Atlas-enhanced intelligent game and operation planning optimization auxiliary decision-making system and method

Abstract

The invention relates to the technical field of auxiliary decision making, and discloses a graph-enhanced intelligent game and operation planning optimization auxiliary decision making system and method, and the method comprises the steps: inputting multi-source heterogeneous data into a space-time semantic mapping frame, and generating a unified space-time semantic data structure; generating a cognitive-spatio-temporal manifold model through a cognitive spatio-temporal manifold generation algorithm; obtaining user interaction data, and generating an adaptive view expression through an interaction response manifold deformation algorithm; obtaining user interaction behavior data, and generating a cognitive balance view through a cognitive load perception projection algorithm; generating an interactive decision-making scene through an immersive space-time scene rendering technology; according to the method, a complex game and operation planning optimization problem is mapped into a visual space-time scene, so that a decision maker can foresee and control a decision result in a virtual space-time space, and the understandability and controllability of game and operation planning optimization are greatly enhanced.

Description

Technical Field

[0001] The present invention relates to the field of decision support technology, and more specifically, to a graph-enhanced intelligent game and operations optimization decision support system and method. Background Art

[0002] In the field of complex game and operational optimization decision-making, the existing decision support system has significant technical bottlenecks. Traditional methods mainly use two-dimensional charts and statistical panels to present data, and their technical limitations are reflected in: Dilemma of expressing high-dimensional abstract data: Existing visualization technologies are difficult to effectively process complex decision-making data with more than three dimensions, which makes it difficult for decision makers to intuitively understand the nonlinear relationship between multi-dimensional data. Experiments show that when the data dimension exceeds 7 dimensions, the decision-making understanding efficiency of traditional systems decreases by 73%; Time dimension fragmentation problem: The existing system uses discrete time slices to present decision impacts, and cannot establish a continuous correlation model for multiple time scales. Statistics show that this fragmented presentation increases the probability of decision makers ignoring long-term impacts by 58%; Collaborative decision-making cognitive bias: Traditional multi-user systems only realize data sharing and lack cognitive alignment mechanisms. Studies have shown that cognitive differences between decision makers with different professional backgrounds can lead to 42% of effective information being lost during the collaborative process; Dimensionality reduction information loss paradox: In order to adapt to human cognitive ability, existing visualization methods generally use linear dimensionality reduction techniques such as PCA, but this will lead to the loss of key decision features. Test data shows that traditional methods lose an average of 36% of decision-related features during dimensionality reduction.

[0003] In modern military command and control systems, commanders need to simultaneously process multi-dimensional information such as terrain, weather, troop status, logistics, enemy situation, etc., and make decisions in an environment where the battlefield situation changes rapidly. Traditional combat simulation systems usually display this information in multiple interfaces, causing commanders to frequently switch perspectives and making it difficult to form a unified battlefield cognition. The survey shows that in a high-intensity combat simulation environment, commanders need to switch interfaces 8.3 times per minute on average, and 41% of key information is ignored during the perspective switching process.

[0004] In the field of e-sports and strategy games, high-level players need to handle dozens of variables such as unit attributes, resource allocation, map control, skill cooldown, etc., and respond at the millisecond level. Existing game interface designs mainly rely on two-dimensional plane layouts and linear timelines, and players need to form psychological models through long-term experience. Data shows that even professional players can only process about 42% of key strategic information at the same time under high-pressure environments, resulting in a significant decline in decision-making quality.

[0005] In scenarios such as large-scale disaster emergency response and multinational corporate strategic planning, decision makers from different professional fields need to jointly participate in complex decision-making. Due to differences in professional terminology, thinking patterns, and focus, traditional systems cannot provide effective cognitive alignment support. Actual case analysis shows that in cross-domain collaborative decision-making, about 67% of communication time is spent on clarifying basic concepts and explaining professional perspectives, rather than discussing the decision itself.

[0006] These technical defects make it difficult for existing systems to meet the triangular requirements of "high-dimensional data understanding - multi-time scale reasoning - group cognitive collaboration" in modern complex decision-making scenarios, seriously restricting the quality and efficiency of decision-making. Summary of the invention

[0007] The present invention provides a graph-enhanced intelligent game and operations optimization auxiliary decision-making system and method to solve the technical problems in the above-mentioned related technologies.

[0008] The present invention provides a graph-enhanced intelligent game and operations optimization auxiliary decision-making method, comprising the following steps: Obtain multi-source heterogeneous data and input them into the spatiotemporal semantic mapping framework to generate a unified spatiotemporal semantic data structure; The unified spatiotemporal semantic data structure generates a cognitive-spatiotemporal manifold model through a cognitive spatiotemporal manifold generation algorithm; Obtain user interaction data and cognitive-space-time manifold model, and generate adaptive view expression through interactive response manifold deformation algorithm; Obtain user interaction behavior data and adaptive view expression, and generate cognitive balance view through cognitive load-aware projection algorithm; Through immersive space-time scene rendering technology, interactive decision-making scenarios and multimodal interaction interfaces are generated.

[0009] Furthermore, the spatiotemporal semantic mapping framework adopts a multi-head self-attention mechanism to realize cross-domain semantic feature extraction and alignment:

[0010] in Indicates the aligned data. Represents the source data, , and Represent the semantic descriptors of time dimension, space dimension and action dimension respectively, is the semantic alignment function.

[0011] Furthermore, the interactive response manifold deformation algorithm calculates the manifold deformation process through the following partial differential equation:

[0012] in represents the user attention distribution at time t, Represents the decision operation performed by the user, represents the attention response operator, represents the decision response operator, Represents cognitive-spacetime manifold.

[0013] Furthermore, the interactive response manifold deformation algorithm adopts an improved implicit Euler method for numerical solution:

[0014] in represents the state of the manifold at time t, express The manifold state at the moment, is the time step.

[0015] Furthermore, the cognitive load perception projection algorithm is implemented by the following steps: analyzing the user interaction behavior data to calculate the cognitive load level value:

[0016] in represents the calculated cognitive load level value, Represents the user's historical interaction records. represents the user's gaze pattern data, Indicates the user's operation response time data. Evaluate functions for cognitive load; Determine the dimension and method of data projection according to the cognitive load level value:

[0017] in represents the original high-dimensional data, Represents the current decision context information, represents the adaptive projection result, is the adaptive projection function; Generate semantic explanations for the projection results:

[0018] in Represents the original data, Represents the generated projection interpretation text, Generate functions for interpretation.

[0019] Furthermore, the cognitive load-aware projection algorithm automatically selects an appropriate dimensionality reduction method according to the cognitive load value: when When , the nonlinear dimensionality reduction method that retains the most information is used; when When , a dimensionality reduction method that balances information retention and computational efficiency is used; when When , the computationally simplest linear dimensionality reduction method is used; in Indicates a low threshold of cognitive load, Indicates a high threshold of cognitive load, Represents the cognitive load value.

[0020] Furthermore, the immersive spatiotemporal scene rendering technology is implemented by the following steps: performing spatiotemporal scene mapping to convert the cognitive balance view into a visual space representation:

[0021] in Represents a set of visual style parameters, Indicates user preference configuration, is the scene mapping function, represents visual scene data, Representation of cognitive-space-time manifold model; Based on the visual space representation, interactive scene rendering is performed to generate an interactive scene:

[0022] in Indicates the interaction parameter configuration, represents the set of physical dynamic parameters, represents the generated interactive scene, Rendering function for the scene.

[0023] Furthermore, after generating the cognitive balance view, the method further includes: Obtaining the interaction data of multiple decision makers and the cognitive balance view, and generating a collective cognitive structure and cognitive guidance strategy through a multi-agent cognitive fusion algorithm; The collective cognitive structure and cognitive guidance strategy are provided as additional input to the immersive spatiotemporal scene rendering technology to enhance the collaborative decision-making capability of the interactive decision-making scenario.

[0024] Furthermore, the multi-agent cognitive fusion algorithm is implemented by the following steps: Extract individual cognitive models from the interaction data of each decision maker:

[0025] in, represents the constructed i-th user cognitive model, represents the interaction history data of the i-th user, represents the attention distribution data of the i-th user, represents the decision behavior data of the i-th user, Extract functions for cognitive models; Multiple individual cognitive models are integrated into a collective cognitive model through the graph neural network algorithm:

[0026] in, represents a graph neural network function, Represents the adjacency matrix between models, characterizing the complementary or conflicting relationships between different individual cognitive models. Indicates from index arrive A collection of individual cognitive models, represents the total number of decision makers, represents a collective cognitive model; Based on the collective cognition model, identify potential blind spots in group cognition:

[0027] in, represents the set of identified cognitive blind spots, A data representation representing the complete decision space, is the detection threshold parameter, represents the blind spot detection function; Generate guidance strategies for identified cognitive blind spots:

[0028] in, represents the set of cognitive guidance strategies generated, represents the set of cognitive blind spots, represents the current decision context, represents the policy generation function.

[0029] A graph-enhanced intelligent game and operational optimization decision-making support system, which is used to execute the above-mentioned graph-enhanced intelligent game and operational optimization decision-making support method, comprising: The spatiotemporal semantic mapping module is used to convert multi-source heterogeneous data into a unified spatiotemporal semantic data structure; A cognitive manifold generation module, which is used to transform the unified spatiotemporal semantic data structure into a cognitive-spatiotemporal manifold model; The manifold deformation module is used to adjust the cognitive-spatiotemporal manifold model according to user interaction data to generate adaptive view representations; The cognitive load perception module is used to analyze user interaction behavior data, dynamically adjust the dimensions and details of data projection, and generate a cognitive balance view; A spatiotemporal scene rendering module for converting cognitive balance views into interactive decision-making scenarios; Multimodal interaction module, used to provide gesture interaction, voice interaction and collaborative interaction functions; The decision simulation module is used to support users in planning decision paths and performing simulations in scenarios.

[0030] The beneficial effects of the present invention are: Cognitive-space-time manifold mapping technology: Innovatively proposes to map game and operational optimization problems from abstract mathematical space to cognitive-space-time manifold, so that abstract data can be visualized into perceptible and operable space-time scenes. This mapping process is different from traditional data visualization technology. It not only considers the geometric characteristics of data, but also incorporates human cognitive laws, realizing a paradigm shift from "mathematical expression" to "cognitive expression".

[0031] Interaction-induced manifold deformation mechanism: A breakthrough in the bidirectional coupling of human cognition and data expression. In traditional visualization systems, data expression is preset and static, while in this implementation, data expression will dynamically deform according to the user's cognitive focus and interactive behavior, achieving real-time adaptation of data expression and cognitive process.

[0032] Cognitive Adaptive Projection Algorithm: Innovatively solves the contradiction between "data complexity" and "cognitive comprehensibility". Unlike traditional methods that improve comprehensibility by simplifying data, this algorithm retains the full complexity of the data, but dynamically adjusts the dimensions and details of the data presentation according to cognitive focus, achieving a dynamic balance between complexity and comprehensibility at different attention levels.

[0033] Collective cognitive enhancement mechanism: It pioneered the integration and enhancement of cognitive models of multiple decision makers. The system can capture and integrate the cognitive patterns of multiple decision makers, build a "super-individual cognitive structure" that transcends individuals, and actively identify group cognitive blind spots, guiding decision makers to focus on neglected areas of decision space. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a flow chart of a graph-enhanced intelligent game and operations optimization decision-making assistance method of the present invention. DETAILED DESCRIPTION

[0035] The subject matter described herein will now be discussed with reference to example implementations. It should be understood that the discussion of these implementations is only to enable those skilled in the art to better understand and implement the subject matter described herein, and the functions and arrangements of the elements discussed may be changed without departing from the scope of protection of the present specification. Various examples may omit, replace, or add various processes or components as needed. In addition, the features described in some examples may also be combined in other examples.

[0036] Implementation method 1: a graph-enhanced intelligent game and operations optimization decision-making assistance method, such as Figure 1 As shown, the following steps are included: Step 100, obtaining multi-source heterogeneous data and inputting them into a spatiotemporal semantic mapping framework to generate a unified spatiotemporal semantic data structure; Specifically, multi-source heterogeneous data such as game strategy data, operations optimization target data, historical decision data, and environmental constraint data are input into the spatiotemporal semantic mapping framework. After processing, a normalized data structure with unified spatiotemporal semantics is output, which supports unified retrieval and analysis operations across data sources.

[0037] The specific implementation of this step includes: Step 101, semantically analyze the input multi-source heterogeneous data to extract semantic features in three dimensions: time, space, and decision; Step 102, align and fuse data from different sources according to a unified spatiotemporal semantic model, which is defined as:

[0038] in, Indicates the aligned data. Represents the source data, , and Semantic descriptors representing the time dimension, space dimension, and action dimension respectively. It is a semantic alignment function that uses a multi-head self-attention mechanism to extract and align cross-domain semantic features.

[0039] Works through the following steps: Construct independent attention heads for each dimension (T, S, A); Each attention head calculates the semantic similarity under its dimension; Merge the attention outputs of each dimension to generate a unified semantic representation.

[0040] Step 103: Establish a spatiotemporal semantic index structure to form a spatiotemporal semantic unified database that can be quickly searched. The index structure adopts an improved R-tree structure, which expands the traditional R-tree to support mixed indexing of multi-dimensional spatiotemporal attributes and semantic attributes.

[0041] The unified spatiotemporal semantic data structure output by this step contains the following contents: A collection of data items with a uniform time stamp; A collection of data items with a unified spatial reference; A collection of data items with uniform decision semantics tags; Spatiotemporal semantic indexing supporting multi-dimensional retrieval.

[0042] This step is different from traditional data preprocessing which only focuses on the unification of data formats. This method achieves deep unification of data semantics, especially the semantic expression of time, space and decision-making behavior, laying the foundation for the subsequent construction of cognitive-space-time manifold.

[0043] Step 200, the unified spatiotemporal semantic data structure generates a cognitive-spatiotemporal manifold model through a cognitive spatiotemporal manifold generation algorithm; The cognitive-space-time manifold model is a mathematical representation structure that maps unified space-time semantic data into cognitive space, making the relationship between data conform to human cognitive characteristics and providing a basic representation structure for subsequent human-computer interaction.

[0044] The specific implementation of this step includes: Step 201, define the mathematical representation of cognitive-spacetime manifold. The cognitive-spacetime manifold is defined as:

[0045] in, Represents cognitive-spacetime manifold, Represents the set of vertices on the manifold, corresponding to the key nodes in the data space; Represents the set of edges between vertices, representing the relationship between nodes; Represents the mapping matrix from physical spacetime to cognitive space; Represents a metric tensor on a manifold, which is used to describe the cognitive complexity of different regions on the manifold.

[0046] Step 202, extracting nonlinear manifold structure from unified spatiotemporal semantic data. This process is completed using a cognitive spatiotemporal manifold generation algorithm, which combines the principles of diffusion maps and Riemannian geometry, retaining the advantages of diffusion maps in capturing discrete relationships, and utilizing the powerful ability of Riemannian geometry in expressing continuous nonlinear space, forming a unified framework that can adaptively express complex cognitive-spatial relationships, specifically defined as:

[0047] in, Represents cognitive-spacetime manifold, represents the manifold generating function, Represents unified spatiotemporal semantic data, represents the smoothing parameter, Represents the neighborhood size. This function is implemented by iteratively optimizing the following objective function:

[0048] in, Represents data points and The similarity between represents the regularization term of the manifold, is the balance parameter, and is the index of the data point, where the summation symbol For all data points To sum, , is the total number of data points, represents the norm of a vector, and Represents data points and The mapping coordinates of .

[0049] Step 203, embed the manifold into the visualization space, and optimize the manifold according to human cognitive characteristics, including reducing cognitive distortion, strengthening the visual significance of key decision points, balancing the visibility of local details and global structures, etc.

[0050] The specific implementation steps of the cognitive space-time manifold generation algorithm include: Standardize the unified spatiotemporal semantic data to eliminate the dimensional differences of different dimensions, using the classic mean-standard deviation standardization method; Construct an initial adjacency graph and use the k-nearest neighbor algorithm to connect the k nearest neighbors for each data point; Calculate the weight matrix of the initial adjacency graph. The weights between neighboring points are based on the Gaussian kernel function of the distance between points, and the weights of non-neighboring points are zero. Calculate the similarity matrix: Based on the distance between data points, use the Gaussian kernel function to calculate the similarity; Construct the Laplacian matrix, which is the degree matrix minus the weight matrix:

[0051] in is a diagonal matrix with diagonal elements represents the degree of node i, is the Laplace matrix.

[0052] Solve the generalized eigenvalue problem and obtain the eigenvector corresponding to the minimum non-zero eigenvalue. The calculation formula is as follows:

[0053] in is the mapping matrix, containing the coordinates mapped to the cognitive space, is the characteristic value.

[0054] Use these feature vectors as the initial embedding of the manifold; Introducing cognitive preference tensor , adjust the manifold metric to accommodate human cognitive characteristics:

[0055] in is the metric tensor on the manifold, is the local coordinate transformation matrix, is the cognitive preference tensor, which characterizes the cognitive importance in different directions.

[0056] Cognitive Preference Tensor Constructed by:

[0057] in represents the cognitive dimension priority basis tensor, represents the corresponding weight coefficient, Represents the number of cognitive dimensions. The cognitive dimension priority basis tensor is obtained from cognitive psychology experimental data and represents the differences in human cognitive preferences in processing different types of information.

[0058] Compute the geodesic distance on the manifold based on the epistemic metric tensor as a distance metric in epistemic space:

[0059] in Indicates connection and All possible paths of , represents the tangent vector of the path, Indicates the minimum value, express and Distance in cognitive space.

[0060] The cognitive-space-time manifold model output by this step contains the following contents: Data node set: represents the key data points in the decision space; Relation edge set: represents the semantic relationship between nodes; Cognitive metric tensor: describes the cognitive difficulty of different regions on the manifold; Manifold mapping function: defines the mapping from the original data space to the cognitive space.

[0061] The core framework of the cognitive-spacetime manifold model consists of a graph structure representation system, a Riemann manifold representation system, a graph-manifold mapping framework, a cognitive metric field system, and a spacetime evolution mechanism, including: Graph structure representation system: Discrete relations are represented using standard graph theory components: Vertex set representation: Use heterogeneous graph nodes to represent multiple types of entities. Each vertex is in the form of a tensor, containing feature vectors and attribute metadata.

[0062] Edge set representation: Multiple edge structures are used to express complex relationships, and each edge has a weight matrix and a relationship type tensor.

[0063] Graph topology characteristics: Utilize homology groups, continuous homology and graph analysis structural features, and apply community detection and identification functional sub-modules.

[0064] Riemannian manifold characterization system: Continuous space representation based on standard differential geometry building blocks: Manifold structure definition: Graph embedding and isometric embedding are used to establish a d-dimensional Riemann manifold, and coordinate atlas and transformation functions are used to define the local structure.

[0065] Metric tensor field: Define the Riemann metric tensor, characterize the distance measure on the manifold, and use semi-positive definite matrices to ensure the validity of the metric.

[0066] Geodesic equation: Derive the geodesic differential equation based on the Euler-Lagrange equation and apply the calculus of variations to solve the shortest path.

[0067] Curvature characteristics: Use the Riemann curvature tensor, section curvature and Riemann scalar to analyze the curvature characteristics of manifolds.

[0068] Graph-manifold mapping framework: Combining standard methods of topological data analysis and manifold learning: Embedding mapping: The mapping relationship is established using manifold learning algorithms such as geodesic multidimensional scaling, Laplace eigenmap and diffusion mapping.

[0069] Structure-preserving principle: follow the topological invariance principles such as homeomorphic mapping, isometric mapping and conformal mapping to maintain the original structural characteristics.

[0070] Regularization strategy: Entropy regularization, differential regularization and topological regularization are used to control the mapping quality.

[0071] Cognitive Metric Field System: Based on standard theories from cognitive science and information geometry: Metric field representation: constructing cognitive distance metrics using information metrics, Fisher information matrices, and relative entropy.

[0072] Attention modulation mechanism: Adjusting local metric properties based on cognitive control theory and selective attention principles.

[0073] Multi-scale representation: Apply wavelet transform and scale space theory to realize multi-resolution representation structure.

[0074] Spatiotemporal evolution mechanism: Integrating the theory of dynamical systems and differential equations: Vector field representation: Use the vector field on the manifold to represent the time evolution direction, and use Lie derivatives to describe the rate of change on the manifold.

[0075] Manifold deformation: Computing spatiotemporal deformations based on flow models, parallel transport, and exponential mapping.

[0076] Evolution equations: Application of partial differential equations, level set methods and variational integration to describe the dynamic characteristics of the system.

[0077] This step is different from the traditional low-dimensional embedding method that only focuses on geometric structure preservation. This algorithm also takes into account the laws of human cognition and measures the tensor The concept of "distance" in cognitive space is defined so that data points with strong decision relevance are visually closer, while data points with weak decision relevance are visually farther away, thereby creating a decision space representation that is consistent with human intuition.

[0078] Step 300, obtaining user interaction data and a cognitive-space-time manifold model, and generating an adaptive view expression through an interactive response manifold deformation algorithm; Adaptive view expression refers to a manifold structure that is dynamically adjusted according to the user's cognitive focus. It can reflect the user's interaction intentions and cognitive needs in real time, so that data expression and user cognitive process form a closed-loop interaction.

[0079] The specific implementation of this step includes: Step 301, obtain the user interaction operation and convert it into a manifold deformation operator:

[0080] in, represents the user attention distribution at time t, Represents the decision operation performed by the user, Represents cognitive-spacetime manifold, and They are the attention response operator and the decision response operator respectively. is the manifold deformation operator.

[0081] Step 302, the interactive response manifold deformation algorithm calculates the deformation process of the manifold according to the user's interactive operation. The specific calculation formula is as follows:

[0082] The partial differential equation is based on the application of partial differential equations in physical simulation (such as heat diffusion equation, elastic deformation equation) and implicit Euler method, which is a standard technique for solving ordinary differential equations in numerical analysis; The physical simulation technology is applied to the field of cognitive interaction, and two new operators are defined: : Attention response operator, capturing the impact of user attention on the manifold; : Decision response operator, simulating the impact of user decision behavior on the manifold; The specific forms of these two operators are obtained by fitting a large amount of user interaction experimental data.

[0083] In order to ensure the stability and continuity of the deformation process, the improved implicit Euler method is used for numerical solution:

[0084] in, represents the state of the manifold at time t, express The manifold state at the moment, is the time step.

[0085] Compared with the traditional implicit Euler method, the improved implicit Euler method makes it more suitable for the characteristics of cognitive-space-time manifold by specifically processing the attention response operator and the decision response operator.

[0086] Step 303: perform corresponding manifold deformation operations for different types of user interaction operations: Attention focusing operation: According to the user's attention area, the display ratio and detail level of the corresponding part of the manifold are adjusted so that the area is expanded in the visual space and the details are enhanced, while maintaining the contextual information of the surrounding area; Decision operation: When a user is detected to perform a decision operation, the predicted distribution of the decision results is calculated, and the decision trajectory and possible result distribution map are generated on the manifold; Hypothetical scenario generation: When a user proposes a hypothetical scenario, a corresponding scenario branch structure is created on the manifold, and the decision evolution path under different assumptions is calculated.

[0087] The adaptive view expression output by this step contains the following contents: Deformed manifold structure: a manifold model that reflects the user's cognitive focus; Attention-enhanced regions: high-detail representation of the region in the manifold where the user is paying attention; Decision trajectory: predicted decision outcome path; Scenario branching: Manifold variations under different assumptions.

[0088] In the traditional human-computer interaction process, data expression and user interaction are two separate processes. This step organically integrates the two through the manifold deformation mechanism, so that data expression can be dynamically adjusted according to the user's cognitive state and interaction intention, achieving a true "human-computer dialogue" effect.

[0089] Step 400, obtaining user interaction behavior data and adaptive view expression, and generating a cognitive balance view through a cognitive load-aware projection algorithm; The cognitive balance view is a data presentation method that strikes a balance between complexity and comprehensibility. It dynamically adjusts the dimension and level of detail of data presentation according to the user's current cognitive load status to ensure efficient information transmission.

[0090] The specific implementation of this step includes: Step 401: Analyze the user interaction behavior data to evaluate the current cognitive load level:

[0091] in, represents the calculated cognitive load level value, Represents the user's historical interaction records. represents the user's gaze pattern data, Indicates the user's operation response time data. It is a cognitive load evaluation function, which adopts a weighted fusion method. The specific weight coefficient is learned from a large amount of user experimental data through a machine learning method.

[0092] Step 402: Determine the dimension and method of data projection according to the cognitive load assessment result:

[0093] in, represents the adaptive projection result, represents the original high-dimensional data, represents the cognitive load level value, Represents the current decision context information, is the adaptive projection function.

[0094] The core idea of ​​the cognitive load-aware projection algorithm is to retain more dimensions and details when the cognitive load is low, and to reduce dimensions and simplify expressions when the cognitive load is high, but at the same time ensure that key decision information is not lost. The algorithm automatically selects the appropriate dimensionality reduction method based on the cognitive load value, including: when When using the most information-retaining nonlinear dimensionality reduction method (such as t-SNE (t-distributed stochastic neighbor embedding), UMAP (uniform manifold approximation and projection), Isomap (isometric mapping), etc.); when When using dimensionality reduction methods that balance information retention and computational efficiency (such as MDS (multidimensional scaling analysis), Sammon mapping, etc.); when When using linear dimensionality reduction methods with the simplest calculations (such as PCA (principal component analysis), LDA (linear discriminant analysis), etc.); in Indicates a low threshold of cognitive load, Indicates a high threshold of cognitive load, Represents the cognitive load value.

[0095] Step 403, generate semantic explanation for the projection result to improve interpretability:

[0096] in, Represents the generated projection interpretation text, represents the adaptive projection result, Represents the original data, Represents the current context information. Generate functions for interpretation.

[0097] The function is transformed from mathematical representation to semantic description through the following four steps: Feature attribution analysis: Identify which original features have the greatest impact on the projection results; This step performs dimension importance quantification, and calculates the statistical correlation between the dimensionality reduction result and the original high-dimensional features through the backward feature tracking algorithm. Typical implementations usually involve principal component contribution, global sensitivity analysis, or SHAP value calculation to determine which original features have a dominant influence on the current projected representation. This process is in line with the principle of feature importance attribution in interpretable machine learning.

[0098] Topological structure analysis: detecting intrinsic topological structures in the projected space; This step detects the intrinsic topological structure in the projected space through unsupervised learning methods and divides the observed data into cohesive subgroups. Common implementations include density-based spatial clustering, hierarchical clustering or spectral clustering techniques, and determine the optimal clustering parameter configuration through internal evaluation indicators such as silhouette coefficient and Davies-Bouldin index to achieve adaptive recognition of topological structure.

[0099] Semantic model construction: mapping statistical features into domain semantic expressions; This step realizes the mapping transformation from statistical features to domain semantics. Through the feature-semantic dictionary and natural language generation template, the identified data patterns are converted into semantic expressions that conform to domain terms. This process follows the content planning and surface realization principles in natural language generation (NLG), mapping quantitative statistical features into qualitative semantic descriptions and establishing a bridge between data and domain knowledge.

[0100] Adaptive adjustment of cognitive load: Dynamically adjust the complexity of explanation according to the user's cognitive state In the final stage, the granularity and complexity of semantic interpretation are dynamically adjusted according to the user's cognitive state and decision context. By applying cognitive load theory and relevant principles of information presentation, the system can establish a balance between information completeness and cognitive economy, providing interpretation content with optimal information density for users with different cognitive needs and professional levels.

[0101] The cognitive balance view output by this step includes the following: Adaptive dimensional projection data: dynamically adjust the dimensional data representation according to cognitive load; Key decision-making features highlighting mark: highlight the features that are crucial to the decision; Projection interpretation text: semantic interpretation of the current projection view; Cognitive load metric: A quantitative representation of the current user cognitive state.

[0102] This step is different from the traditional method of improving comprehensibility by simplifying data. This method retains the full complexity of the data, but dynamically adjusts the dimensions and details of the data presentation according to the user's cognitive state, achieving a dynamic balance between complexity and comprehensibility. The system does not preset a fixed data expression form, but dynamically generates the most suitable expression for the current situation based on the user's cognitive needs and abilities.

[0103] Step 500, obtaining the interaction data and cognitive balance view of multiple decision makers, and outputting the collective cognitive structure and cognitive guidance strategy through a multi-agent cognitive fusion algorithm; The collective cognitive structure is a fusion representation of the cognitive models of multiple decision makers, which can capture the complementarity and blind spots of group thinking, while the cognitive guidance strategy is an attention guidance scheme generated for the identified cognitive blind spots.

[0104] The specific implementation of this step is as follows: Step 501, extracting individual cognitive models from the interaction data of each decision maker:

[0105] in, represents the constructed i-th user cognitive model, represents the interaction history data of the i-th user, represents the attention distribution data of the i-th user, represents the decision behavior data of the i-th user, The cognitive model extraction function is a well-known technology based on user modeling and behavior analysis. The innovation lies in the comprehensive analysis of three kinds of data: interaction history, attention distribution and decision-making behavior, to build a complete user cognitive model.

[0106] Step 502: fuse multiple individual cognitive models into a collective cognitive model:

[0107] in, represents the generated collective cognitive model, represents a set of n extracted individual cognitive models, represents the fusion weight parameter, Represents the current collaboration context information, represents the model fusion function, Indicates the starting index of the collection. Represents the total number of decision makers. The model fusion function uses graph neural network (GNN) technology, which regards individual cognitive models as nodes in the graph and defines the complementary or conflicting relationships between nodes through the relationship adjacency matrix to achieve nonlinear cognitive model fusion.

[0108] The multi-agent cognitive fusion algorithm does not perform simple model averaging, but instead implements weighted nonlinear fusion through graph neural networks, which can capture the structural relationship between different cognitive models:

[0109] in, represents a graph neural network function, Represents the adjacency matrix between models, characterizing the complementary or conflicting relationships between different individual cognitive models. Indicates from index arrive A collection of individual cognitive models, represents the total number of decision makers, Represents a collective cognitive model.

[0110] Step 503, based on the collective cognition model, identifying potential blind spots in group cognition:

[0111] in, represents the set of identified cognitive blind spots, represents the collective cognitive model, A data representation that represents the complete decision space, is the detection threshold parameter, represents the blind spot detection function. Cognitive blind spots refer to areas that exist in the decision space but are weakly represented in the collective cognitive model.

[0112] The blind spot detection function is essentially a combination of coverage analysis and anomaly detection. It performs the following steps: Divide the decision space into grid cells; Calculate the “attention density” of the collective cognitive model for each unit; Identify areas where attention density is below threshold but decision importance is high; Label these areas as a collection of cognitive blind spots .

[0113] Step 504: Generate a guidance strategy for the identified cognitive blind spots:

[0114] in, represents the set of cognitive guidance strategies generated, represents the set of cognitive blind spots, represents the collective cognitive model, represents the current decision context, represents the policy generation function.

[0115] The strategy generation function specifically performs the following steps: Select appropriate guidance template from the strategy library according to the current context and blind spot characteristics; Calculate an “attention-guiding priority” for each blind spot; Generate a guidance strategy based on priorities that includes visual cues, interaction suggestions, and explanatory text; Combining multiple local strategies into a cognitive guidance strategy set .

[0116] Common strategies include: visual highlighting, dynamic zooming, interactive prompts, comparison frames, and semantic interpretation.

[0117] The output of this step includes the following: Collective cognitive model: a fusion representation of cognitive patterns of multiple decision makers; Cognitive Blind Spot Map: Identifying critical areas in the decision space that are collectively overlooked; Cognitive guidance strategies: attention guidance schemes generated for cognitive blind spots; Cognitive complementarity index: quantifies the degree of complementarity between cognitive patterns of different decision makers.

[0118] This step is different from the traditional collaborative decision-making method that only focuses on the aggregation of opinions and the resolution of differences. This method deeply analyzes and integrates the cognitive process of decision makers. It can not only integrate the cognitive advantages of multiple decision makers, but also actively identify blind spots in collective thinking and direct the attention of decision makers to these neglected areas through guiding strategies, thereby realizing the excavation and enhancement of collective wisdom.

[0119] Step 600, generating an interactive decision-making scene and a multimodal interactive interface through immersive spatiotemporal scene rendering technology.

[0120] This step takes the output of the previous steps (cognitive-space-time manifold model, adaptive view expression, cognitive balance view, collective cognitive structure and cognitive guidance strategy) as input, converts and processes them through immersive space-time scene rendering technology to generate interactive decision-making scenarios and multimodal interaction interfaces.

[0121] An interactive decision-making scenario refers to a visual environment that can intuitively display the decision space and support user operations and simulations, while a multimodal interactive interface provides a variety of human-computer interaction methods, allowing decision makers to interact naturally with the decision-making scenario.

[0122] The specific implementation of this step is as follows: Step 601, perform spatiotemporal scene mapping to convert the cognitive-spatiotemporal manifold into a visual space representation:

[0123] in, represents visual scene data, Represents the cognitive-space-time manifold model, Represents a set of visual style parameters, Indicates user preference configuration, It is the scene mapping function.

[0124] Step 602: Perform interactive scene rendering based on visual space representation:

[0125] in, represents the generated interactive scene, represents visual scene data, Indicates the interaction parameter configuration. represents the set of physical dynamic parameters, Rendering function for the scene.

[0126] The scene mapping function and the scene rendering function are based on well-known technologies in the field of computer graphics and visualization, including: information visualization mapping technology, physical engine and dynamics simulation and interactive rendering technology.

[0127] Step 603: construct a multimodal interaction interface to implement the following interaction methods: (1) Gesture interaction module: captures user gestures and converts them into scene operation instructions; (2) Voice interaction module: parses user voice commands and executes corresponding scene control and query functions; (3) Collaborative interaction module: synchronizes multi-user operations and maintains a consistent scene state.

[0128] The input of each interaction mode is converted into corresponding manifold deformation operations through a unified interaction-response framework, achieving seamless connection between user intention and data expression.

[0129] Step 604, providing decision path planning and simulation functions:

[0130] in, represents the generated decision path data, represents a sequence of actions specified by the user, Represents the cognitive-space-time manifold model, represents the set of constraints applied, Represents the decision simulation function.

[0131] The decision simulation function combines Monte Carlo simulation and Markov decision process (MDP). The specific execution process is as follows: The user-specified action sequence Mapping to manifold on the trajectory; Consider constraints at each decision point The following possible actions; Use hybrid models (combining deterministic and probabilistic models) to predict decision outcomes: For results under deterministic constraints, causal models are used for calculation; For uncertain factors, Monte Carlo simulation is used to generate possible distributions of outcomes; Generating decision path data Include: Main path (most likely outcome); uncertainty bounds (probability distribution of possible outcomes); Key decision points (nodes with high path sensitivity).

[0132] These paths are visualized as time-evolving trajectories on the manifold, allowing decision makers to intuitively assess the long-term impact of decision sequences.

[0133] The output of this step includes: Interactive decision-making scenarios: operable three-dimensional or multi-dimensional scenario environments; Multimodal interactive interface: a human-computer interaction interface that supports gestures, voice, and collaborative operations; Decision Simulation Engine: A tool component used to calculate and visualize decision results; Multi-user collaboration layer: A collaborative infrastructure that supports the simultaneous participation of multiple decision makers.

[0134] The interactive decision-making scenario in this step has the following characteristics: Dynamic responsiveness: The scene can respond to user operations in real time and provide immediate feedback; Multi-scale navigation: supports seamless switching from macro overview to micro details; Predictive display: proactively predict and visualize the possible results of user actions; Collaborative consistency: Maintain consistency of multi-user views and synchronization of operations.

[0135] This step is different from traditional data visualization technology that only provides static or limited interactive views. This method transforms the abstract decision space into an intuitive space-time scene and supports rich interactive methods, allowing decision makers to perceive, manipulate and plan decisions as intuitively as in the physical world, truly realizing an "immersive decision-making" experience.

[0136] The technical effects of this implementation are as follows: This implementation method achieves a paradigm shift from "passive data display" to "active cognitive guidance" by integrating abstract spatiotemporal data analysis with human cognitive processes into a unified interactive cognitive-spatiotemporal manifold, producing the following significant technical effects: Improved decision-making understanding efficiency: The system transforms complex game and operational optimization problems into intuitive spatiotemporal scenarios, allowing decision makers to quickly understand the essence of the problem. Practical applications show that the time to understand complex game decisions using this system is shortened by 81%, which means that decision makers can master decision problems of the same complexity in less than one-fifth of the time.

[0137] Cognitive load is significantly reduced: Through cognitive adaptive projection technology, the system can dynamically adjust the data presentation method according to the user's cognitive state, effectively reducing cognitive load. Tests show that when processing 64-dimensional high-dimensional data, the user's cognitive load score is reduced by 68%, while maintaining 92% of the data information fidelity.

[0138] Improved team collaboration efficiency: The collective cognitive enhancement mechanism enables team members to share cognitive patterns and complement each other's cognitive blind spots, improving team collaboration efficiency by 156%. In heterogeneous teams (members with different professional backgrounds), this improvement is even more significant, reaching 203%.

[0139] Decision quality has been significantly improved: By visualizing the decision path and result distribution, decision makers can more comprehensively evaluate the impact of decisions, and the quality of decisions has been improved by 43%. This improvement is particularly significant in decision-making scenarios that require balancing short-term and long-term benefits.

[0140] Effective elimination of decision-making blind spots: The system can proactively identify and guide decision makers to focus on neglected decision factors. Experiments show that the system can detect 94% of decision-making blind spots in traditional methods, effectively preventing decision-making bias and risk neglect.

[0141] Breakthrough in high-dimensional complex decision-making capabilities: The system successfully supports the visualization and interaction of complex decision-making problems involving 64 dimensions and 12 time scales, breaking through the dimensional limitations of traditional visualization methods and enabling decision makers to intuitively grasp the ultra-high-dimensional decision space.

[0142] Immersive decision-making experience: The interactive space-time scenarios provided by the system make the decision-making process intuitive and immersive. The user satisfaction score is 187% higher than that of traditional decision support systems, and the willingness to use the system is increased by 212%.

[0143] Accelerated knowledge transfer: By visualizing abstract decision-making problems, the system significantly accelerates the knowledge transfer between domain experts and novices, increasing novices’ understanding of complex decision-making problems by 276% and their decision-making ability by 152%.

[0144] The core breakthrough of this implementation method is to establish a dynamic balance mechanism between data complexity and cognitive comprehensibility, making complex decision-making problems intuitive and controllable without sacrificing the complexity and integrity of the problem. Instead of simplifying data to adapt to cognition, the system dynamically adjusts the data presentation dimensions and details according to cognitive focus through cognitive adaptive projection technology, allowing complexity and comprehensibility to coexist at different attention levels, fundamentally solving the problem of understanding and decision-making of high-dimensional complex data.

[0145] An application example of implementation mode 1 is as follows: Urban traffic management departments face complex traffic flow optimization decision-making challenges, which require collaborative decision-making at multiple time scales (real-time, short-term, and long-term) based on multi-source data (including real-time traffic flow, historical traffic patterns, weather conditions, special event arrangements, etc.). Decision-makers include traffic control center operators, urban planners, and emergency response teams, who need to quickly identify patterns, predict trends, and formulate intervention measures in complex high-dimensional traffic data.

[0146] Traditional traffic management systems usually only provide basic data visualization, lack intuitive display of high-dimensional data, and are difficult to support collaborative analysis and decision-making by multiple decision makers. At the same time, a large amount of traffic data often leads to information overload, making it difficult for decision makers to quickly identify key information and respond.

[0147] This system achieves a dynamic balance between data complexity and cognitive comprehensibility by mapping complex traffic data onto cognitive-space-time manifolds and dynamically adjusting data presentation based on the user's cognitive state, significantly improving the efficiency and quality of traffic decision-making.

[0148] Example input data: The multi-source heterogeneous data processed by the system include: Some examples of real-time traffic flow data are shown in Table 1: Table 1: Real-time traffic flow data (partial sample)

[0149] Some examples of historical traffic pattern data are shown in Table 2: Table 2: Historical traffic pattern data (partial sample)

[0150] Some examples of data related to the environment and events are shown in Table 3: Table 3: Environmental and event data (partial sample)

[0151] Some examples of the traffic intervention database are shown in Table 4: Table 4: Intervention database (partial sample)

[0152] Technical effect verification: In order to verify the technical effect of this implementation method in the urban traffic flow management scenario, we compared the decision-making efficiency and quality performance of the traditional traffic management platform and this system in three representative traffic management scenarios. The verification process adopted a controlled experimental method, with 20 traffic management professionals using the traditional platform and this system to complete the same decision-making task.

[0153] Verification of improved decision-making understanding efficiency: Verification object: Whether this system can significantly improve decision makers' efficiency in understanding complex traffic data.

[0154] The experimental results of decision understanding efficiency comparison are shown in Table 5: Table 5: Comparison results of decision understanding efficiency

[0155] The test results show that compared with traditional systems, this system significantly improves decision makers’ efficiency in understanding complex traffic situations: In the peak-hour traffic hotspot identification task, the number of hotspots that decision makers could identify increased by 126.9%, and the time required for identification decreased by 73.3%; In the analysis of the impact of special events, decision makers’ accuracy in estimating the impact range increased by 32.1%, and the recognition rate of secondary impact areas (such as congestion diffusion areas) increased by 89.3%; In the identification of abnormal traffic patterns, decision makers' anomaly identification accuracy increased by 37.4% and the time to propose response measures decreased by 72.4%.

[0156] These results verify that the system effectively reduces the cognitive burden of understanding complex data through the cognitive-space-time manifold model and the interaction-induced manifold deformation mechanism, enabling decision makers to quickly and accurately grasp the overall picture of traffic conditions and identify key problem points and their causes.

[0157] Verification of improving the efficiency of multi-agent collaborative decision-making: Verification object: Whether this system can significantly improve the efficiency of multiple decision makers in collaboratively solving complex traffic problems.

[0158] The comparison results of multi-agent collaborative decision-making efficiency are shown in Table 6: Table 6: Comparison results of multi-agent collaborative decision-making efficiency

[0159] Experimental results show that this system has significant advantages in multi-agent collaborative decision-making scenarios: In the formulation of traffic control plans for large-scale events, the team using this system saved 72.4% of time and reduced negotiation rounds by 61.4% compared with traditional methods; In the coordinated response to traffic emergencies, the time to formulate response plans was reduced by 76.7%, and the time for cross-departmental coordination was reduced by 78.2%; In the adaptive adjustment of seasonal traffic patterns, the time for multi-department collaborative planning was reduced by 72.0%, while the planning adaptability score increased by 24.1%.

[0160] These results verify that this system effectively integrates the professional perspectives of different decision makers through the collective cognitive enhancement mechanism, identifies and compensates for cognitive blind spots in collective decision-making, and significantly improves the efficiency and quality of team collaborative decision-making. In addition, the sharing mechanism of interactive decision-making scenarios enables all decision makers to collaborate in the same cognitive space, greatly reducing information asymmetry and communication costs.

[0161] Decision quality improvement verification: After the system was actually deployed in the "M City Smart Traffic Management Center", a three-month follow-up evaluation was conducted on the impact on the quality of urban traffic management decisions. By comparing the changes in key performance indicators of traffic management before and after the system deployment, the effect of improving decision-making quality was verified.

[0162] The comparison results of decision quality 3 months before and after system deployment are shown in Table 7: Table 7: Comparison of decision quality (average value 3 months before and after system deployment)

[0163] The actual deployment results verify that this system significantly improves decision-making quality: Traffic congestion management: average vehicle speed on trunk roads during peak hours increased by 22.5%, and cross-regional congestion diffusion events decreased by 63.0%; Traffic incident response: the average incident response time was reduced by 63.0%, and the effectiveness of intervention measures was increased by 22.1%; Resource utilization: The balance of road network capacity utilization increased by 43.1%, and the frequency of traffic light control optimization increased by 195.7%; In terms of user experience: Drivers reported a 35.4% increase in travel time predictability and a 26.6% increase in citizen traffic satisfaction.

[0164] These actual operation data show that this system not only significantly improves the efficiency of decision makers' understanding of complex traffic data and the efficiency of multi-agent collaborative decision-making, but more importantly, these efficiency improvements are transformed into improved quality of actual traffic management. By providing more intuitive and comprehensive decision support, the system enables decision makers to formulate more accurate and effective traffic management strategies, thereby achieving overall optimization of urban traffic flow.

[0165] The application of this implementation method in the urban traffic flow management scenario verifies the following key technical innovations and their effects: Cognitive-spatial unified expression of heterogeneous data: The multi-source heterogeneous traffic data is transformed into a unified cognitive-spatial manifold model, which verifies its significant effect in solving the problems of data fragmentation and multi-source heterogeneous data fusion. In practical applications, the completeness of decision makers' understanding of traffic conditions has increased by 87.3%.

[0166] Interaction-induced manifold deformation mechanism: It realizes the dynamic adaptation of data expression and user cognitive focus, and verifies the effectiveness of highlighting key decision areas while maintaining a global perspective. Experimental results show that the accuracy of data understanding in the decision-maker's focus area has increased by 91.5%, and the ability to grasp the global situation has increased by 63.8%.

[0167] Cognitive load-aware adaptive projection: Dynamically adjusts the data presentation dimension and level of detail based on the user's current cognitive state, verifying its effectiveness in balancing data complexity and comprehensibility. Tests show that users' decision accuracy under high cognitive load conditions increased by 76.2%.

[0168] Collective cognitive enhancement and blind spot compensation: Effectively integrating the professional perspectives of multiple decision makers and identifying and compensating for cognitive blind spots has proven its remarkable effectiveness in improving the efficiency and quality of team collaborative decision-making. In actual applications, cross-departmental collaboration efficiency has increased by 75.6%, and the collective decision-making blind spot recognition rate has increased by 82.3%.

[0169] Immersive spatiotemporal decision-making scenario interaction: It provides an intuitive and immersive decision-making experience, which verifies its effectiveness in reducing the cognitive burden of complex decision-making. Tests show that the cognitive burden of decision-makers in complex scenario analysis was reduced by 68.5%, and decision confidence increased by 56.7%.

[0170] This implementation method applies the "co-creative decision-making scenario system of hyper-dimensional space-time graph" to the urban traffic flow management scenario. Through the combined effect of the above-mentioned technical innovations, it successfully achieves the technical goal of "expressing complex game and operational optimization problems with intuitive space-time scenarios", and has made breakthrough progress in balancing data complexity and cognitive comprehensibility, significantly improving the decision-making efficiency and quality in complex decision-making scenarios.

[0171] The above describes an embodiment of the present invention, but this embodiment is not limited to the above-mentioned specific implementation mode. The above-mentioned specific implementation mode is merely illustrative and not restrictive. Under the guidance of this embodiment, ordinary technicians in this field can also make more forms of equivalent embodiments, all of which are within the protection of this embodiment.

Claims

1. A graph-enhanced intelligent game and operations optimization decision-making assistance method, characterized in that: The following steps are involved: Obtain multi-source heterogeneous data and input them into the spatiotemporal semantic mapping framework to generate a unified spatiotemporal semantic data structure; The unified spatiotemporal semantic data structure generates a cognitive-spatiotemporal manifold model through a cognitive spatiotemporal manifold generation algorithm; Obtain user interaction data and cognitive-space-time manifold model, and generate adaptive view expression through interactive response manifold deformation algorithm; Obtain user interaction behavior data and adaptive view expression, and generate cognitive balance view through cognitive load-aware projection algorithm; Through immersive space-time scene rendering technology, interactive decision-making scenarios and multimodal interaction interfaces are generated.

2. The graph-enhanced intelligent game and operations optimization decision-making auxiliary method according to claim 1 is characterized in that: The spatiotemporal semantic mapping framework adopts a multi-head self-attention mechanism to achieve cross-domain semantic feature extraction and alignment: ； in Indicates the aligned data. Represents the source data, , and Represent the semantic descriptors of time dimension, space dimension and action dimension respectively, is the semantic alignment function.

3. The graph-enhanced intelligent game and operations optimization decision-making auxiliary method according to claim 2 is characterized in that: The interactive response manifold deformation algorithm calculates the manifold deformation process through the following partial differential equation: ； in represents the user attention distribution at time t, Represents the decision operation performed by the user, represents the attention response operator, represents the decision response operator, Represents cognitive-spacetime manifold.

4. The graph-enhanced intelligent game and operations optimization decision-making auxiliary method according to claim 3 is characterized in that: The interactive response manifold deformation algorithm uses an improved implicit Euler method for numerical solution: ； in represents the state of the manifold at time t, express The manifold state at the moment, is the time step.

5. The graph-enhanced intelligent game and operations optimization decision-making auxiliary method according to claim 4 is characterized in that: The cognitive load perception projection algorithm is implemented by the following steps: analyzing the user interaction behavior data to calculate the cognitive load level value: ； in represents the calculated cognitive load level value, Represents the user's historical interaction records. represents the user's gaze pattern data, Indicates the user's operation response time data. Evaluate functions for cognitive load; Determine the dimension and method of data projection according to the cognitive load level value: ； in represents the original high-dimensional data, Represents the current decision context information, represents the adaptive projection result, is the adaptive projection function; Generate semantic explanations for the projection results: ； in Represents the original data, Represents the generated projection interpretation text, Generate functions for interpretation.

6. The graph-enhanced intelligent game and operational optimization decision-making auxiliary method according to claim 5 is characterized in that: The cognitive load-aware projection algorithm automatically selects an appropriate dimensionality reduction method based on the cognitive load value: when When , the nonlinear dimensionality reduction method that retains the most information is used; when When , a dimensionality reduction method that balances information retention and computational efficiency is used; when When , the simplest linear dimensionality reduction method is used; in Indicates a low threshold of cognitive load, Indicates a high threshold of cognitive load, Represents the cognitive load value.

7. The graph-enhanced intelligent game and operations optimization decision-making auxiliary method according to claim 6 is characterized in that: The immersive spatiotemporal scene rendering technology is implemented by the following steps: performing spatiotemporal scene mapping to convert the cognitive balance view into a visual space representation: ； in Represents a set of visual style parameters, Indicates user preference configuration, is the scene mapping function, represents visual scene data, Representation of cognitive-space-time manifold model; Based on the visual space representation, interactive scene rendering is performed to generate an interactive scene: ； in Indicates the interaction parameter configuration, represents the set of physical dynamic parameters, represents the generated interactive scene, Rendering function for the scene.

8. The graph-enhanced intelligent game and operational optimization decision-making auxiliary method according to claim 7 is characterized in that: After generating the cognitive balance view, the method further includes: Obtaining the interaction data of multiple decision makers and the cognitive balance view, and generating a collective cognitive structure and cognitive guidance strategy through a multi-agent cognitive fusion algorithm; The collective cognitive structure and cognitive guidance strategy are provided as additional input to the immersive spatiotemporal scene rendering technology to enhance the collaborative decision-making ability of the interactive decision-making scenario.

9. The graph-enhanced intelligent game and operational optimization decision-making auxiliary method according to claim 8, characterized in that: The multi-agent cognitive fusion algorithm is implemented by the following steps: Extract individual cognitive models from the interaction data of each decision maker: ； in, represents the constructed i-th user cognitive model, represents the interaction history data of the i-th user, represents the attention distribution data of the i-th user, represents the decision behavior data of the i-th user, Extract functions for cognitive models; Multiple individual cognitive models are integrated into a collective cognitive model through the graph neural network algorithm: ； in, represents a graph neural network function, Represents the adjacency matrix between models, characterizing the complementary or conflicting relationships between different individual cognitive models. Indicates from index arrive A collection of individual cognitive models, represents the total number of decision makers, represents a collective cognitive model; Identify potential blind spots in group cognition based on the collective cognition model: ； in, represents the set of identified cognitive blind spots, A data representation representing the complete decision space, is the detection threshold parameter, represents the blind spot detection function; Generate guidance strategies for identified cognitive blind spots: ； in, represents the set of cognitive guidance strategies generated, represents the set of cognitive blind spots, represents the current decision context, represents the policy generation function.

10. A graph-enhanced intelligent game and operations optimization decision-making support system, characterized in that: The method is used to execute the graph-enhanced intelligent game and operational optimization decision-making assistance method described in any one of claims 1 to 9, comprising: The spatiotemporal semantic mapping module is used to convert multi-source heterogeneous data into a unified spatiotemporal semantic data structure; A cognitive manifold generation module, which is used to transform the unified spatiotemporal semantic data structure into a cognitive-spatiotemporal manifold model; The manifold deformation module is used to adjust the cognitive-spatiotemporal manifold model according to user interaction data to generate adaptive view representations; The cognitive load perception module is used to analyze user interaction behavior data, dynamically adjust the dimensions and details of data projection, and generate a cognitive balance view; A spatiotemporal scene rendering module for converting cognitive balance views into interactive decision-making scenarios; Multimodal interaction module, used to provide gesture interaction, voice interaction and collaborative interaction functions; The decision simulation module is used to support users in planning decision paths and performing simulations in scenarios.

Images