cognitive learning feedback information data;

[0091] In the embodiment of the present application, the evaluation system of the corresponding cognitive interaction quality is established according to the user cognitive interaction record data obtained by the previous analysis, the corresponding evaluation indexes are defined, such as interaction frequency, interaction depth and feedback effectiveness, and the weight distribution method is used based on the previously established user cognitive interaction quality evaluation system, the corresponding weight is set for each evaluation index, the corresponding learning content in the user historical cognitive learning data is comprehensively evaluated, the interaction quality score is calculated, the cognitive learning feedback evaluation score of the user is generated, and finally the user cognitive learning feedback information data is obtained.

[0092] S15: cognitive feature vector fusion conversion according to user historical cognitive learning data, user cognitive interaction record data and user cognitive learning feedback information data, to obtain user cognitive learning feature vector.

[0093] In the embodiment of the present application, the cognitive feature vector is fused and converted by combining the previously analyzed user historical cognitive learning data, user cognitive interaction record data and user cognitive learning feedback information data, so as to convert various data into feature data by using a feature engineering method, for example, obtaining features such as learning content theme, learning difficulty level and learning time length from the user's historical cognitive learning data, obtaining features such as click rate, viewing time length and discussion participation from the user's cognitive interaction record data, and obtaining features such as positive, neutral and negative feedback sentiment index from the user cognitive learning feedback information data, and generating a final feature vector by using a feature superposition method, for example, A, B, C], wherein A is a user cognitive learning content feature set, B is a user cognitive learning interaction frequency feature set, and C is a user cognitive learning feedback sentiment feature set, and finally obtaining a user cognitive learning feature vector.

[0094] Further, the cognitive learning trajectory analysis of the user historical cognitive learning data comprises:

[0095] The cognitive learning time and content extraction processing of the user historical cognitive learning data obtains a user historical cognitive learning timestamp and user historical cognitive learning content.

[0096] In the embodiment of the present application, the user historical cognitive learning data is extracted from the user's historical cognitive learning database, and the user historical cognitive learning data is filtered by using SQL query to obtain the timestamp and content of each learning activity. The timestamp will be converted into a standard format for time series analysis. The extracted learning content will be stored in a structured data table, which contains the unique identifier, learning theme, timestamp and content description of each learning activity. Then, the extracted data is cleaned and formatted by using the Pandas library of Python to ensure the clear correspondence between the timestamp and the content. Finally, the user historical cognitive learning timestamp and user historical cognitive learning content are obtained.

[0097] Preferably, the cognitive learning theme mining analysis of the user historical cognitive learning content obtains various user historical cognitive learning activity themes.

[0098] In the embodiment of the present application, the cognitive learning theme of the extracted user historical cognitive learning content is analyzed, the natural language processing (NLP) technology is used, the LDA (latent Dirichlet allocation) model is applied for theme modeling, the learning content is processed by using a word segmentation tool (such as jieba), stop words are removed, a bag-of-words model is formed, the bag-of-words model is input into the LDA model for training, the main learning theme of the user is identified, the learning activity theme list of each user is generated, and finally the historical cognitive learning activity theme of each user is obtained.

[0099] Preferably, the relevance analysis is performed on the historical cognitive learning activity theme of each user to obtain the learning activity behavior correlation relationship between the cognitive learning activity themes of each user.

[0100] In the embodiment of the present application, the relevance analysis is performed on the historical cognitive learning activity theme of each user to obtain the learning activity behavior correlation relationship between the cognitive learning activity themes of each user.

[0101] Preferably, the cognitive learning behavior correlation graph analysis is performed on the historical cognitive learning content of the user based on the learning activity behavior correlation relationship between the cognitive learning activity themes of each user to generate the cognitive learning behavior correlation graph of the user.

[0102] In the embodiment of the present application, the learning behavior correlation graph of the corresponding learning content theme in the historical cognitive learning content of the user is connected and constructed by combining the learning activity behavior correlation relationship between the cognitive learning activity themes of each user obtained by the previous correlation analysis, the graph structure in graph theory is used, each learning theme in the historical cognitive learning content of the user is taken as a node, the correlation relationship between the nodes is taken as an edge, the network analysis tool (such as NetworkX) is used for graph construction, the learning preference and behavior mode of the user are understood by analyzing the node degree, clustering coefficient and other indexes of the graph, and the corresponding cognitive learning behavior correlation graph is generated to intuitively show the correlation between different learning themes, and finally the cognitive learning behavior correlation graph of the user is generated.

[0103] Preferably, the cognitive learning trajectory analysis is performed on the cognitive learning behavior correlation graph of the user based on the historical cognitive learning timestamp of the user to obtain the historical cognitive learning activity trajectory of the user.

[0104] In the embodiment of the present application, by combining the previously extracted user historical cognitive learning time stamp, the learning process track of the learning content theme node in the corresponding cognitive learning behavior association relationship path in the user historical cognitive learning behavior association graph is identified and analyzed, so as to arrange the extracted time stamp in time sequence and combine the association graph, identify the cognitive learning activity track of the user in the corresponding time sequence, and identify and analyze the specific learning path of the user. By calculating the time distribution of the user on each learning theme, the learning track graph of the user is constructed, the learning activity and its change of the user on different themes are presented in a visual manner, and then the learning habit and potential problems are revealed, and finally the user historical cognitive learning activity track is analyzed.

[0105] Further, the cognitive feature vector fusion conversion according to the user historical cognitive learning data, the user cognitive interaction record data and the user cognitive learning feedback information data comprises:

[0106] The cognitive learning content features of the user are analyzed to obtain user cognitive learning content features, wherein the user cognitive learning content features include user cognitive learning content themes, user cognitive learning difficulty levels and user cognitive learning time lengths;

[0107] In the embodiment of the present application, by collecting the historical cognitive learning data of the user, including learning records, theme contents and completion time, etc., and by using a data analysis tool (such as the Pandas library of Python) to arrange and clean the user historical cognitive learning data, and by using a natural language processing technology to extract learning content themes, using a TF-IDF algorithm to calculate the weight of the theme to determine the main content concerned by the user, at the same time, using a statistical analysis method (such as the average value and the standard deviation) to evaluate the difficulty level of the learning content, and calculating the time length of each learning, so as to analyze and generate the cognitive learning content features of the user, covering the learning content theme, the learning difficulty level and the learning time length, and finally obtaining the user cognitive learning content features.

[0108] Preferably, the user interaction frequency features of the user cognitive interaction record data are analyzed to obtain user cognitive learning interaction frequency features, wherein the user cognitive learning interaction frequency features include user cognitive learning click rates, user cognitive learning viewing time lengths and user cognitive learning discussion participation degrees;

[0109] In the embodiment of the present application, the cognitive interaction record data of the user is collected, including click logs, viewing time and discussion records, and the click rate of the user is calculated through data mining technology (such as time series analysis), so as to obtain the participation of the user in different learning contents, the viewing time is used to identify the concentration of the user by comparing the actual viewing time of each learning module of the user with the preset time, and the discussion participation is evaluated by analyzing the speaking frequency and reply of the user in the discussion area, so as to analyze the corresponding learning interaction frequency characteristics, including the click rate, the viewing time and the discussion participation, and finally obtain the cognitive learning interaction frequency characteristics of the user.

[0110] Preferably, the cognitive learning feedback information data of the user is analyzed to obtain cognitive learning feedback emotional characteristics of the user, wherein the cognitive learning feedback emotional characteristics of the user include positive feedback emotional index of the user, neutral feedback emotional index of the user and negative feedback emotional index of the user.

[0111] In the embodiment of the present application, the cognitive learning feedback information data of the user obtained by the previous feedback analysis is statistically analyzed to obtain the learning feedback emotional characteristics of the user, and the emotional mathematical characteristic parameters are analyzed, so as to use the emotional analysis technology (such as VADER or TextBlob) to score the emotional feedback information of the user, classify the feedback content into positive, neutral or negative, and calculate the emotional index of each feedback type to obtain the emotional tendency of the user in the learning process, the positive feedback emotional index is based on the number of positive evaluations divided by the total number of evaluations, the neutral feedback emotional index is based on the number of comments without obvious tendency divided by the total number of evaluations, and the negative feedback emotional index is based on the number of negative evaluations divided by the total number of evaluations, so as to statistically analyze and generate the corresponding learning feedback emotional characteristics, and finally obtain the cognitive learning feedback emotional characteristics of the user.

[0112] Preferably, the cognitive learning content characteristics, the cognitive learning interaction frequency characteristics and the cognitive learning feedback emotional characteristics of the user are fused and converted into a cognitive feature vector.

[0113] In the embodiment of the present application, the cognitive learning content characteristics, the cognitive learning interaction frequency characteristics and the cognitive learning feedback emotional characteristics of the user obtained by the previous analysis are fused, so as to normalize different characteristics by using a feature standardization method, ensure that each feature is in the same order of magnitude, and fuse each feature to generate a final feature vector in a feature superposition manner, for example [A, B, C], wherein A is a set of cognitive learning content characteristics of the user, B is a set of cognitive learning interaction frequency characteristics of the user, and C is a set of cognitive learning feedback emotional characteristics of the user, the vector can be used for subsequent model training and learning effect prediction, fully reflects the cognitive state and learning habit of the user, and finally obtains the cognitive learning feature vector of the user.

[0114] Further, the hierarchical cognitive reasoning and learning network construction module includes the following functions:

[0115] Based on the low-level cognitive module in the large language model, the user cognitive learning feature vector is analyzed for low-level task identification to obtain a user cognitive knowledge low-level learning task.

[0116] The user cognitive knowledge low-level learning task is processed for task content classification to obtain a user cognitive low-level learning task content classification matrix.

[0117] A high-level cognitive reasoning framework capable of processing low-level learning tasks is constructed through a high-level cognitive module in the large language model, and based on the high-level cognitive reasoning framework, a high-level cognitive reasoning path analysis is performed on the corresponding low-level learning tasks in the user cognitive low-level learning task content classification matrix, to obtain high-level cognitive reasoning paths between the low-level learning tasks.

[0118] Based on the high-level cognitive reasoning paths between the low-level learning tasks, a high-level learning reasoning adjustment analysis is performed on the corresponding user cognitive knowledge low-level learning tasks to obtain a user cognitive knowledge high-level learning reasoning relationship.

[0119] An adaptive cognitive learning network is constructed according to the user cognitive knowledge low-level learning task and the user cognitive knowledge high-level learning reasoning relationship to generate an initial architecture of the user adaptive cognitive learning network.

[0120] As an embodiment of the present application, referring to Figure 3 , it is Figure 1 the function flow diagram of the middle-level cognitive reasoning and learning network construction module, in this embodiment, the functions of the hierarchical cognitive reasoning and learning network construction module include:

[0121] S21: Based on the low-level cognitive module in the large language model, the user cognitive learning feature vector is analyzed for low-level task identification to obtain a user cognitive knowledge low-level learning task.

[0122] In the embodiment of the present application, the user cognitive learning feature vector obtained by previous fusion and conversion is analyzed for learning task identification by using the low-level cognitive module in the large language model, wherein the input feature vector includes learning content features, learning interaction frequency features, and learning feedback emotion features, etc. information, so as to analyze these input feature vectors by a specific machine learning algorithm (such as decision tree or support vector machine) in the low-level cognitive module, and identify the corresponding learning tasks from them, which include information memory, concept understanding, etc., thereby forming a set of user cognitive knowledge low-level learning tasks, and finally obtaining a user cognitive knowledge low-level learning task.

[0123] S22: performing task content classification processing on the user cognitive knowledge low-level learning task to obtain a user cognitive knowledge low-level learning task content classification matrix;

[0124] In the embodiment of the present application, by classifying the task content of the previously identified user cognitive knowledge low-level learning task, the low-level learning tasks are classified according to the theme, difficulty and relevance by constructing a task content classification matrix, and the learning tasks are grouped by using a clustering algorithm (such as K-means or hierarchical clustering), and the identification of different categories is generated, for example, the learning tasks can be divided into memory tasks, understanding tasks and application tasks, etc., thereby forming a clear classification matrix, identifying the characteristics of each task, and finally obtaining a user cognitive knowledge low-level learning task content classification matrix.

[0125] S23: constructing a high-level cognitive reasoning framework capable of processing low-level learning tasks through a high-level cognitive module in a large language model, and performing high-level cognitive reasoning path analysis on the corresponding low-level learning tasks in the user cognitive knowledge low-level learning task content classification matrix based on the high-level cognitive reasoning framework, to obtain the high-level cognitive reasoning path between the low-level learning tasks;

[0126] In the embodiment of the present application, a high-level cognitive reasoning framework capable of processing low-level learning tasks is constructed by using a high-level cognitive module in a large language model, to define the basic structure of high-level cognitive reasoning, including reasoning rules, logical relationships and reasoning paths, and to perform high-level reasoning path analysis on the user's low-level learning task content classification matrix by using a reasoning engine (such as Prolog or a self-defined rule engine), the model deduces the relationship between the low-level learning tasks through the established reasoning rules, thereby generating a high-level cognitive reasoning path, so as to better understand the learning process and needs of the user, and finally obtaining the high-level cognitive reasoning path between the low-level learning tasks.

[0127] S24: performing high-level learning reasoning adjustment analysis on the corresponding user cognitive knowledge low-level learning task based on the high-level cognitive reasoning path between the low-level learning tasks, to obtain the user cognitive knowledge high-level learning reasoning relationship;

[0128] In the embodiment of the present application, the learning reasoning relationship of the corresponding user cognitive knowledge low-level learning task is adjusted by combining the high-level cognitive reasoning path between the low-level learning tasks obtained through the previous reasoning analysis, to delete or reconstruct the low relationship strength path in the high-level reasoning path by using the adjustment method in graph theory, to optimize the learning reasoning relationship between the learning tasks, and the reasoning path can also be analyzed by using matrix operation, to ensure that the logical relationship between the tasks is clear and efficient, thereby generating an optimized high-level learning reasoning relationship, and finally obtaining the user cognitive knowledge high-level learning reasoning relationship.

[0129] S25: Constructing a self-adaptive cognitive learning network according to the low-level learning tasks of the user cognitive knowledge and the high-level learning inference relationship of the user cognitive knowledge, to generate an initial architecture of the user self-adaptive cognitive learning network.

[0130] In the embodiment of the present application, the construction of the self-adaptive cognitive learning network is performed according to the low-level learning tasks and the high-level learning inference relationship of the user cognitive knowledge obtained through previous analysis, which includes defining the nodes and edges of the network, wherein the nodes represent the learning tasks and the edges represent the inference relationship between the tasks, and the network is optimized by using the graph neural network (GNN) or reinforcement learning algorithm, so that the network can be adaptively adjusted according to the feedback of the user, thereby connecting to form an initial architecture of the user self-adaptive cognitive learning network, which can dynamically respond to the learning situation of the user and improve the learning efficiency, and finally generate an initial architecture of the user self-adaptive cognitive learning network.

[0131] Further, the high-level learning inference adjustment analysis of the corresponding user cognitive knowledge low-level learning task based on the high-level cognitive inference path between each low-level learning task includes:

[0132] The cognitive inference path relationship connection processing of the corresponding user cognitive knowledge low-level learning task based on the high-level cognitive inference path between each low-level learning task is performed to generate a high-level cognitive inference path relationship graph between each low-level learning task.

[0133] In the embodiment of the present application, the inference path relationship connection processing of the corresponding user cognitive knowledge low-level learning task based on the high-level cognitive inference path between each low-level learning task is performed to use the node and edge construction method in graph theory, regard each low-level learning task as a node, and the inference path relationship between the tasks as an edge, traverse each node through an algorithm (such as depth-first search or breadth-first search), identify the inference relationship between them, and use a data visualization tool (such as Gephi) to generate a high-level cognitive inference path relationship graph, which visualizes the identified relationship, ensures that the inference relationship of all low-level learning tasks is presented in a structured form, and finally generates a high-level cognitive inference path relationship graph between each low-level learning task.

[0134] Preferably, the inference relationship strength of the high-level cognitive inference path relationship graph between each low-level learning task is evaluated and calculated by using a learning task inference relationship strength calculation formula, to obtain the cognitive inference relationship strength between each low-level learning task.

[0135] In the embodiment of the present application, by combining the time variable parameter, the reasoning relationship influence weight between each low-level learning task, the reasoning path related number, the reasoning path number influence decay coefficient, the task content similarity, the task content similarity influence decay coefficient, the context correlation parameter, the context correlation influence decay coefficient and the related parameters, a suitable learning task reasoning relationship strength calculation formula is constructed to evaluate and calculate the high-level cognitive reasoning path relationship graph between each low-level learning task, so as to quantitatively calculate the reasoning relationship strength between each task, and finally obtain the cognitive reasoning relationship strength between each low-level learning task.

[0136] Preferably, based on the cognitive reasoning relationship strength between each low-level learning task, the high-level cognitive reasoning path relationship graph between each low-level learning task is analyzed and adjusted for high-level learning reasoning, so as to obtain the user cognitive knowledge high-level learning reasoning relationship.

[0137] In the embodiment of the present application, by combining the cognitive reasoning relationship strength between each low-level learning task obtained by the previous quantitative calculation, the corresponding high-level cognitive reasoning path relationship graph is analyzed and adjusted, so as to apply the strength-based weighted graph adjustment algorithm to delete or adjust the edges with low cognitive reasoning relationship strength, optimize the connection relationship between each low-level learning task, and ensure that the graph reflects a more accurate learning path. This process uses the clustering algorithm in machine learning to cluster tasks with similar strength, so as to overall evaluate the high-level reasoning relationship, thereby generating a new high-level learning reasoning relationship, effectively displaying the best learning reasoning path taken by the user in the learning process, and finally obtaining the user cognitive knowledge high-level learning reasoning relationship.

[0138] Further, the learning task reasoning relationship strength calculation formula is specifically:

[0139]

[0140] In the formula, R ij is the cognitive reasoning relationship strength between the i th low-level learning task and the j th low-level learning task, i and j are item degree measurement parameters of the low-level learning task, T is the upper limit of the time range of the reasoning relationship strength calculation, t is the time variable parameter, w ij is the reasoning relationship influence weight between the i th low-level learning task and the j th low-level learning task, s ij (t) is the reasoning path related number between the i th low-level learning task and the j th low-level learning task at time t, a is the reasoning path number influence decay coefficient, c ij (t) is the task content similarity between the i th low-level learning task and the j th low-level learning task at time t, β is the task content similarity influence decay coefficient, pij (t) is the context correlation parameter between the i-th low-level learning task and the j-th low-level learning task at time t, γ is the attenuation coefficient of the context correlation effect, and η is the correction coefficient of the cognitive reasoning relationship strength.

[0141] The present invention uses a specific mathematical model and has been verified to obtain a formula for calculating the strength of learning task reasoning relationships. This formula is used to evaluate the strength of reasoning relationships in the high-level cognitive reasoning path relationship map between low-level learning tasks. The learning task reasoning relationship strength calculation formula combines the number of reasoning paths, task content similarity, and contextual relevance, providing a comprehensive framework for evaluating cognitive reasoning relationships between low-level learning tasks. This multi-dimensional evaluation method. Through integral calculation (from 0 to T), the formula can take into account changes in time and reflect the strength of reasoning relationships between learning tasks in different time periods. This dynamic characteristic enables the analysis to adapt to changes in the learning process. In addition, by introducing reasoning relationship influence weights, the calculation formula can adjust the relative importance of different tasks according to actual conditions, thereby improving the flexibility and accuracy of the evaluation. By using the influence attenuation coefficient, the change of task relevance over time can be effectively controlled, so that the influence of the corresponding reasoning relationship on the current analysis is reduced, thereby ensuring that the results are more timely and relevant. By combining task content similarity with contextual relevance, the formula can more comprehensively evaluate the relationship between tasks, which is particularly important for complex learning tasks because they are often affected by multiple factors. In addition, by introducing a correction coefficient, the calculation results can be adjusted according to specific needs to adapt to specific application scenarios or compensate for potential deviations of the model, thereby improving the credibility of the results. In summary, this formula fully considers the strength of the cognitive reasoning relationship R between the i-th low-level learning task and the j-th low-level learning task. ij , the item measurement parameters i and j of the low-level learning task, the upper limit of the time range T for calculating the strength of the reasoning relationship, the time variable parameter t, the influence weight w of the reasoning relationship between the i-th low-level learning task and the j-th low-level learning task ij , the number of related reasoning paths s between the i-th low-level learning task and the j-th low-level learning task at time t ij (t), the attenuation coefficient α of the number of reasoning paths, the task content similarity c between the i-th low-level learning task and the j-th low-level learning task at time t ij (t), the attenuation coefficient of task content similarity β, the context correlation parameter p between the i-th low-level learning task and the j-th low-level learning task at time t ij (t), context relevance attenuation coefficient γ, correction coefficient η of cognitive reasoning relationship strength, according to the cognitive reasoning relationship strength R between the i-th low-level learning task and the j-th low-level learning taskij The inter-relationship between the above parameters constitutes a function relationship:

[0142]

[0143] The formula can realize the reasoning relationship strength evaluation and calculation process of the high-level cognitive reasoning path relationship graph between each low-level learning task, and through the introduction of the correction coefficient η of the cognitive reasoning relationship strength, the calculation process can be adjusted according to the error situation, thereby improving the accuracy and applicability of the learning task reasoning relationship strength calculation formula.

[0144] Further, the self-adaptive cognitive learning network construction according to the user cognitive knowledge low-level learning task and the user cognitive knowledge high-level learning reasoning relationship comprises:

[0145] The user cognitive knowledge low-level learning task is configured with a base node to obtain a user cognitive low-level learning task base node.

[0146] In the embodiment of the present application, the user cognitive knowledge low-level learning task obtained through previous identification analysis is configured with a base node to adopt a knowledge graph method to label the relationship between the knowledge points mastered by the user and the to-be-learned task content in the user cognitive knowledge low-level learning task, for example, if the user is learning mathematics, the base node includes "basic arithmetic", "geometric concept", etc. In the configuration process, the learning content of the user is mapped with the knowledge points by using a classification algorithm (such as a decision tree or a random forest) to ensure that the definition and range of each knowledge point are clear and explicit, thereby forming a systematic low-level learning task base node, and finally obtaining a user cognitive low-level learning task base node.

[0147] Preferably, the user cognitive low-level learning task base node is converted into a self-adaptive learning node to obtain a user cognitive low-level self-adaptive learning node.

[0148] In the embodiment of the present application, the learning style and preference of the user in each user cognitive low-level learning task base node are analyzed by using a self-adaptive learning algorithm, similar learning behaviors are classified through cluster analysis, variables (such as learning speed, memory, etc.) affecting the learning effect are identified, and then these variables are associated with the base node to generate a self-adaptive learning node, and a genetic algorithm is used to optimize the relationship between the nodes to ensure that the learning node not only reflects the existing knowledge of the user, but also contains personalized learning paths and strategies. The self-adaptive learning node formed should have a dynamic updating capability to adapt to new demands generated by the user in the learning process, and finally a user cognitive low-level self-adaptive learning node is obtained.

[0149] Preferably, the corresponding user cognitive low-level adaptive learning nodes are adaptively cognitively learned and network constructed based on the user cognitive knowledge high-level learning inference relationship, to generate an initial architecture of the user adaptive cognitive learning network.

[0150] In the embodiment of the present application, the connection integration of the cognitive learning network of the corresponding user cognitive low-level adaptive learning nodes is performed by combining the previously analyzed user cognitive knowledge high-level learning inference relationship, to utilize the graph neural network technology, regard the low-level adaptive learning nodes as nodes in the graph, learn the inference relationship as edges, design the network structure, and in the construction process, use the deep learning model to train the relationship between nodes, extract the implicit features generated in the user learning process, thereby optimizing the connection mode of the learning network, through multiple iterations and verifications, ensure that the network structure can effectively support the inference ability of the user in the complex cognitive learning task, thereby forming a dynamic and efficient adaptive cognitive learning network architecture, and finally connected to generate the initial architecture of the user adaptive cognitive learning network.

[0151] Further, the cognitive learning interaction bottleneck simulation analysis module includes the following functions:

[0152] Obtaining user real-time cognitive learning behavior data;

[0153] In the embodiment of the present application, the cognitive learning activities of the user are monitored in real time by using sensors, learning management systems (LMS) and data collection tools, these tools include learning progress trackers, online assessment platforms and interactive learning software, and through these systems, the collected data includes the learning time, answering speed, number of incorrect answers, frequency of participating in discussions and usage of learning resources of the user, etc., all data is aggregated to the central database through the API interface, to ensure the accuracy and timeliness of the data, this process provides a comprehensive view of the dynamic changes of the user in the learning environment, and finally integrated to obtain the user real-time cognitive learning behavior data.

[0154] Preferably, the user real-time cognitive learning behavior data is processed for learning behavior feature extraction to obtain user cognitive learning behavior features; the user real-time cognitive learning behavior data is cognitively learned and interaction mode recognized and analyzed based on the user cognitive learning behavior features, to obtain a user cognitive learning interaction behavior mode;

[0155] In the embodiments of the present application, the user real-time cognitive learning behavior data previously collected in real time is cleaned and preprocessed to remove noise and outliers, and the user learning behavior is analyzed in depth by using feature extraction algorithms such as principal component analysis (PCA) and clustering algorithm, and key features are extracted, including user learning efficiency (such as the ratio of time required to complete the task to accuracy), learning mode (such as the ratio of active learning to passive learning), and emotional state (such as anxiety level during learning), etc., so as to obtain user cognitive learning behavior features. At the same time, by combining the previously extracted user cognitive learning behavior features, a machine learning model (such as random forest or support vector machine) is applied to identify and analyze the cognitive learning interaction mode of the corresponding user cognitive learning behavior process in the user real-time cognitive learning behavior data, so as to identify the learning behavior interaction mode of the user, such as self-regulated learning, social learning or cooperative learning mode, and finally obtain the user cognitive learning interaction behavior mode.

[0156] Preferably, the user cognitive learning behavior features are divided and fused based on the user cognitive learning interaction behavior mode to obtain the learning feature fusion vector corresponding to each user cognitive learning interaction behavior;

[0157] In the embodiments of the present application, the corresponding user cognitive learning behavior features are divided into the same interaction behavior mode by combining the user cognitive learning interaction behavior mode obtained by the previous identification and analysis, so as to use clustering analysis to group the behavior features with similar interaction behavior modes, and generate the corresponding feature fusion vector for each group of behavior features. These vectors will contain the common features of each group of user cognitive learning behavior, such as typical performance and interaction mode in a specific learning task. The generation of the fusion vector uses the weighted average method to ensure that the influence degree of each behavior feature reflects its importance in the actual learning process, and finally obtains the learning feature fusion vector corresponding to each user cognitive learning interaction behavior.

[0158] Preferably, the learning feature fusion vector corresponding to each user cognitive learning interaction behavior is input into the user adaptive cognitive learning network initial architecture to perform cognitive learning interaction simulation analysis, and the cognitive learning interaction simulation process corresponding to each user cognitive learning interaction behavior is obtained;

[0159] In the embodiment of the present application, by inputting the learning feature fusion vector corresponding to each user cognitive learning interaction behavior generated by previous fusion into the user adaptive cognitive learning network initial architecture constructed previously, at this time, the user adaptive cognitive learning network framework is used to simulate and analyze the cognitive learning interaction process to predict the behavior performance and learning effect of each user in a specific learning environment, the network training uses the back propagation algorithm, and the network weight is optimized through multiple iterations, so that the accurate cognitive learning interaction process is simulated, the simulation result shows the expected path and learning achievement of the user in the learning process, and finally the cognitive learning interaction simulation process corresponding to each user cognitive learning interaction behavior is obtained.

[0160] Preferably, the cognitive learning interaction simulation process corresponding to each user cognitive learning interaction behavior is subjected to interactive simulation process bottleneck identification analysis to obtain the user cognitive learning interaction simulation process bottleneck point.

[0161] In the embodiment of the present application, after the cognitive learning interaction simulation process corresponding to each user cognitive learning interaction behavior is completed, the bottleneck identification analysis of the interactive simulation process is performed to identify the bottleneck points in the cognitive learning interaction simulation process by applying data mining techniques such as outlier detection and time series analysis, the bottleneck points include understanding obstacles of specific knowledge points, frequent error reactions or obvious lag of learning progress, etc., in the analysis process, the bottleneck data is displayed by combining the visualization tool, so that the identification result is more intuitive, and finally the corresponding user cognitive learning interaction simulation process bottleneck point is obtained.

[0162] Further, the cognitive learning network feedback optimization module comprises the following functions:

[0163] The cognitive learning interaction bottleneck feature analysis is performed on the user cognitive learning interaction simulation process bottleneck point to obtain user cognitive learning interaction process bottleneck feature description data.

[0164] In the embodiment of the present application, the cognitive learning bottleneck feature of the user cognitive learning interaction simulation process bottleneck point (for example, the stagnation time of the user on a specific knowledge learning point is too long) obtained by previous analysis is subjected to statistical analysis to statistically analyze the bottleneck performance of the user in a specific cognitive learning situation, so as to reveal the feature description factors affecting the cognitive learning efficiency, such as the distraction of the user, the unstable mastery of the knowledge point or the low interactive participation, etc., and help the system to more deeply understand the obstacles of the user in the cognitive learning process, for example, a user frequently produces errors when learning mathematical concepts due to lack of basic knowledge, resulting in low learning efficiency, and finally the user cognitive learning interaction process bottleneck feature description data is obtained.

[0165] Preferably, the user adaptive cognitive learning network initial architecture is analyzed for learning network optimization target based on the user cognitive learning interaction process bottleneck feature description data, and a user cognitive learning network optimization target list is obtained.

[0166] In the embodiment of the present application, the user adaptive cognitive learning network initial architecture used in the cognitive learning interaction process is analyzed for learning network optimization target identification based on the user cognitive learning interaction process bottleneck feature description data obtained by previous analysis, so as to map the bottleneck feature data to the structure of the learning network by using the system thinking theory, and to clearly define the optimization targets, including reducing the cognitive load of the user, improving the depth of knowledge mastery, and improving the learning initiative, wherein each optimization target corresponds to a specific quantitative index, such as a 20% reduction in error rate, a 15% improvement in learning initiative, etc., and by constructing an optimization target list, the relevance and feasibility between the optimization targets are ensured, and finally a user cognitive learning network optimization target list is obtained.

[0167] Preferably, the user cognitive learning network optimization target list is analyzed for learning network optimization strategy, and a user cognitive learning network optimization adjustment strategy scheme is obtained.

[0168] In the embodiment of the present application, after the optimization target list of the user cognitive learning network is determined, the learning network optimization strategy is analyzed, so as to evaluate the performance of the current learning network by using the model prediction technology, and compare it with the optimization target, then, a variety of optimization strategies are designed by using the reinforcement learning algorithm, for example, for users with too high cognitive load, an optimization strategy based on hierarchical teaching method is implemented, which provides step-by-step guidance and timely feedback to relieve learning pressure, at the same time, the strategy is evaluated by using the simulation tool to ensure the effectiveness and operability of the strategy, so as to form a comprehensive optimization adjustment strategy scheme, which covers different cognitive learning links and individual differences of users, and finally the user cognitive learning network optimization adjustment strategy scheme is obtained.

[0169] Preferably, the user adaptive cognitive learning network initial architecture is optimized for cognitive learning network feedback based on the user cognitive learning network optimization adjustment strategy scheme, so as to generate a user adaptive cognitive learning optimization adjustment network.

[0170] In the embodiment of the present application, the initial architecture of the user adaptive cognitive learning network is optimized and adjusted by combining the user cognitive learning network optimization adjustment strategy scheme obtained through previous analysis, so as to analyze the new data of the user in the learning process in real time by using the machine learning algorithm, and respond to the learning needs of the user by adjusting the network parameters and the learning path, for example, if a user shows a very high frustration at a certain link, the learning content will be dynamically adjusted, the corresponding basic knowledge review materials will be provided, and the interactive exercises will be increased to enhance the cognitive learning confidence, so as to generate the corresponding adaptive cognitive learning optimization network, which will effectively improve the learning efficiency and effect of the user, realize the goal of personalized cognitive learning, and finally optimize the generation of the user adaptive cognitive learning optimization adjustment network.

[0171] Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting, the scope of the present application being defined by the appended claims and not by the above description, therefore all variations falling within the meaning and scope of the equivalent requirements of the application file are intended to be included in the above description which is merely a specific embodiment of the present application, enabling a person skilled in the art to understand or implement the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to these embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A large language model-based adaptive cognitive learning network simulation system, characterized in that, Comprise the following modules: User cognitive learning feature vector conversion module, for obtaining user historical cognitive learning data, user cognitive interaction record data and user cognitive learning feedback information data, and performing cognitive feature vector fusion conversion according to the user historical cognitive learning data, the user cognitive interaction record data and the user cognitive learning feedback information data, to obtain the user cognitive learning feature vector; Hierarchical cognitive reasoning and learning network construction module, for hierarchical cognitive reasoning analysis of the user cognitive learning feature vector based on a large language model, to obtain the user cognitive knowledge low-level learning task and the user cognitive knowledge high-level learning reasoning relationship; Adaptive cognitive learning network construction based on the user cognitive knowledge low-level learning task and the user cognitive knowledge high-level learning reasoning relationship, to generate the user adaptive cognitive learning network initial architecture; wherein the hierarchical cognitive reasoning analysis of the user cognitive learning feature vector based on the large language model comprises: Low-level task identification analysis of the user cognitive learning feature vector based on the low-level cognitive module in the large language model, to obtain the user cognitive knowledge low-level learning task; wherein the user cognitive knowledge low-level learning task includes memory task, understanding task and application task; Task content classification processing of the user cognitive knowledge low-level learning task, to obtain the user cognitive low-level learning task content classification matrix; High-level cognitive reasoning framework for processing low-level learning tasks is constructed through the high-level cognitive module in the large language model, and high-level cognitive reasoning path analysis of the corresponding low-level learning task in the user cognitive low-level learning task content classification matrix is performed based on the high-level cognitive reasoning framework, to obtain the high-level cognitive reasoning path between the low-level learning tasks; High-level learning reasoning adjustment analysis of the corresponding user cognitive knowledge low-level learning task based on the high-level cognitive reasoning path between the low-level learning tasks, to obtain the user cognitive knowledge high-level learning reasoning relationship; Cognitive learning interaction bottleneck simulation analysis module, for obtaining user real-time cognitive learning behavior data, and performing cognitive learning interaction bottleneck simulation analysis of the user adaptive cognitive learning network initial architecture based on the user real-time cognitive learning behavior data, to obtain the user cognitive learning interaction simulation process bottleneck point; Cognitive learning network feedback optimization module, for cognitive learning network feedback optimization adjustment of the user adaptive cognitive learning network initial architecture based on the user cognitive learning interaction simulation process bottleneck point, to generate the user adaptive cognitive learning optimization adjustment network.

2. The large language model based adaptive cognitive learning network simulation system according to claim 1, wherein, The user cognitive learning feature vector analysis module comprises the following functions: Obtain the user historical cognitive learning data by obtaining the historical cognitive learning data record from the user cognitive learning platform database; Obtain the user historical cognitive learning activity trajectory by performing cognitive learning trajectory analysis on the user historical cognitive learning data; Obtain the user cognitive interaction record data by performing cognitive interaction analysis and recording on the user historical cognitive learning data based on the user historical cognitive learning activity trajectory; construct a user cognitive interaction quality evaluation system according to the user cognitive interaction record data, and perform cognitive learning feedback analysis on the user historical cognitive learning data based on the user cognitive interaction quality evaluation system to obtain user cognitive learning feedback information data; perform cognitive feature vector fusion conversion on the user historical cognitive learning data, the user cognitive interaction record data, and the user cognitive learning feedback information data to obtain a user cognitive learning feature vector.

3. The large language model based adaptive cognitive learning network simulation system according to claim 2, wherein, The cognitive learning trajectory analysis on the user historical cognitive learning data includes: perform cognitive learning time and content extraction processing on the user historical cognitive learning data to obtain user historical cognitive learning timestamps and user historical cognitive learning content; perform cognitive learning theme mining analysis on the user historical cognitive learning content to obtain various user historical cognitive learning activity themes; perform correlation analysis on the various user historical cognitive learning activity themes to obtain learning activity behavior correlation relationships between the various user cognitive learning activity themes; perform cognitive learning behavior correlation graph analysis on the user historical cognitive learning content based on the learning activity behavior correlation relationships between the various user cognitive learning activity themes to generate a user historical cognitive learning behavior correlation graph; perform cognitive learning trajectory analysis on the user historical cognitive learning behavior correlation graph based on the user historical cognitive learning timestamps to obtain user historical cognitive learning activity trajectories.

4. The large language model based adaptive cognitive learning network simulation system according to claim 2, wherein, The cognitive feature vector fusion conversion on the user historical cognitive learning data, the user cognitive interaction record data, and the user cognitive learning feedback information data includes: perform cognitive learning content feature analysis on the user historical cognitive learning data to obtain user cognitive learning content features, wherein the user cognitive learning content features include user cognitive learning content themes, user cognitive learning difficulty levels, and user cognitive learning time lengths; perform user interaction frequency feature analysis on the user cognitive interaction record data to obtain user cognitive learning interaction frequency features, wherein the user cognitive learning interaction frequency features include user cognitive learning click rates, user cognitive learning viewing time lengths, and user cognitive learning discussion participation degrees; perform cognitive learning feedback feature analysis on the user cognitive learning feedback information data to obtain user cognitive learning feedback sentiment features, wherein the user cognitive learning feedback sentiment features include user cognitive learning positive feedback sentiment indexes, user cognitive learning neutral feedback sentiment indexes, and user cognitive learning negative feedback sentiment indexes; perform cognitive feature vector fusion conversion on the user cognitive learning content features, the user cognitive learning interaction frequency features, and the user cognitive learning feedback sentiment features to obtain a user cognitive learning feature vector.

5. The large language model based adaptive cognitive learning network simulation system according to claim 1, wherein, The high-level learning reasoning adjustment analysis on the corresponding user cognitive knowledge low-level learning tasks based on the high-level cognitive reasoning paths between the low-level learning tasks includes: perform cognitive reasoning path relationship connection processing on the corresponding user cognitive knowledge low-level learning tasks based on the high-level cognitive reasoning paths between the low-level learning tasks to generate a high-level cognitive reasoning path relationship graph between the low-level learning tasks; The learning task reasoning relationship strength calculation formula is used to evaluate and calculate the high-level cognitive reasoning path relationship graph between the low-level learning tasks, to obtain the cognitive reasoning relationship strength between the low-level learning tasks. The high-level cognitive reasoning path relationship graph between the low-level learning tasks is adjusted and analyzed based on the cognitive reasoning relationship strength between the low-level learning tasks, to obtain the high-level learning reasoning relationship of the user cognitive knowledge.

6. The large language model based adaptive cognitive learning network simulation system according to claim 5, wherein, The learning task reasoning relationship strength calculation formula is specifically: ； Where, For the A low-level learning task and The strength of cognitive reasoning relationships between low-level learning tasks, and are all item-level measurement parameters for low-level learning tasks, The upper limit of the time range for calculating the strength of the inference relationship, is the time variable parameter, For the A low-level learning task and The reasoning relationship between low-level learning tasks affects the weight, For the A low-level learning task and low-level learning tasks in time The number of relevant reasoning paths on , is the attenuation coefficient of the number of inference paths, For the A low-level learning task and low-level learning tasks in time The similarity of the task content on is the attenuation coefficient of task content similarity, For the A low-level learning task and low-level learning tasks in time Context-dependent parameters on is the context-dependent impact attenuation coefficient, is the correction coefficient of the strength of cognitive reasoning relationship.

7. The large language model based adaptive cognitive learning network simulation system according to claim 1, wherein, The self-adaptive cognitive learning network construction based on the low-level learning task of the user cognitive knowledge and the high-level learning reasoning relationship of the user cognitive knowledge includes: The low-level learning task base node of the user cognitive knowledge is configured, to obtain the low-level learning task base node of the user cognitive knowledge; The low-level adaptive learning node of the user cognitive knowledge is transformed based on the low-level learning task base node of the user cognitive knowledge, to obtain the low-level adaptive learning node of the user cognitive knowledge; The corresponding low-level adaptive learning node of the user cognitive knowledge is constructed into a self-adaptive cognitive learning network based on the high-level learning reasoning relationship of the user cognitive knowledge, to generate an initial architecture of the self-adaptive cognitive learning network of the user.

8. The large language model based adaptive cognitive learning network simulation system of claim 1, wherein, The cognitive learning interaction bottleneck simulation analysis module includes the following functions: Real-time cognitive learning behavior data of the user is obtained; Learning behavior feature extraction processing is performed on the real-time cognitive learning behavior data of the user, to obtain cognitive learning behavior features of the user; The real-time cognitive learning behavior data of the user is analyzed based on the cognitive learning behavior features of the user, to obtain a cognitive learning interaction behavior mode of the user; The cognitive learning behavior features of the user are divided and fused based on the cognitive learning interaction behavior mode of the user, to obtain a learning feature fusion vector corresponding to each cognitive learning interaction behavior of the user; The cognitive learning interaction simulation process corresponding to each cognitive learning interaction behavior of the user is obtained by inputting the learning feature fusion vector corresponding to each cognitive learning interaction behavior of the user into the initial architecture of the self-adaptive cognitive learning network of the user. The cognitive learning interaction bottleneck feature of the user is analyzed based on the cognitive learning interaction simulation process bottleneck point of the user, to obtain cognitive learning interaction process bottleneck feature description data of the user; 9. The large language model based adaptive cognitive learning network simulation system according to claim 1, wherein, The learning network optimization target analysis is performed on the initial architecture of the self-adaptive cognitive learning network of the user based on the cognitive learning interaction process bottleneck feature description data of the user, to obtain a learning network optimization target list of the user; The learning network optimization strategy analysis is performed on the learning network optimization target list of the user, to obtain a learning network optimization adjustment strategy scheme of the user; The cognitive learning network feedback optimization adjustment is performed on the initial architecture of the self-adaptive cognitive learning network of the user based on the learning network optimization adjustment strategy scheme of the user, to generate an optimized and adjusted self-adaptive cognitive learning network of the user. ​ ​

Images