16 results about "Cognitive diagnosis" patented technology

The application provides a dynamic graph neural cognitive diagnosis method fusing multi-dimensional features, and the method comprises the following steps: collecting multi-modal data of an online learning platform, and obtaining a multi-dimensional feature splicing vector based on the multi-modal data; based on the multi-dimensional feature splicing vector and by using a graph neural network, obtaining an aggregation matrix from an embedding matrix of a current knowledge point, a hidden state matrix of the current knowledge point, an embedding matrix of a neighboring knowledge point, and a hidden state matrix of the neighboring knowledge point; performing a forgetting process on the embedding matrix of the knowledge point by using a forgetting vector, inputting the aggregation matrix into a memory gate structure model, and capturing a time sequence feature embedding matrix of the knowledge point; based on the time sequence feature embedding matrix and by using the graph neural network, constantly updating the hidden state matrix of the knowledge point, and obtaining an updated knowledge mastery degree of the learner at the next moment. The application comprehensively diagnoses the learning effect of the learner and updates the knowledge mastery state of the learner, and greatly improves the accuracy of the diagnosis result.

The application provides a cognitive diagnosis method and system based on causal inference to eliminate problem difficulty bias, relates to the technical field of educational data mining, and specifically includes the following steps: preprocessing the historical record of a student's answer to a problem to obtain a training set composed of a student, a problem, problem difficulty and a score; introducing a problem difficulty variable on the basis of an existing diagnosis model, eliminating bias by using causal inference, constructing a cognitive diagnosis model, and training the cognitive diagnosis model by using the training set; inputting a student to be diagnosed, a problem and problem difficulty into the trained cognitive diagnosis model to obtain a cognitive diagnosis result of the student output by the diagnosis model; the application provides a general cognitive diagnosis framework, creates a cognitive diagnosis model based on an existing diagnosis model, eliminates the adverse effects of problem difficulty bias by using causal inference technology, and performs unbiased training on the model to diagnose the cognitive level of the student, thereby improving the accuracy and efficiency of cognitive diagnosis.

This application belongs to the field of intelligent assessment technology, specifically disclosing a matrix optimization method, apparatus, and device based on a cognitive diagnostic model. This application removes the questions to be verified from the correlation matrix to be optimized and the corresponding answer data from the answer matrix; it determines the student's knowledge point mastery vector and question feature vector based on the target cognitive diagnostic model; and it sequentially verifies multiple knowledge points associated with the questions to be verified according to a preset positive testing strategy, optimizing the correlation matrix by correcting the knowledge points to be optimized. By removing the questions to be verified and their corresponding answer data from their respective matrices, the accuracy of the input model's data is improved. The preset positive testing strategy ensures that knowledge points in the correlation matrix to be optimized are corrected only when sufficient and significant statistical evidence is obtained, thereby effectively improving the accuracy and efficiency of the optimization matrix.

The application relates to the technical field of artificial intelligence, and discloses an emotional conversation generation method and device, electronic equipment and a storage medium, the method comprising the following steps: constructing an emotional conversation dataset and constructing a plurality of intelligent agents based on a pre-trained large language model, which respectively play the roles of a seeker and a consultant; obtaining current conversation text received by a first intelligent agent corresponding to the seeker, mapping a target cognitive distortion classification system of the current conversation text based on the emotional conversation dataset; determining the cognitive bias of the current user based on the target cognitive distortion classification system; determining a target intervention strategy of the current conversation text based on the cognitive bias by using a second intelligent agent corresponding to the consultant, a preset hybrid reward mechanism and a cognitive strategy reinforcement learning algorithm; and optimizing reply text in the target intervention strategy to generate a target reply text, so that the cognitive diagnosis and cognitive intervention strategy accuracy for the seeker are improved, and accurate identification and professional intervention of cognitive distortion of the seeker are realized.

This invention relates to the field of artificial intelligence-based educational assessment, and discloses a cognitive diagnostic model and method based on knowledge text. The model includes a text feature extraction module for extracting and generating knowledge point association vectors representing the degree of association between questions and preset knowledge points; a knowledge proficiency embedding module for mapping knowledge proficiency vectors; a question attribute embedding module for mapping question knowledge difficulty vectors and question discrimination scalars; a knowledge association attention module for adjusting attention weights and generating weighted knowledge representations; and an adaptive nonlinear prediction module containing a multi-layer KANLinear structure composed of B-spline basis functions and outputting the prediction probability of correct student answers. This invention solves the problem of fragmented internal module functions in existing models and features high interpretability and independence from manual annotation.

The application provides a kind of personalized learning path recommendation method based on reinforcement learning, it is related to educational data mining technical field.The method first constructs learner simulator according to the learning record of scholar, the simulator can judge the learning level of learner;Then the knowledge relationship graph between the knowledge points contained in the exercise is automatically constructed by the concept map automatic construction model based on text classification and association rule mining;Based on the knowledge relationship graph and the cognitive diagnosis model, an exercise navigation module is designed to select potential candidate exercises;After the reinforcement learning agent selects the action in the action space, the state transition is determined in the state space, the model parameters are updated according to the loss function and optimization strategy, and the reinforcement learning model is optimized;Finally, the designed reinforcement learning model is used to recommend exercises for learners, and the reinforcement learning model parameters are updated according to the learning situation of learners.The method can recommend efficient and reasonable learning path for learners.

This invention relates to a deep cognitive diagnostic method based on graph neural networks, belonging to the field of deep learning technology. The invention aims to represent the cognitive diagnostic task as a representation learning and prediction problem on a heterogeneous graph composed of student nodes, item nodes, and ability nodes. While maintaining the student-item-score triplet input format, the item-ability mapping relationship represented by the Q-matrix is ​​transformed into an explicit connection relationship in the graph structure. Furthermore, the message passing mechanism of the graph neural network is used to achieve information interaction and joint modeling between different types of nodes, thereby improving the accuracy of student answer prediction, the granularity of student ability assessment, and the interpretability of the model output.

This invention discloses a dynamic teaching strategy optimization method for personalized learning, relating to the field of intelligent teaching optimization. The method includes acquiring emotional, behavioral, and academic data; using a cognitive diagnostic unit to perform correlation and feature fusion processing on the behavioral and academic data to obtain a knowledge point mastery probability matrix and a knowledge gap association strength matrix; using an affective computing unit to extract and fuse features from the emotional data to obtain affective fusion features; classifying and scoring the affective fusion features to obtain affective intensity and affective stability scores; constructing a fusion reinforcement learning model; and generating and adjusting dynamic teaching strategies based on the knowledge point mastery probability matrix, the knowledge gap association strength matrix, the affective intensity score, and the affective stability score. This invention improves the timeliness and accuracy of cognitive diagnostic parameters using the above method, supporting precise identification of knowledge weaknesses.

The application discloses a kind of multi-dimensional evaluation method and system of VR teaching based on user behavior feedback, according to Bloom's taxonomy of educational objectives, construct the multi-dimensional evaluation framework covering cognition, skill and emotion / meta-cognition, and set observable behavior proxy index for each dimension;Through VR equipment, real-time collection of user multi-modal behavior data and teaching context, knowledge state is identified and backstepped by using Transformer and cognitive diagnosis model;Combining graph neural network and attention mechanism, multi-granularity evaluation of micro, meso and macro three layers fusion is realized;At the same time, based on historical behavior, individualized ability curve is constructed, novice and expert behavior are dynamically distinguished, evaluation standard is adaptively adjusted, and growth is emphasized rather than absolute score;Finally, the intelligent feedback for students and teachers is generated, and teaching intervention is driven.The system realizes the whole process, objective and individualized VR teaching evaluation, effectively improves teaching accuracy and learning effectiveness.

The application discloses a kind of based on multi-objective optimization's selecting subject method, comprising:1 obtains student and subject relevant data;2 training bottom cognitive diagnosis model;3 based on accuracy, diversity and security constructs multi-objective optimization problem;4 using heuristic multi-objective evolutionary algorithm to solve multi-objective optimization problem, obtain the student's selecting subject set.The application based on multi-objective evolutionary algorithm balances test accuracy, diversity and security, and by calculating and the answer prediction accuracy of the student similar to its ability level to the test accuracy of the student more intuitive and effective modeling, to ensure the prediction accuracy of student knowledge level ability.

This invention relates to a cognitive diagnosis and reverse thinking training education system based on dual AI collaboration, belonging to the fields of educational technology and artificial intelligence technology. The system includes: acquiring and preprocessing multimodal data from students' learning process, outputting multidimensional cognitive feature vectors; dynamically dividing cognitive levels based on these vectors, constructing differentiated cognitive profiles, and generating personalized training task matrices by combining them with a mapping rule base; locating cognitive gaps and logical error paradigms, calling a metacognitive question chain library to generate a progressive question sequence, and combining it with scenario-based verification tasks to form a set of error correction and knowledge reinforcement; constructing a pre-construction common difficulty training module for a group cognitive graph, building a dynamic mastery assessment index system, and generating multidimensional ability feedback reports. This achieves accurate diagnosis of students' cognitive states and personalized training adaptation, efficiently repairing knowledge gaps, strengthening reverse thinking abilities, and improving the pertinence and effectiveness of learning and training.

This invention belongs to the fields of artificial intelligence and educational big data mining and analysis. Specifically, it is a method and system for constructing a multi-dimensional profile of college students based on dynamic and static knowledge graphs and large models. The method includes steps such as task initiation and academic path retrieval plan generation, dynamic and static knowledge graph fusion retrieval and cross-subgraph semantic aggregation, data decoupling and cognitive diagnosis, and feature fusion and profile generation. This enables cognitive diagnosis and multi-dimensional profile construction for college students. This invention addresses the problems in existing technologies, such as insufficient integration of dynamic and static data, lack of analysis of college students' learning processes, and lack of objective evaluation indicators.

This invention discloses a neural collapse-driven dual long-tail cognitive diagnosis method and system. Based on the NeuralCDM model, the method introduces a regularization constraint module, which employs two pseudo-label generation methods: dynamic clustering based on embedding vectors and classification based on statistical profiling. The former directly performs K-Means clustering on the high-dimensional feature vectors of students or exercises during training; the latter dynamically divides all students into four classes based on the statistical properties of student feature vectors. Furthermore, two geometric regularization losses are designed for students and exercises: intra-class cohesion regularization and inter-class separation regularization. By jointly optimizing the diagnostic loss and geometric constraints, this invention can induce a highly discriminative structured feature space while maintaining reasonable feature compactness, thereby significantly improving the model's accuracy in long-tail cognitive diagnosis tasks.

This invention relates to the field of learning path planning technology, and provides an adaptive AI learning path planning system based on learning profiles. The system includes a multi-source learning data acquisition module, a data preprocessing module, a three-dimensional cognitive diagnosis module, a multi-dimensional dynamic learning profile construction module, a subject knowledge graph module, an adaptive learning path planning module, and a dynamic path adjustment module. The multi-source learning data acquisition module collects multi-source learning data from students, including basic information data, historical learning data, initial assessment data, and real-time interaction data during the learning process. By integrating dynamic memory states based on the Ebbinghaus forgetting curve, the system can track students' memory retention in real time, ensuring the rationality of the learning sequence, avoiding knowledge gaps, improving learning effectiveness, achieving multi-objective optimization, improving the rationality of path planning, and enhancing the accuracy of cognitive diagnosis, thus providing a more accurate foundation for learning profiles and path planning.

The application discloses a cold start cognitive diagnosis method based on a multi-granularity context enhanced large language model. The application comprises the following steps: firstly, a mixed skill graph is constructed by fusing the semantic association and the time sequence association between skills, historical skill records are extracted, and micro-skill historical information is classified and labeled; secondly, a perception state encoder is constructed, a cognitive state vector is obtained by fusing a question and environmental characteristics, and medium-scale answer performance information of a learning trajectory similar peer is retrieved; then, the objective correct rate of multi-dimensional educational attributes is extracted and mapped into a text difficulty label as macro reference correction information; finally, the multi-granularity context is converted into a structured prompt word to input a large language model, and the final prediction result and a diagnosis report are output through a four-step diagnosis reasoning chain. The application is suitable for an educational intelligent tutoring system, a new type of framework combining multi-granularity teaching context and large model reasoning is designed, and the prediction accuracy and interpretability of the model in a cold start data sparse environment are significantly improved.