Auxiliary decision-making data processing method for grassroots governance based on nonlinear mapping function

Abstract

The present application relates to the technical field of grassroots governance and intelligent decision support system, and more particularly to a grassroots governance auxiliary decision-making data processing method and a data processing platform, which comprises event data collection and preprocessing, recursive call and collaborative optimization based on optimization algorithm, path simulation and optimization through virtual environment and multi-dimensional probability network, and adaptive decision adjustment based on real-time feedback. Through the present application, the system can realize efficient event processing and global optimization, enhance the decision-making ability and response speed in dealing with complex events, and significantly improve the overall efficiency of grassroots governance.

Description

[0001] Cross-reference to related applications

[0002] This application is a divisional application of Chinese patent application No. 2024112757868, filed on September 12, 2024, entitled "Data Processing Method and Data Processing Platform for Grassroots Governance Auxiliary Decision Making". Technical Field

[0003] This invention relates to the technical field of grassroots governance and intelligent decision support systems, and in particular to a data processing method and data processing platform for grassroots governance auxiliary decision-making. Background Technology

[0004] Grassroots governance involves handling numerous daily issues and responding to emergencies. In this process, the efficient collection, analysis, and processing of various event data is crucial to the success of grassroots governance. Traditional manual processing methods are no longer sufficient to meet the increasingly complex needs of grassroots governance, especially in modern cities where event types are diverse and frequent, requiring timely and accurate decision support. Therefore, modern grassroots governance is increasingly relying on intelligent and data-driven technologies to help improve management efficiency and decision-making accuracy.

[0005] Existing grassroots governance technologies (Chinese invention patent, publication number: CN117271847A, title: A method and system for assisting grassroots governance business processing based on a large model) mainly rely on business assistance processing methods based on large models. These technologies construct grassroots governance event models, train the models using labeled event data, and classify and analyze events within the scope of the trained model. While this technology can improve the efficiency of event processing to some extent, it still has some significant shortcomings:

[0006] Existing technologies rely on predefined rules and labels for model training, which makes the system perform poorly when faced with unforeseen complex events. Due to limitations in model training cycles and data labeling, the system's response accuracy will significantly decrease when the event type or data exceeds the scope of model training.

[0007] For security reasons, existing technologies deploy models in intranet environments, limiting the models' ability to access internet content. While this improves data security, it also results in a lack of flexibility and update capability when the system is dealing with events related to external data.

[0008] Existing technologies mainly rely on static rules and model reasoning, lacking the ability to dynamically adjust and optimize, which is particularly inadequate when events develop rapidly or situations are complex and changeable. Summary of the Invention

[0009] To address the numerous problems existing in the prior art, this invention provides a data processing method and platform for grassroots governance auxiliary decision-making. Through multi-level recursive optimization and real-time feedback mechanisms, combined with virtual environment simulation and multi-dimensional probabilistic networks, this invention forms a dynamic and highly flexible decision support system. Its core lies in utilizing advanced algorithms and data processing technologies to achieve intelligent processing of the entire process from event data collection, processing, optimization to execution. Ultimately, this system significantly improves the efficiency and accuracy of decision-making in grassroots governance, ensuring efficient responses to complex events.

[0010] A method for processing data to support grassroots governance decision-making includes the following steps:

[0011] Collect event data, preprocess the data, and classify and prioritize the preprocessed event data according to the event type and urgency to generate high-priority event data;

[0012] High-priority event data is distributed to multiple distributed intelligent nodes. Each distributed intelligent node performs event classification, feature extraction, and preliminary decision-making on the high-priority event data. Preliminary decision data is generated by transforming the data model and using a hybrid inference system. Subsequently, the preliminary decision data is optimized through a non-linear decision feedback mechanism to generate optimized preliminary decision data.

[0013] The initial decision data is distributed to models at multiple levels, and the optimization results of each level model are recursively called and collaboratively optimized through a cross-level recursive optimization algorithm to generate cross-level recursive optimization data.

[0014] Based on cross-level recursive optimization data, a virtual environment is created to simulate event processing paths, and path optimization is performed through a multi-dimensional probabilistic network to finally generate optimal path data.

[0015] Event processing is performed based on the optimal path data. The execution results are analyzed and adaptive adjustments are made. The optimized decision data is then updated to the global knowledge base and synchronized to the distributed intelligent nodes.

[0016] Preferably, the data preprocessing includes noise removal of the event data by using a denoising algorithm to filter environmental noise in the event data; format conversion of the event data to unify data from different sources into a standardized format; and feature extraction of the event data by using a feature extraction algorithm to identify and extract key features.

[0017] Preferably, the classification and priority setting steps are implemented by an algorithm based on event type and urgency, wherein the urgency is set by calculating the potential impact of the event and the processing time requirements, and the generated high-priority event data includes a comprehensive index of time sensitivity and processing urgency.

[0018] Preferably, the transformed data model dynamically adjusts the event data features using the following calculation expression:

[0019]

[0020] in, This represents the transformed feature data; This represents a transformation function related to the event type; Indicates the initial characteristics of the event data; This represents parameters related to the characteristics of the event type.

[0021] Preferably, the hybrid reasoning system includes a rule-based reasoning module and a statistical reasoning module. The rule-based reasoning module performs logical reasoning on structured data based on a predefined set of rules, while the statistical reasoning module performs statistical analysis on unstructured data based on a probability model. The two modules are combined to generate preliminary decision data.

[0022] Preferably, the preliminary decision data is optimized using the following nonlinear mapping function:

[0023]

[0024] in, This represents the optimized preliminary decision data; Represents a nonlinear mapping function; This indicates preliminary decision-making data; This represents the cumulative weighting factor in the feedback loop.

[0025] Preferably, the preliminary decision data is distributed to models at multiple levels, and the optimization results of the multiple-level models are recursively called and collaboratively optimized using the following recursive optimization expression:

[0026]

[0027] in, Indicates the first The data is optimized through recursion. This indicates data optimization for the departmental model; This indicates that the event type model optimizes the data. , and These represent the weight coefficients in the recursive optimization process.

[0028] Preferably, the virtual environment is created by simulation-driven learning based on historical and real-time data, and performs probabilistic inference and path optimization on the simulation results of event processing paths through a multi-dimensional probabilistic network. The path optimization is selected using the following expression:

[0029]

[0030] in, This represents the probability value of the optimal path; Indicates the first The probability of the simulation result for each path.

[0031] Preferably, the event processing based on the optimal path data is performed, the execution result data is collected in real time, the execution result is analyzed through an adaptive algorithm, correction decision data is generated, the correction decision data is input into the global knowledge base, the decision rules and models in the global knowledge base are updated, and the updated global knowledge base data is synchronized to all distributed intelligent nodes.

[0032] A system for implementing the aforementioned grassroots governance auxiliary decision-making data processing method includes:

[0033] The data acquisition module is used to collect event data and perform noise removal, format conversion, and feature extraction on the event data to generate standardized event data.

[0034] The data classification and priority setting module is used to classify and prioritize standardized event data according to event type and urgency, and generate high-priority event data;

[0035] The distributed intelligent node module includes multiple distributed intelligent nodes, which are used to receive high-priority event data, classify the high-priority event data, extract features from the high-priority event data, generate preliminary decision data through transforming the data model and hybrid inference system, and then optimize the preliminary decision data through a nonlinear decision feedback mechanism to generate optimized preliminary decision data.

[0036] The cross-level recursive optimization module is used to distribute the initial optimization decision data to models at multiple levels, and to recursively call and collaboratively optimize the optimization results of models at each level through the cross-level recursive optimization algorithm to generate cross-level recursive optimization data.

[0037] The event cognition and simulation module is used to create a virtual environment based on cross-level recursive optimization data, simulate event processing paths, and optimize the simulation results through a multi-dimensional probabilistic network to generate optimal path data.

[0038] The execution and feedback module is used to perform event processing based on the optimal path data, collect and analyze the execution result data, make adaptive adjustments, generate correction decision data, and update the correction decision data to the global knowledge base;

[0039] The global knowledge base module is used to store and update optimized decision data, and synchronize the updated data to multiple distributed intelligent nodes to ensure that the system can continuously optimize and respond to future event processing needs.

[0040] Compared with the prior art, the advantages and beneficial effects of the present invention are as follows:

[0041] This invention achieves coordination and globally optimal decision-making between models at different levels through a cross-level recursive optimization algorithm. This technology overcomes the limitations of existing models and can flexibly adjust and optimize decision paths when facing complex events.

[0042] This invention achieves dynamic optimization and continuous improvement of decision-making through a real-time feedback mechanism and adaptive algorithm; this enables the system to continuously optimize decisions based on actual execution during event processing, thereby improving the accuracy of decisions and the effectiveness of execution.

[0043] This invention achieves efficient optimization of event processing paths through virtual environment simulation and multidimensional probabilistic networks. This technology not only solves the limitations of data interaction in existing technologies, but also greatly improves the system's ability to perform multidimensional analysis and select the optimal path in complex environments. Attached Figure Description

[0044] Figure 1 This is a schematic flowchart of the method of the present invention;

[0045] Figure 2 This is a schematic diagram of cross-level recursive optimization in this invention;

[0046] Figure 3 This is a schematic diagram of the virtual environment and multidimensional probabilistic network in this invention;

[0047] Figure 4 This is a schematic diagram illustrating the synchronization between the global knowledge base and distributed intelligent nodes in this invention;

[0048] Figure 5 This is a structural block diagram of the data processing platform of the present invention. Detailed Implementation

[0049] The embodiments of the present disclosure will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the disclosure. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of the present disclosure for ease of explanation. However, it will be apparent that one or more embodiments may be practiced without these specific details. Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concepts of the present disclosure.

[0050] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this disclosure. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0051] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0052] like Figure 1 As shown, a data processing method for grassroots governance auxiliary decision-making includes the following steps:

[0053] Collect event data, preprocess the data, and classify and prioritize the preprocessed event data according to the event type and urgency to generate high-priority event data;

[0054] Preferably, the data preprocessing includes noise removal of the event data by using a denoising algorithm to filter environmental noise in the event data; format conversion of the event data to unify data from different sources into a standardized format; and feature extraction of the event data by using a feature extraction algorithm to identify and extract key features.

[0055] Noise removal aims to clean up irrelevant or interfering information in data to ensure its purity. In this invention, noise removal primarily targets environmental noise in event data. This is typically achieved using denoising algorithms that automatically identify and filter out interfering components in the data, such as background noise in surveillance videos or external electromagnetic interference in sensor data. The main effect of noise removal is to improve the signal-to-noise ratio of the data, ensuring that subsequent processing steps are based on high-quality data. For example, when processing audio data rich in environmental noise, using Fourier transform for frequency domain analysis can effectively remove high-frequency noise, retaining only event-relevant audio signals. This processing improves the accuracy of audio data feature extraction, providing a solid foundation for subsequent event classification and decision-making.

[0056] Format conversion aims to unify data from diverse sources into a standardized format. In grassroots governance, data may originate from various heterogeneous sources, such as text, audio, video, and sensor data. The formats and structures of these data vary, and directly using them for decision-making can lead to compatibility issues and low processing efficiency. Through format conversion, the system can transform all data into a standardized format, making it suitable for the needs of subsequent processing modules. For example, sensor data is typically in time-series format, while text data may be natural language descriptions. Format conversion allows these different types of data to be transformed into a unified structured data format, facilitating unified processing and analysis in subsequent steps.

[0057] Feature extraction involves further processing preprocessed data to extract key features. Feature extraction algorithms analyze the internal structure of the data to identify the most representative information for event classification and decision-making. For example, when processing video data, convolutional neural networks (CNNs) can be used to automatically extract object contours and motion trajectories from images. These features help the system quickly identify the nature of events and their potential trends. Through feature extraction, the system can transform massive amounts of raw data into a compact feature set, significantly reducing the complexity of data processing while retaining information crucial for event judgment.

[0058] Preferably, the classification and priority setting steps are implemented by an algorithm based on event type and urgency, wherein the urgency is set by calculating the potential impact of the event and the processing time requirements, and the generated high-priority event data includes a comprehensive index of time sensitivity and processing urgency.

[0059] The core of the classification process lies in categorizing events into different classes based on their characteristics. These categories are typically based on predefined event types, such as security incidents, environmental incidents, and public health incidents. Each event type has its unique processing flow and resource requirements. Therefore, the classification algorithm automatically assigns events to the appropriate category based on relevant characteristics such as text descriptions, geographical locations, and the time of occurrence. For example, when processing large amounts of municipal alarm data, the system can use Natural Language Processing (NLP) technology to extract keywords from event descriptions and combine this with historical data analysis to automatically identify whether it is a security incident or a public safety hazard. This event type-based classification not only helps improve the system's response speed but also ensures the rational allocation of resources, allowing different types of events to receive the most suitable handling.

[0060] After classification, the prioritization step further distinguishes the processing order of events by calculating their urgency. The urgency assessment is primarily based on two core factors: the potential impact of the event and the required processing time. Potential impact assessment involves quantifying the possible consequences of the event, such as casualties, property damage, and social instability. The required processing time, based on the nature of the event and historical data, estimates the timeframe within which a response must be received to avoid adverse consequences. Urgency is typically calculated using a comprehensive algorithm that combines potential impact and processing time requirements to generate a comprehensive urgency index. For example, for a fire alarm, the system considers the potential danger zone (e.g., proximity to residential areas), current weather conditions (e.g., wind speed and direction), and the availability of rescue forces to quickly calculate the event's urgency index. The system then prioritizes events based on this index, handling those with the highest urgency indices first.

[0061] The generated high-priority event data includes a comprehensive indicator of time sensitivity and processing urgency. Time sensitivity refers to how sensitive an event is to changes in external conditions within a specific time window, which directly affects the timeliness of event processing. Processing urgency reflects the potentially serious consequences of untimely event handling. The high-priority event data generated by combining these two indicators ensures that when faced with multiple concurrent events, the system can prioritize processing those events that require the most immediate response, reducing the negative impact of delays.

[0062] In one embodiment, a grassroots governance system is used in response to emergencies. For example, in an extreme weather warning system, the system receives a large number of alarm messages, including flooding, fallen trees, and power outages. The system first classifies these events into different types based on the descriptions of the alarm messages, and then calculates the urgency level of each event. For example, a flooding event might be assigned a higher urgency index due to its potential for traffic paralysis and the risk of people being trapped, while a power outage event's urgency index is set based on the population density and the importance of power supply to the affected area. Ultimately, the system prioritizes the flooding event, dispatching emergency rescue forces to the scene quickly, while the power department addresses the power outage issue later.

[0063] High-priority event data is distributed to multiple distributed intelligent nodes. Each distributed intelligent node performs event classification, feature extraction, and preliminary decision-making on the high-priority event data. Preliminary decision data is generated by transforming the data model and using a hybrid inference system. Subsequently, the preliminary decision data is optimized through a non-linear decision feedback mechanism to generate optimized preliminary decision data.

[0064] The allocation of distributed intelligent nodes is the starting point of this step. High-priority event data is first distributed to multiple distributed intelligent nodes, which can be located in different physical locations or logically represent different processing units in the system. The advantage of distributed processing is that it can improve the parallel processing capability of the system, avoid overloading a single node, and thus improve the overall processing speed and fault tolerance of the system. In a practical application scenario, such as a city's emergency response system, distributed intelligent nodes may be distributed in different areas of the city. Each node is responsible for processing high-priority event data within its area. In this way, when multiple emergency events occur simultaneously, the system can rationally allocate tasks among the nodes to ensure a rapid response.

[0065] On each distributed intelligent node, event classification and feature extraction are performed first. The classification step further refines the events, potentially reclassifying them based on more specific local conditions or more detailed event descriptions. Feature extraction analyzes the event data at a higher dimension, extracting the most useful features for subsequent decision-making. The feature extraction process may involve various techniques, such as deep learning-based feature extraction algorithms, which can automatically identify key patterns and relationships from event data. For example, when processing surveillance video data, intelligent nodes can use convolutional neural networks (CNNs) to extract human action features from events, thereby better determining the nature of the event and its potential future development.

[0066] After classification and feature extraction, the intelligent node enters the preliminary decision-making stage. Preliminary decision-making, based on the feature data extracted in the previous steps, utilizes a transformed data model and a hybrid reasoning system to make initial judgments and response suggestions for the event. The transformed data model refers to dynamically adjusting the model's parameters according to the characteristics of different types of events to adapt to different event types and situations. This dynamic adjustment makes the model more flexible and adaptable, thereby improving the accuracy of the preliminary decision. The hybrid reasoning system combines the advantages of rule-based reasoning and statistical reasoning. The former can perform logical judgments based on a predefined set of rules, while the latter performs uncertainty analysis based on historical data and probabilistic models. This hybrid reasoning approach enables the system to quickly process events with high certainty and also make relatively reliable preliminary judgments when facing uncertainty.

[0067] After generating preliminary decision data, the system optimizes this data through a nonlinear decision feedback mechanism. The principle of this mechanism is to dynamically adjust the weighting factors in the feedback loop, combining the deviation between the actual feedback and the model output to correct and optimize the preliminary decision data. This step uses a series of nonlinear mapping functions to fine-tune the decision path, reducing errors and improving decision accuracy. For example, in an emergency evacuation, the preliminary decision might suggest a certain evacuation route, but feedback analysis reveals that the actual evacuation time for that route exceeds expectations. The system will then readjust the weighting factors through the nonlinear decision feedback mechanism to optimize a new, more effective evacuation route.

[0068] In one embodiment, within a complex emergency management system, when a high-priority fire alarm is triggered, the system rapidly distributes the alarm data to several intelligent nodes near the fire's location. These nodes further classify and extract features based on information such as the fire's specific location, time, and weather conditions, including the speed of fire spread and the density of surrounding buildings. Subsequently, the system combines these features, dynamically adjusts the parameters of the decision model by transforming the data model, and provides preliminary decision suggestions through a hybrid inference system, such as formulating a preliminary evacuation plan. Then, based on on-site feedback, such as actual evacuation time and personnel flow, the system adjusts the evacuation plan in real time through a non-linear decision feedback mechanism to ensure the safety and efficiency of the final evacuation routes.

[0069] Preferably, the transformed data model dynamically adjusts the event data features using the following calculation expression:

[0070]

[0071] in, This represents the transformed feature data; This represents a transformation function related to the event type; Indicates the initial characteristics of the event data; This represents parameters related to the characteristics of the event type.

[0072] Transformation data models enhance the adaptability of decision-making models to different types of events by dynamically adjusting data features. First, the initial features of the event data... These are the basic information extracted from the raw data. These features typically contain basic descriptions of the event, such as time, location, and number of participants. However, simply relying on these initial features is insufficient to accurately handle different types of complex events. Therefore, a transformation function is introduced. It weights and adjusts the initial features according to the specific type of event (such as natural disasters, social security incidents, public health incidents, etc.), so that these features show higher relevance and importance in the handling of different types of events.

[0073] Transformation function The design of the transformation function is typically based on historical data and expert knowledge. It can automatically adjust during the decision-making process. For example, for a fire incident, the transformation function may amplify characteristic data related to fire spread (such as wind speed and humidity), while for a traffic accident, the transformation function may focus more on characteristics such as road conditions and traffic flow. Through such dynamic adjustments, the system can flexibly adjust the data input according to different event types, ensuring that when processing each event, the system receives the most relevant and decision-making-valuable data.

[0074] In this model, parameters It acts as a correction factor, used to fine-tune the transformed feature data. Parameters These parameters are typically related to the specific characteristics of an event and can be further refined based on the uniqueness of the event. For example, in a particular event, in addition to the usual characteristics such as the number of infections and the transmission route, It can also reflect the distribution of local medical resources or the public's awareness of protection, and these characteristics can have a significant impact on the final decision.

[0075] The introduction of a transformed data model significantly improves the system's flexibility and accuracy. By dynamically adjusting the initial features, the system can generate more accurate feature data based on different event types. This improves the decision-making model's ability to handle different events. This flexibility ensures that the system maintains efficient and accurate response capabilities when faced with complex and ever-changing events.

[0076] In one embodiment, an emergency response scenario is presented within an urban management system. Assume a multi-hazard concurrency situation where the system needs to handle different types of events simultaneously, such as earthquakes, fires, and floods. For earthquake events, a data model is transformed. This might increase the weighting of geological features, such as earthquake intensity and focal depth, while for fire events, the system might focus more on features related to fire propagation, such as wind speed and building density. Through dynamic adjustments, the characteristic data of these events... This will better meet actual needs, ensuring that the system can base its analysis and decision-making on the most relevant data, and ultimately formulate the most effective response measures.

[0077] Preferably, the hybrid reasoning system includes a rule-based reasoning module and a statistical reasoning module. The rule-based reasoning module performs logical reasoning on structured data based on a predefined set of rules, while the statistical reasoning module performs statistical analysis on unstructured data based on a probability model. The two modules are combined to generate preliminary decision data.

[0078] The rule-based reasoning module is a component of the hybrid reasoning system, responsible for performing logical reasoning on structured data. Structured data typically includes formatted and organized information, such as records in a database, data read by sensors, or predefined event information. The rule-based reasoning module performs reasoning operations on this data based on a predefined set of rules, usually developed by domain experts based on years of experience and historical data, and contains the processing logic for specific events. For example, in emergency management, rules for fire incidents might include: if the fire spread rate exceeds a certain threshold, priority should be given to evacuating people from the affected area. This rule-based reasoning approach can quickly arrive at clear decisions, and is particularly suitable for event handling scenarios with well-defined logic and clear conditions.

[0079] Unlike the rule-based reasoning module, the statistical reasoning module primarily handles unstructured data, such as text, images, and videos. This type of data is typically complex and difficult to process using simple rules. The statistical reasoning module uses probabilistic models to perform statistical analysis on unstructured data. These models can learn from extensive historical data to infer the probability or trend of certain events. For example, when processing social media data, the system can use the statistical reasoning module to analyze public reaction trends to an event, determining its urgency or potential trajectory. The advantage of statistical reasoning lies in its ability to handle data with high uncertainty, providing a basis for decision-making through probabilistic analysis, enabling the system to make reasonable judgments even in the face of ambiguity or the unknown.

[0080] The core of the hybrid reasoning system lies in combining the strengths of these two modules: precise rule-based reasoning for structured data and comprehensive statistical analysis for unstructured data, thereby generating more comprehensive preliminary decision-making data. This combination ensures the system's flexibility and accuracy when processing different types of data. For example, in an emergency response scenario, when the system receives an alarm message about a potential public safety incident, the rule-based reasoning module can quickly determine whether immediate action is needed based on existing rules, while the statistical reasoning module can further analyze relevant social media data to assess public attention and the speed of dissemination of the incident. By combining the results of these two reasoning methods, the system can arrive at a more comprehensive preliminary decision, such as whether to immediately initiate an emergency response or conduct further observation first.

[0081] In one embodiment, this describes a scenario involving the handling of major public events in urban management. For example, an urban management system receives an alarm about a subway accident. The rule-based reasoning module can quickly derive a preliminary response plan based on structured data about the accident, such as subway train number, number of affected passengers, and location of the accident, such as closing relevant lines and evacuating passengers. Simultaneously, the statistical reasoning module can assess the social impact and spread of the event by analyzing related discussions on social media. If the statistical analysis shows that the event has attracted significant attention in a short period and elicits a strong public reaction, the system may adjust its initial decision and activate a higher-level emergency response to prevent further escalation of the incident.

[0082] Preferably, the preliminary decision data is optimized using the following nonlinear mapping function:

[0083]

[0084] in, This represents the optimized preliminary decision data; Represents a nonlinear mapping function; This indicates preliminary decision-making data; This represents the cumulative weighting factor in the feedback loop.

[0085] In this invention, the introduction of a nonlinear decision feedback mechanism plays a crucial role in improving the accuracy and response speed of decision-making. This mechanism optimizes the initial decision data through a nonlinear mapping function, ensuring that the final decision better adapts to actual conditions and environmental changes.

[0086] In this mechanism, preliminary decision data This initial decision is generated based on the system's earlier decision-making module and includes preliminary response suggestions for the current event. However, in actual execution, this initial decision is often influenced by various dynamic factors, such as changes in the external environment, real-time resource availability, and further development of the event. Therefore, relying solely on this initial decision data may not be sufficient to handle complex and ever-changing real-world scenarios. In this case, a nonlinear mapping function... It played a key role.

[0087] Nonlinear mapping function A nonlinear mapping function is a complex mathematical model used to perform a nonlinear transformation on input data (i.e., the sum of preliminary decision data and cumulative weighting factors). The purpose of this nonlinear transformation is to adjust decisions based on the nonlinear relationships within the data, making them more aligned with actual needs. Compared to linear models, nonlinear mapping functions can handle more complex data relationships and more accurately reflect changes in the real world. For example, in dealing with a public health emergency, preliminary decision data might be based on a linear prediction of the current number of infections, but the actual spread of infection often exhibits nonlinear characteristics (such as exponential growth). Through a nonlinear mapping function, the system can more accurately adjust the prediction of infection spread and optimize resource allocation and response strategies.

[0088] Cumulative weighting factor This is the feedback component in the mechanism. The cumulative weighting factor in the feedback loop represents the empirical data accumulated by the system through real-time monitoring and analysis during the decision-making process. This empirical data may include previous decision results, environmental changes, and the actual development of events. Over time, the cumulative weighting factor gradually reflects the importance of this feedback information, thus playing a role in correcting and optimizing the nonlinear mapping function. For example, in a city emergency evacuation plan, the initial decision may be based on standard traffic flow data, but as the actual evacuation proceeds, the system will continuously collect actual traffic data through the feedback loop and adjust the cumulative weighting factor accordingly. This allows for real-time optimization of decisions, ultimately leading to a more reasonable and effective evacuation plan.

[0089] Through a nonlinear decision feedback mechanism, the system can perform real-time decision optimization when facing complex and dynamic environments. This process ensures that the system not only relies on initial static decisions but also dynamically adjusts based on feedback during actual operation. Ultimately, this mechanism can greatly improve the system's response speed and decision accuracy, avoiding inefficient or even erroneous decisions caused by discrepancies between initial and actual conditions.

[0090] In one embodiment, this is applied in an urban flood control emergency management system. When the system receives a flood warning, it generates preliminary decision data. This may include suggested evacuation areas and required resource allocation. However, the initial decisions may need to be adjusted as the flood actually develops. Through a nonlinear decision feedback mechanism, the system can monitor flood level changes, wind speed, rainfall, and other factors in real time, and continuously adjust the cumulative weighting factors through the feedback loop. At this point, the nonlinear mapping function The system will re-optimize evacuation decisions based on this latest data, such as adjusting evacuation routes and increasing or decreasing resource allocation. Ultimately, the system will use this optimized decision data... Guide flood control and emergency response actions to ensure that cities can respond most effectively to sudden floods.

[0091] The initial decision data is distributed to models at multiple levels, and the optimization results of each level model are recursively called and collaboratively optimized through a cross-level recursive optimization algorithm to generate cross-level recursive optimization data.

[0092] Preferably, the preliminary decision data is distributed to models at multiple levels, and the optimization results of the multiple-level models are recursively called and collaboratively optimized using the following recursive optimization expression:

[0093]

[0094] in, Indicates the first The data is optimized through recursion. This indicates data optimization for the departmental model; This indicates that the event type model optimizes the data. , and These represent the weight coefficients in the recursive optimization process.

[0095] Cross-level recursive optimization is a highly complex and crucial step used to integrate and optimize decision models across multiple levels. The main objective of this step is to enable models at different levels to coordinate with each other through recursive calls and collaborative optimization, thereby generating more accurate and globally optimal decision data.

[0096] Cross-level recursive optimization algorithms improve the overall decision-making performance by distributing initial decision data across multiple levels of models and performing recursive calls and collaborative optimization among these models. Figure 2 As shown, the "level" here can be understood as different decision dimensions or modules in the system, such as departmental models and event type models. Each level model is optimized independently and generates its corresponding optimization data. However, relying solely on the optimization results of a single level may lead to local optima rather than global optima. Therefore, cross-level recursive optimization algorithms repeatedly adjust and optimize the results of each level by recursively calling different level models to ensure global consistency and optimality of the decision.

[0097] exist middle, As the core data for recursive calls, it is continuously adjusted and optimized in each round of recursion. Departmental model optimization data. Optimize data with event type model These represent the output results of different levels of the model in the current optimization round. These results are weighted by coefficients. , and Combined to generate new recursive optimization data The weighting coefficients may be dynamically adjusted during the recursive process to accommodate the importance and influence of different levels of the model. For example, in some cases, the results of the departmental model may be more important than those of the event type model, and the system will adjust the weighting coefficients accordingly. and To highlight the influence of the departmental model.

[0098] The main advantage of cross-level recursive optimization lies in its ability to optimize decisions globally, ensuring that models at different levels can work collaboratively and thus avoiding the limitations of local optima. For example, in grassroots governance, one department might focus on optimizing resource allocation, while another focuses on event classification and prioritization. Through cross-level recursive optimization, these decisions at different levels can be coordinated to ultimately generate a globally optimal decision that comprehensively considers effective resource utilization, time urgency, and event impact.

[0099] In terms of effectiveness, cross-level recursive optimization ensures global consistency and overall optimization in decision-making, which is particularly important when dealing with complex events. The system can not only optimize at the local level but also integrate the optimization results from various levels into a more refined and comprehensive decision solution through recursive calls. This recursive optimization process improves the system's decision-making efficiency and accuracy, especially when facing complex events with multiple dimensions and variables, significantly enhancing the system's responsiveness and decision-making effectiveness.

[0100] In one embodiment, a comprehensive emergency management system needs to coordinate multiple departments (such as fire, medical, and transportation) and event types (such as fire, flood, and earthquake) to respond to large-scale natural disasters. Preliminary decision-making data. After optimization, the results are distributed to models at various levels, such as the resource allocation model for the fire department and the evacuation model for the traffic management department. Each department optimizes independently, generating optimization results. and Through a cross-level recursive optimization algorithm, the system gradually adjusts these results, ensuring that decisions across all departments are coordinated and that resources and actions are optimally allocated globally. As the recursive calls proceed, the system ultimately generates a comprehensive response plan, ensuring efficient collaboration among departments and minimizing disaster losses.

[0101] like Figure 3 As shown, based on cross-level recursive optimization data, a virtual environment is created to simulate event processing paths, and path optimization is performed through a multi-dimensional probabilistic network to finally generate optimal path data.

[0102] Preferably, the virtual environment is created by simulation-driven learning based on historical and real-time data, and performs probabilistic inference and path optimization on the simulation results of event processing paths through a multi-dimensional probabilistic network. The path optimization is selected using the following expression:

[0103]

[0104] in, This represents the probability value of the optimal path; Indicates the first The probability of the simulation result for each path.

[0105] In this invention, event processing path simulation and path optimization based on cross-level recursive optimization data are key steps used by the system to determine the optimal action plan. This process involves creating a virtual environment for simulation and combining it with a multi-dimensional probabilistic network to infer and optimize event processing paths, thereby generating optimal path data to support the execution of decisions.

[0106] This process first relies on cross-level recursive optimization data, generated by the recursive optimization algorithm in the previous step, representing the optimal combination of outputs from different levels of models. Based on this data, the system creates a virtual environment. This virtual environment is not merely a simple simulation space; it is a dynamic, digitally-based scenario grounded in real-world conditions. This virtual environment learns by combining historical and real-time data. Specifically, it extracts patterns from the processing history of past events and, in conjunction with currently acquired real-time data, simulates the possible future development paths of events.

[0107] Simulation-driven learning in virtual environments is achieved through advanced machine learning algorithms. These algorithms continuously adjust environmental parameters based on data input to more accurately simulate various possible development scenarios of an event. For example, when simulating a flood event, the virtual environment combines historical flood data (such as water levels, rainfall, and terrain features) with current real-time meteorological data to predict the possible development path and impact range of the flood. This simulation not only provides predictability of the event's possible outcomes but also offers rich data support for subsequent path optimization.

[0108] The role of multidimensional probabilistic networks is crucial in this process. A multidimensional probabilistic network is a complex mathematical model used to calculate and infer the success probability of different event handling paths. Simulation results of multiple event handling paths generated in a virtual environment are input into the multidimensional probabilistic network. The system performs probabilistic analysis based on the characteristics of these paths, thereby evaluating the potential success rate of each path in actual execution. The core of path optimization lies in selecting the optimal path—that is, the path with the highest probability of success—to ensure that the final decision maximizes the achievement of the expected goals.

[0109] The meaning is to select the path with the highest probability value from all simulation results as the optimal path. This optimization process fully utilizes the advantages of probabilistic analysis. By evaluating the feasibility and potential success rate of different paths under actual conditions, the system can select the path most likely to succeed. For example, in urban emergency evacuation decision-making, the system may generate multiple evacuation route simulation results. Path optimization will select the route with the lowest probability of obstruction and the fastest travel speed as the final evacuation route to maximize the safety of personnel.

[0110] By employing path optimization based on cross-level recursive data optimization, virtual environment simulation, and multi-dimensional probabilistic networks, the system significantly improves the accuracy and operability of decision-making. It not only provides a systematic approach to predicting and selecting optimal course of action but also addresses complex and ever-changing real-world conditions, ensuring a high success rate in decision execution.

[0111] In one embodiment, an emergency response scenario occurs during a large-scale public emergency. Assuming a city faces an emergency, the system needs to formulate the optimal evacuation plan. First, the system generates simulations of multiple evacuation routes based on cross-level recursive optimization data. These simulations take into account historical experience in handling similar events, as well as currently collected environmental and population flow data. Subsequently, the system inputs these simulation results into a multi-dimensional probabilistic network for probabilistic path analysis. Finally, the system selects the evacuation route with the highest success rate through an optimization expression. This route is then pushed to the emergency response team as the optimal decision data, ensuring that the evacuation operation can be carried out quickly and efficiently, minimizing casualties.

[0112] Event processing is performed based on the optimal path data. The execution results are analyzed and adaptive adjustments are made. The optimized decision data is then updated to the global knowledge base and synchronized to the distributed intelligent nodes.

[0113] like Figure 4 As shown, preferably, the event processing based on the optimal path data is performed, the execution result data is collected in real time, the execution result is analyzed through an adaptive algorithm, correction decision data is generated, the correction decision data is input into the global knowledge base, the decision rules and models in the global knowledge base are updated, and the updated global knowledge base data is synchronized to all distributed intelligent nodes.

[0114] Within the framework of this invention, performing event processing based on optimal path data, along with real-time analysis and adaptive adjustments, are crucial steps to ensure continuous system optimization in actual operation. The core of this process lies in utilizing the optimal path data generated through prior optimization to execute specific event processing, while simultaneously monitoring and analyzing the execution results in real time, dynamically adjusting decisions, and updating the optimized decisions to the global knowledge base to ensure the system can continuously learn and improve.

[0115] After generating optimal path data, the system first uses this data to guide the specific execution of event handling. This step means that the system puts the previously derived optimal decision-making plan into practice, such as evacuating personnel according to the optimized route during an emergency evacuation. The execution process in this stage involves several real-time variables, such as personnel movement speed, environmental changes, and resource consumption. These variables may deviate from the initial expectations, therefore, the system needs to monitor them in real time.

[0116] Real-time collection of execution results data is a crucial part of this process. The system acquires various types of data generated during event handling in real time through sensors, monitoring equipment, and communication systems. For example, during personnel evacuation, the system collects information such as crowd density, traffic flow, and evacuation speed in real time. This data provides a foundation for subsequent analysis and decision-making adjustments.

[0117] The system analyzes the real-time collected execution result data using an adaptive algorithm. The core of this algorithm is its ability to dynamically adjust decision parameters based on real-time data, thereby optimizing the current decision path. The algorithm adjusts weights and corrects the model based on the difference between the actual and expected results, generating corrective decision data. For example, if the system detects that the actual transit time for a certain evacuation route is longer than expected, possibly due to sudden traffic congestion, the adaptive algorithm will automatically analyze the cause and suggest changing evacuation routes or adjusting evacuation speeds to ensure the smooth execution of the overall evacuation plan.

[0118] After generating corrective decision data, the system inputs this data into the global knowledge base. The global knowledge base is the system's core data storage and knowledge management platform; it records all historical decisions, model parameters, rule sets, and related optimization data. By inputting the latest corrective decision data into the global knowledge base, the system can continuously enrich and improve its knowledge system, enabling it to better utilize past experience data when facing similar events, thereby improving the quality and efficiency of decision-making.

[0119] Furthermore, synchronizing the updated global knowledge base data to all distributed intelligent nodes is a crucial step in ensuring the collaborative operation of the entire system. Distributed intelligent nodes rely on the latest data and models in the global knowledge base when executing tasks. Therefore, after each knowledge base update, the system synchronizes these updates to all intelligent nodes, enabling them to operate based on the latest optimized decisions when handling similar events subsequently. This synchronization process ensures the overall consistency and coordination of the system, avoiding decision-making errors or inefficiencies caused by data inconsistencies between different nodes.

[0120] Through this process, the system can collect and analyze data in real time during actual event handling, making adaptive adjustments to ensure the accuracy and effectiveness of decisions. This continuous feedback and adjustment mechanism not only improves the success rate of current event handling but also continuously optimizes the overall performance of the system, enabling it to cope with increasingly complex and changing realities in long-term operation.

[0121] In one embodiment, an urban traffic management system is used in response to sudden natural disasters. Assuming an earthquake occurs, the system generates optimal evacuation routes and uses this route data to guide traffic evacuation in the city. As the evacuation progresses, the system monitors traffic flow, congestion, and average vehicle speed on each road in real time. When the system detects severe congestion on a main road due to a traffic accident, an adaptive algorithm immediately analyzes and generates corrective decision data, suggesting replanning evacuation routes or diverting some traffic. Subsequently, this corrective data is input into a global knowledge base and synchronized to other distributed intelligent nodes, enabling the entire traffic system to adjust its strategies quickly to ensure the smooth implementation of the evacuation plan and the safety of personnel.

[0122] like Figure 5 As shown, a data processing platform for implementing the grassroots governance auxiliary decision-making data processing method includes:

[0123] The data acquisition module collects event data and performs noise removal, format conversion, and feature extraction to generate standardized event data. This module is responsible for collecting event data from multiple sources and performing preliminary processing on this data. First, noise removal technology filters out interfering information to ensure data purity. Next, format conversion technology unifies data from different sources into a standardized format for easier subsequent processing. Finally, feature extraction algorithms identify and extract key features to generate standardized event data. Through these steps, the system ensures the quality and consistency of the input data, laying the foundation for subsequent decision-making.

[0124] The data classification and prioritization module categorizes and prioritizes standardized event data based on event type and urgency, generating high-priority event data. This module classifies standardized event data and assigns priorities based on event type and urgency. The classification process relies on predefined rules and algorithms to categorize events into different types, such as natural disasters and social security incidents. Prioritization generates high-priority event data by calculating the potential impact of events and processing time requirements. In this way, the system can prioritize events with the greatest impact on grassroots governance, ensuring the rational allocation and use of resources.

[0125] The distributed intelligent node module comprises multiple distributed intelligent nodes that receive high-priority event data, classify and extract features from it, and generate preliminary decision data through a transformed data model and a hybrid inference system. Subsequently, a nonlinear decision feedback mechanism optimizes the preliminary decision data, generating optimized preliminary decision data. The core of this module lies in its distributed processing capability, enabling rapid response and preliminary decision-making even under large-scale, multi-event concurrency.

[0126] The cross-level recursive optimization module distributes the initial decision data to models at multiple levels and recursively calls and collaboratively optimizes the optimization results of each level of the model using a cross-level recursive optimization algorithm, generating cross-level recursive optimized data. This module distributes the optimized initial decision data to models at multiple levels through a recursive optimization algorithm, and performs recursive calls and collaborative optimization among these models. Ultimately, it generates cross-level recursive optimized data. The key to this process lies in the collaborative work of the models at each level, ensuring the global optimality of the decision and avoiding the limitations of local optima.

[0127] The event cognition and simulation module creates a virtual environment based on cross-level recursive optimization data to simulate event processing paths. It then optimizes the simulation results using a multi-dimensional probabilistic network to generate optimal path data. This module combines historical and real-time data to accurately predict event development trends and selects the processing path with the highest success rate through probabilistic inference, providing strong support for final decision-making.

[0128] The execution and feedback module is used to perform event processing based on optimal path data, collect and analyze execution result data, make adaptive adjustments, generate corrective decision data, and update the corrective decision data to the global knowledge base. After obtaining optimal path data, the system performs event processing based on this data and collects execution result data in real time. Through adaptive algorithms, the system analyzes the execution results, generates corrective decision data, and updates this data to the global knowledge base. The core of this module lies in real-time feedback and adaptive adjustment, ensuring that the system can dynamically optimize decisions during event processing, improving decision accuracy and response speed.

[0129] The global knowledge base module stores and updates optimized decision data, synchronizing the updated data to multiple distributed intelligent nodes to ensure the system can continuously optimize and respond to future event handling needs. This synchronization mechanism enables continuous system optimization, ensuring that each node uses the latest decision models and rules when dealing with future events, thereby improving the system's overall responsiveness and decision quality.

[0130] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects.

[0131] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for base layer governance aided decision data processing based on a non-linear mapping function, characterized in that, The method comprises the following steps: collecting event data, preprocessing the data, and classifying and prioritizing the preprocessed event data according to event types and emergency levels to generate high-priority event data; allocating the high-priority event data to multiple distributed intelligent nodes, each of which classifies events, extracts features, and makes preliminary decisions on the high-priority event data, generates preliminary decision data through a transformed data model and a hybrid inference system, and then optimizes the preliminary decision data through a nonlinear decision feedback mechanism to generate optimized preliminary decision data; the transformed data model is used to dynamically adjust the parameters of the model according to the characteristics of different types of events to adapt to different event types and situations; the nonlinear decision feedback mechanism adjusts the weight factor in the feedback loop dynamically, combines the deviation between the actual feedback and the model output, and corrects and optimizes the preliminary decision data; distributing the optimized preliminary decision data to multiple levels of models and recursively calling and cooperatively optimizing the optimization results of each level of model through a cross-level recursive optimization algorithm to generate cross-level recursive optimization data; the multiple levels of models include department models and event type models, representing different decision dimensions or modules; the cross-level recursive optimization algorithm recursively calls between different levels of models to repeatedly adjust and optimize the results of each level; The optimization results of the multiple levels of models are recursively called and cooperatively optimized by the following recursive optimization expression: ; wherein, denotes the recursively optimized data; denotes the department model optimized data; denotes the event type model optimized data; , and denote the weight coefficients in the recursive optimization process; based on the cross-level recursive optimization data, a virtual environment is created for simulation of the event handling path, and the simulation result of the event handling path is probabilistically inferred and path optimized through a multi-dimensional probability network, to finally generate optimal path data; the path optimization is performed through the following expression: wherein, a probability value representing an optimal path; a probability value representing a simulation result of the i-th path; based on the optimal path data, performing event processing, collecting execution result data in real time, analyzing the execution result through an adaptive algorithm, generating correction decision data, inputting the correction decision data into a global knowledge base, updating decision rules and models in the global knowledge base, and synchronizing the updated global knowledge base data to all distributed intelligent nodes; the transformed data model dynamically adjusts the event data features through the following calculation expression: wherein, represents transformed feature data; represents a transformation function associated with the event type; represents initial features of the event data; represents parameters associated with the event type features; the preliminary decision data is optimized through the following nonlinear mapping function: wherein, represents the optimized preliminary decision data; represents a non-linear mapping function; represents the preliminary decision data; represents a cumulative weight factor in the feedback loop.

2. The base layer governance aided decision-making data processing method based on a nonlinear mapping function according to claim 1, characterized in that, The data preprocessing includes noise removal of event data, filtering of environmental noise in event data using a denoising algorithm; format conversion of event data to unify data from different sources into a standardized format; feature extraction of event data to identify and extract key features through a feature extraction algorithm.

3. The base layer governance aided decision data processing method based on the nonlinear mapping function according to claim 1, characterized in that, The classification and priority setting step is based on an algorithm for event types and emergency levels, wherein the emergency level is set by calculating the potential impact and processing time requirement of the event, and the generated high-priority event data contains a comprehensive index of time sensitivity and processing urgency.

4. The base layer governance aided decision-making data processing method based on a nonlinear mapping function according to claim 1, characterized in that, The hybrid inference system includes a rule-based reasoning module and a statistical reasoning module, the rule-based reasoning module performs logical reasoning on structured data according to a predefined rule set, and the statistical reasoning module performs statistical analysis on unstructured data based on a probability model, and the two generate preliminary decision data in combination.

Images