An e-government big data processing system and method

Abstract

This invention belongs to the field of e-government big data processing technology. It discloses an e-government big data processing system and method. The system includes: a traffic data acquisition module, a deep data acquisition module, a data interaction extraction module, a traffic quantitative prediction module, an auxiliary decision generation module, a causal reasoning explanation module, a disadvantaged group bias module, and a data differentiation and processing module. The system analyzes feature vectors to obtain congestion prediction values, analyzes these prediction values, and generates auxiliary decision reports based on the analysis results. It then performs causal inference on the auxiliary decision reports and, based on the inference results and the auxiliary decision reports, generates a causal reasoning report. Finally, it calculates congestion prediction values ​​based on regional type data and complaint assessment data, and corrects these values ​​to obtain revised prediction values. Overall, this invention has significant advantages such as high accuracy in traffic condition prediction, strong data mining capabilities, and a significant role in balancing management.

Description

Technical Field

[0001] This invention relates to the field of e-government big data processing technology, and more specifically, to an e-government big data processing system and method. Background Technology

[0002] E-government refers to a new management model in which state organs comprehensively apply modern information technology, network technology, and office automation technology to conduct office work, management, and provide public services to society in their government activities. In a broad sense, e-government should include all state institutions, while in a narrow sense, it mainly includes administrative organs at all levels that directly undertake the management of national public affairs and social affairs. Thanks to the rapid development of technology, the electronic processing of government affairs has been widely applied. Among them, the smoothness of urban traffic is not only related to the operation of the city, but also closely related to people's travel and life.

[0003] Patent application CN117892849B discloses a method and system for optimizing e-government big data processing. Composed of a permission tagging module, a conflict alignment module, a dataset fusion module, a permission matching module, and a distributed authorization module, it can implement any of the e-government big data processing optimization methods described in this invention. The internal structure of the system collaborates with each other. By mapping different user permissions to corresponding regional levels of departments in a large amount of data within the e-government system, it reduces the difficulty of permission searches for a large amount of business data from corresponding departments in the subsequent authorization matching and distributed storage of sensitive data. Furthermore, it applies targeted encryption measures of different sensitivities to the data and performs adaptive processing before distributed data processing, reducing the time spent on decrypting and encrypting large amounts of sensitive data in e-government big data, thereby simplifying the operation process of the e-government big data processing optimization system.

[0004] However, while the aforementioned e-government big data processing optimization methods and systems have reduced the time required for decrypting and processing large amounts of sensitive data in e-government big data to some extent by setting up multiple modules, thus simplifying the operation process of the e-government big data processing optimization system, e-government processing covers a wide range of areas. Among them, transportation, as the core artery of a city, directly affects the operation of the city and people's travel and life. Therefore, how to effectively address traffic pressure through e-government big data processing systems has become a major problem that needs to be solved in the current e-government processing direction.

[0005] In view of this, the present invention proposes an e-government big data processing system and method to solve the above problems. Summary of the Invention

[0006] To overcome the aforementioned deficiencies of the prior art and to achieve the above objectives, the present invention provides the following technical solution, including:

[0007] The traffic data acquisition module is used to collect traffic datasets, which include traffic flow data, average vehicle speed data, road occupancy data, and event occurrence data.

[0008] The deep data acquisition module is used to collect deep datasets, which include policy impact data, remote sensing congestion data, and complaint assessment data.

[0009] The data interaction extraction module is used to preprocess traffic datasets and deep datasets, and extract feature data to obtain feature vectors;

[0010] Furthermore, the steps of preprocessing the traffic dataset and deep dataset, and extracting feature data include:

[0011] Q1: Clean the data by removing outliers and normalize all sub-data items in the basic dataset to the range of [0, 1] according to the normalization formula;

[0012] Q2: Congestion risk is calculated based on average vehicle speed data and road occupancy data. The specific formula for the calculation is as follows: ;

[0013] Obtain congestion risk characteristic data ,in, This is average vehicle speed data. For the maximum possible speed, This refers to road occupancy data;

[0014] Q3: The potential impact is calculated using traffic flow data and event occurrence data. The specific formula for the calculation is as follows: ;

[0015] Obtain potential impact characteristic data ,in, For traffic flow data, Data on the occurrence of the event;

[0016] Q4: By combining policy impact data and complaint assessment data, social factor characteristics are extracted. The specific calculation formula for the extraction is as follows: ;

[0017] Obtain social factor characteristic data ,in, For policy impact data, For complaint assessment data;

[0018] Q5: Calculate remote sensing congestion data Subtract the absolute value of the event occurrence data to obtain the verification feature data. ;

[0019] Q6: The specific formula for calculating traffic pressure based on congestion risk characteristic data, social factor characteristic data, and remote sensing congestion data is as follows: ;

[0020] Obtain traffic pressure characteristic data ;

[0021] Q7: Package congestion risk characteristic data, potential impact characteristic data, social factor characteristic data, validation characteristic data, and traffic pressure characteristic data to obtain a feature vector;

[0022] The traffic quantification prediction module is used to analyze feature vectors to obtain congestion prediction values;

[0023] Furthermore, the steps for analyzing the eigenvectors include:

[0024] Step 1: Obtain a set of historical feature vectors stored in the database, and compare them with the current time based on the timestamp. Group the comparison results from smallest to largest and label them accordingly. The labeling results are L1, L2, L3, ..., Ln. Use the labeling results as the sample set.

[0025] Step 2: Divide the sample set into 70% training set, 15% test set and 15% validation set, and build the first traffic congestion prediction model based on the sample set.

[0026] Step 3: Input the historical feature vector and the feature vector into the traffic congestion prediction model for calculation. The specific formula for calculation is as follows: ;

[0027] Obtain the first congestion prediction value ,in, To integrate the number of models, It is the Sigmoid activation function. The number of decision trees in a single ensemble model. For the first The dynamic weighting factors of each decision tree. For the first In the group integration model, the first Gradient boosting decision trees For feature vectors, As a weighting factor for the attention mechanism, For multi-head self-attention functions, For the first Historical feature vectors of the ensemble model;

[0028] Step 4: Calculate the error value based on the first congestion prediction value from Step 3. The specific formula for calculation is as follows: ;

[0029] Obtain the error value ,in, For the sample size, This represents the actual congestion level.

[0030] Step 5: Based on the error value in Step 4, judge according to the error threshold. When the error value is less than or equal to the error threshold, output the traffic congestion prediction model. When the error value is greater than the error threshold, return to Step 3 to retrain the first traffic congestion prediction model.

[0031] Step 6: Input the feature vector into the traffic congestion prediction model and output the congestion prediction value;

[0032] Step 7: Output the congestion prediction values ​​to the auxiliary decision generation module;

[0033] The auxiliary decision generation module is used to analyze congestion prediction values ​​and generate auxiliary decision reports based on the analysis results.

[0034] Furthermore, methods for analyzing congestion forecasts and generating decision support reports based on the analysis results include:

[0035] Based on traffic threshold intervals (R1, R2);

[0036] When the congestion prediction value is less than R1, a congestion report and emergency instructions are generated; when the congestion prediction value is greater than or equal to R1 and less than R2, a medium report and optimization instructions are generated; when the congestion prediction value is greater than or equal to R2, a normal report is generated.

[0037] A normal report includes a statement indicating that traffic conditions are predicted to be good, and staff should monitor the situation according to the pre-set workflow.

[0038] The medium-level report includes optimization instructions and states that the predicted traffic conditions are moderate, and staff are requested to monitor whether the traffic congestion situation escalates.

[0039] The congestion report includes emergency instructions and a description of the predicted traffic congestion, requesting staff to immediately arrange for management personnel to go to the scene to direct traffic;

[0040] The optimization instruction contains a set of characters representing the trigger for semaphore optimization;

[0041] Emergency commands contain a set of characters representing the generation of the best route to the congestion site;

[0042] Package the normal report, moderate report, and congestion report to obtain a decision support report;

[0043] The causal reasoning explanation module is used to make causal inferences on the auxiliary decision-making report and, based on the inference results and the auxiliary decision-making report, to obtain a causal reasoning report.

[0044] Furthermore, methods for making causal inferences about decision support reports and combining the inference results with the decision support reports include:

[0045] When the auxiliary decision-making report is a moderate report or a congestion report, the causal reasoning mechanism is triggered;

[0046] The causal inference mechanism is based on a causal inference rule base. It compares the input feature vector, traffic dataset, and deep dataset, and outputs the inference result.

[0047] The inference results are integrated into the corresponding auxiliary decision-making report to obtain a causal reasoning report;

[0048] The vulnerable group tilt module is used to calculate based on regional type data and complaint assessment data, and to correct the congestion prediction value according to the calculation results to obtain the predicted correction value;

[0049] Furthermore, methods for calculating congestion predictions based on regional type data and complaint assessment data, and then revising the congestion predictions based on the calculation results, include:

[0050] Substitute the regional type data and complaint assessment data into the calculation formula: ;

[0051] Obtain the predicted correction value ,in, This is the congestion forecast value. For correction factor, Complaint assessment data for remote areas, Complaint assessment data for the central area;

[0052] when When the value is less than 0.8, the predicted correction value will be returned to the auxiliary decision generation module to replace the congestion prediction value;

[0053] The data differentiation and processing module is used to store key datasets in the database, display causal inference reports through a visualization panel, identify the causal inference reports, and process them based on the identification results.

[0054] Furthermore, the methods for identifying causal reasoning reports and processing them based on the identification results include:

[0055] When the causal reasoning report is rated as medium, an optimization instruction will be sent to the intelligent controller;

[0056] When the causal reasoning report is a congestion report, the emergency instructions are sent to the intelligent route generation tool to obtain the emergency route, and the emergency route is output to the staff receiving terminal;

[0057] Key datasets include traffic datasets, deep datasets, feature vectors, decision support reports, causal inference reports, and predicted correction values;

[0058] Furthermore, S1: Collect traffic datasets, which include traffic flow data, average vehicle speed data, road occupancy data, and event occurrence data;

[0059] S2: Collect deep datasets, which include policy impact data, remote sensing congestion data, and complaint assessment data;

[0060] S3: Preprocess the traffic dataset and deep dataset, and extract feature data to obtain feature vectors;

[0061] S4: Analyze the feature vectors to obtain congestion prediction values;

[0062] S5: Analyze the congestion forecast values ​​and generate a decision support report based on the analysis results;

[0063] S6: Make causal inferences on the auxiliary decision-making report, and obtain a causal reasoning report based on the inference results and the auxiliary decision-making report;

[0064] S7: Calculate based on regional type data and complaint assessment data, and correct the congestion prediction value according to the calculation results to obtain the predicted correction value;

[0065] S8: Store key datasets in the database, display causal reasoning reports through a visualization panel, identify the causal reasoning reports, and process them based on the identification results.

[0066] The technical effects and advantages of the e-government big data processing system and method of the present invention are as follows:

[0067] This invention collects traffic datasets, including traffic flow data, average vehicle speed data, road occupancy data, and event occurrence data. It also collects deep datasets, including policy impact data, remote sensing congestion data, and complaint assessment data. The traffic and deep datasets are preprocessed, and feature data is extracted to obtain feature vectors. These feature vectors are analyzed to obtain congestion prediction values. The analysis of these prediction values ​​generates an auxiliary decision-making report. Causal inference is performed on the auxiliary decision-making report, and based on the inference results and the auxiliary decision-making report, a causal reasoning report is obtained. Calculations are performed based on regional type data and complaint assessment data, and the congestion prediction values ​​are corrected based on the calculation results to obtain revised prediction values. Key datasets are stored in a database, and the causal reasoning report is displayed through a visualization panel for identification. Based on the identification results, the system processes the data to quantify and predict traffic pressure using multi-source data, providing valuable core data support for staff to address traffic congestion. Furthermore, by further mining the congestion prediction values, the invention provides corresponding decision-making suggestions, significantly reducing the workload of staff and effectively lowering the time cost of decision-making. Simultaneously, through causal reasoning explanations of congestion, the invention provides staff with intuitive insights into the main causes of user situations, assisting them in resolving congestion as quickly as possible. Moreover, by prioritizing data from peripheral areas, the invention effectively reduces policy discrimination caused by traditional "data bias," thereby effectively balancing management differences between central and non-central areas. Overall, the invention has significant advantages such as high accuracy in traffic condition prediction, strong data mining capabilities, and a significant role in balancing management. Attached Figure Description

[0068] Figure 1 This is a schematic diagram of an e-government big data processing system according to the present invention;

[0069] Figure 2 This is a schematic diagram of an e-government big data processing method according to the present invention. Detailed Implementation

[0070] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0071] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “a,” “the,” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0072] Depending on the context, the words “if” or “suppose” as used here can be interpreted as “when” or “in response to determination” or “in response to detection.” Similarly, depending on the context, the phrases “if determination” or “if detection (of the stated condition or event)” can be interpreted as “when determination” or “in response to determination” or “when detection (of the stated condition or event)” or “in response to detection (of the stated condition or event).”

[0073] Furthermore, the timing of the steps in the following method embodiments is merely an example and not a strict limitation.

[0074] In practice, the server-side equipment deployed in an e-government big data processing system may consist of one or more devices. This e-government big data processing system can be implemented as: a business instance, a virtual machine, or hardware devices. For example, the e-government big data processing system can be implemented as a business instance deployed on one or more devices in a cloud node. Simply put, this e-government big data processing system can be understood as software deployed on a cloud node, used to provide e-government big data processing services to various user terminals. Alternatively, the e-government big data processing system can also be implemented as a virtual machine deployed on one or more devices in a cloud node. This virtual machine contains application software for managing various user terminals. Or, the e-government big data processing system can also be implemented as a server composed of numerous identical or different types of hardware devices, with one or more hardware devices configured to provide e-government big data processing services to various user terminals.

[0075] In terms of implementation, the e-government big data processing system and the user terminal are mutually compatible. That is, if the e-government big data processing system is implemented as an application installed on a cloud service platform, then the user terminal is implemented as a client that establishes a communication connection with the application; or if the e-government big data processing system is implemented as a website, then the user terminal is implemented as a webpage; or if the e-government big data processing system is implemented as a cloud service platform, then the user terminal is implemented as a mini-program in an instant messaging application.

[0076] like Figure 1 The figure shown is a system architecture diagram of an e-government big data processing system provided in an embodiment of the present invention.

[0077] The e-government big data processing system described in this invention can be hosted on a cloud server. In terms of implementation, it can function as one or more service devices, or as an application installed in the cloud (e.g., a mobile service operator's server, server cluster, etc.), or it can be developed as a website. Depending on the functions implemented, the e-government big data processing system may include a traffic data acquisition module, a deep data acquisition module, a data interaction extraction module, a traffic quantitative prediction module, an auxiliary decision generation module, a causal reasoning and explanation module, a disadvantaged group support module, and a data differentiation and processing module. The modules described in this invention can also be referred to as units, which are a series of computer program segments that can be executed by the processor of an electronic device and perform a fixed function, stored in the memory of the electronic device.

[0078] In this embodiment of the invention, each of the above-mentioned modules in the e-government big data processing system can be implemented independently and can call other modules. Here, "calling" can be understood as a module connecting to multiple modules of another type and providing corresponding services to those connected modules. For example, the sharing and evaluation module can call the same information collection module to obtain information collected by that module. Based on the above characteristics, in the e-government big data processing system provided by this embodiment of the invention, without modifying the program code, the applicable scope of the e-government big data processing system architecture can be adjusted by adding modules and directly calling them, achieving cluster-based horizontal expansion to achieve the goal of quickly and flexibly expanding the e-government big data processing system. In practical applications, the above-mentioned modules can be set in the same device or different devices, or they can be set in virtual devices, such as service instances in a cloud server. Example 1

[0079] Please see Figure 1 As shown in this embodiment, an e-government big data processing system includes:

[0080] The traffic data acquisition module is used to collect traffic datasets, which include traffic flow data, average vehicle speed data, road occupancy data, and event occurrence data.

[0081] It needs to be explained that the system uses inductive loop sensors to collect the number of vehicles passing through a designated area per unit time to obtain traffic flow data; microwave radar speedometers to collect the average vehicle speed within a designated area to obtain average vehicle speed data; video detection cameras to collect the road space occupancy rate within a designated area to obtain road occupancy rate data; and AI event detection cameras to collect abnormal events within a designated area to obtain event occurrence data. Abnormal events refer to situations such as vehicle accidents or construction barriers, but when an abnormal event is detected, the output value of the event data transmission is 1.

[0082] The deep data acquisition module is used to collect deep datasets, which include policy impact data, remote sensing congestion data, and complaint assessment data.

[0083] It needs to be explained that, through the government database API interface, policy text sentiment values ​​are collected to obtain policy impact data. Policy text sentiment values ​​refer to the analysis of keywords using the BERT model. For example, "restricted behavior" is negative, while "travel subsidies" is positive. Negative values ​​are assigned a value of 0, and positive values ​​are assigned a value of 1. Through drone remote sensing equipment, visible light images of congestion density in a designated area are collected, and vehicle pixels are segmented to calculate the congestion ratio, resulting in remote sensing congestion data. The congestion ratio is assigned a value: when the congestion ratio reflects smooth traffic, the remote sensing congestion data is assigned a value of 0; when the congestion ratio reflects congestion, the remote sensing congestion data is assigned a value of 1. Through an intelligent voice analysis platform, the emotional tendencies of citizens' voices in hotline recordings are collected, and scores are output based on ASR-to-text tools and LSTM sentiment analysis models to obtain complaint assessment data. The score scale is -1 to 1, where -1 is negative and 1 is positive.

[0084] The data interaction extraction module is used to preprocess the traffic dataset and the deep dataset, and to extract feature data to obtain feature vectors.

[0085] Further steps, including preprocessing the traffic dataset and deep dataset, and extracting feature data, include:

[0086] Q1: Clean the data by removing outliers and normalize all sub-data items in the basic dataset to the range of [0, 1] according to the normalization formula;

[0087] It should be explained that removing outliers refers to, for example, negative road occupancy data; the base dataset includes traffic datasets and deep datasets; the specific expression of the normalization formula is: ,in Normalized value Any sub-data item of the basic data, This represents the historical maximum value of any given sub-data item. This represents the historical minimum value of any given sub-data item; normalization is used to eliminate the dimensions of all sub-data items in the base dataset.

[0088] Q2: Congestion risk is calculated based on average vehicle speed data and road occupancy data. The specific formula for the calculation is as follows: ;

[0089] Obtain congestion risk characteristic data ,in, This is average vehicle speed data. For the maximum possible speed, This refers to road occupancy data;

[0090] Q3: The potential impact is calculated using traffic flow data and event occurrence data. The specific formula for the calculation is as follows: ;

[0091] Obtain potential impact characteristic data ,in, For traffic flow data, Data on the occurrence of the event;

[0092] Q4: By combining policy impact data and complaint assessment data, social factor characteristics are extracted. The specific calculation formula for the extraction is as follows: ;

[0093] Obtain social factor characteristic data ,in, For policy impact data, For complaint assessment data;

[0094] Q5: Calculate remote sensing congestion data Subtract the absolute value of the event occurrence data to obtain the verification feature data. ;

[0095] Q6: The specific formula for calculating traffic pressure based on congestion risk characteristic data, social factor characteristic data, and remote sensing congestion data is as follows: ;

[0096] Obtain traffic pressure characteristic data ;

[0097] Q7: Package congestion risk characteristic data, potential impact characteristic data, social factor characteristic data, validation characteristic data, and traffic pressure characteristic data to obtain a feature vector;

[0098] It should be explained that the sub-data items in the basic data involved in steps Q2 to Q6 and all subsequent modules are data values ​​that have been cleaned and normalized.

[0099] The traffic quantification prediction module is used to analyze the feature vectors to obtain congestion prediction values;

[0100] Further steps in analyzing the eigenvectors include:

[0101] Step 1: Obtain a set of historical feature vectors stored in the database, and compare them with the current time based on the timestamp. Group the comparison results from smallest to largest and label them accordingly. The labeling results are L1, L2, L3, ..., Ln. Use the labeling results as the sample set.

[0102] Step 2: Divide the sample set into 70% training set, 15% test set and 15% validation set, and build the first traffic congestion prediction model based on the sample set.

[0103] Step 3: Input the historical feature vector and the feature vector into the traffic congestion prediction model for calculation. The specific formula for calculation is as follows: ;

[0104] Obtain the first congestion prediction value ,in, To integrate the number of models, It is the Sigmoid activation function. The number of decision trees in a single ensemble model. For the first The dynamic weighting factors of each decision tree. For the first In the group integration model, the first Gradient boosting decision trees For feature vectors, As a weighting factor for the attention mechanism, For multi-head self-attention functions, For the first Historical feature vectors of the ensemble model;

[0105] It should be explained that the Sigmoid activation function is used to constrain the output value within the immediately following brackets to the range (0,1);

[0106] Step 4: Calculate the error value based on the first congestion prediction value from Step 3. The specific formula for calculation is as follows: ;

[0107] Obtain the error value ,in, For the sample size, This represents the actual congestion level.

[0108] Step 5: Based on the error value in Step 4, judge according to the error threshold. When the error value is less than or equal to the error threshold, output the traffic congestion prediction model. When the error value is greater than the error threshold, return to Step 3 to retrain the first traffic congestion prediction model.

[0109] It should be explained that the error threshold was determined manually and then input into the system;

[0110] Step 6: Input the feature vector into the traffic congestion prediction model and output the congestion prediction value;

[0111] Step 7: Output the congestion prediction values ​​to the auxiliary decision generation module;

[0112] The auxiliary decision generation module is used to analyze the congestion prediction values ​​and generate an auxiliary decision report based on the analysis results.

[0113] Furthermore, methods for analyzing congestion forecasts and generating decision support reports based on the analysis results include:

[0114] Based on traffic threshold intervals (R1, R2);

[0115] It should be explained that the traffic threshold range is determined manually and then input into the system;

[0116] When the congestion prediction value is less than R1, a congestion report and emergency instructions are generated; when the congestion prediction value is greater than or equal to R1 and less than R2, a medium report and optimization instructions are generated; when the congestion prediction value is greater than or equal to R2, a normal report is generated.

[0117] A normal report includes a statement indicating that traffic conditions are predicted to be good, and staff should monitor the situation according to the pre-set workflow.

[0118] The medium-level report includes optimization instructions and states that the predicted traffic conditions are moderate, and staff are requested to monitor whether the traffic congestion situation escalates.

[0119] The congestion report includes emergency instructions and a description of the predicted traffic congestion, requesting staff to immediately arrange for management personnel to go to the scene to direct traffic;

[0120] The optimization instruction contains a set of characters representing the trigger for semaphore optimization;

[0121] Emergency commands contain a set of characters representing the generation of the best route to the congestion site;

[0122] Package the normal report, moderate report, and congestion report to obtain a decision support report;

[0123] The causal reasoning explanation module is used to make causal inferences on the auxiliary decision-making report and obtain a causal reasoning report based on the inference results and the auxiliary decision-making report.

[0124] Furthermore, methods for making causal inferences about decision support reports and combining the inference results with the decision support reports include:

[0125] When the auxiliary decision-making report is a moderate report or a congestion report, the causal reasoning mechanism is triggered;

[0126] The causal inference mechanism is based on a causal inference rule base. It compares the input feature vector, traffic dataset, and deep dataset, and outputs the inference result.

[0127] It should be explained that the causal reasoning rule base is a preset data comparison table. For example, when the congestion risk feature data is greater than 0.6 and the event occurrence data is 1, the inference result output is that the traffic accident event caused the congestion.

[0128] The inference results are integrated into the corresponding auxiliary decision-making report to obtain a causal reasoning report;

[0129] The disadvantaged group tilt module is used to calculate based on regional type data and complaint assessment data, and correct the congestion prediction value according to the calculation results to obtain the prediction correction value.

[0130] Furthermore, methods for calculating congestion predictions based on regional type data and complaint assessment data, and then revising the predictions based on the calculation results, include:

[0131] Substitute the regional type data and complaint assessment data into the calculation formula: ;

[0132] Obtain the predicted correction value ,in, This is the congestion forecast value. For correction factor, Complaint assessment data for remote areas, Complaint assessment data for the central area;

[0133] It should be explained that the region type data is used to divide the data collection area into urban central areas and remote urban areas;

[0134] when When the value is less than 0.8, the predicted correction value will be returned to the auxiliary decision generation module to replace the congestion prediction value;

[0135] The data differentiation and processing module is used to store key datasets in a database, display causal reasoning reports through a visualization panel, identify the causal reasoning reports, and process them based on the identification results.

[0136] Furthermore, methods for identifying causal reasoning reports and processing the results include:

[0137] When the causal reasoning report is rated as medium, an optimization instruction will be sent to the intelligent controller;

[0138] When the causal reasoning report is a congestion report, the emergency instructions are sent to the intelligent route generation tool to obtain the emergency route, and the emergency route is output to the staff receiving terminal;

[0139] Key datasets include traffic datasets, deep datasets, feature vectors, decision support reports, causal inference reports, and predicted correction values;

[0140] In this embodiment, the beneficial effects are achieved by collecting a traffic dataset, including traffic flow data, average vehicle speed data, road occupancy data, and event occurrence data, and a deep dataset, including policy impact data, remote sensing congestion data, and complaint assessment data. The traffic and deep datasets are preprocessed, and feature data is extracted to obtain feature vectors. These feature vectors are analyzed to obtain congestion prediction values. The congestion prediction values ​​are then analyzed, and an auxiliary decision-making report is generated based on the analysis results. Causal inference is performed on the auxiliary decision-making report, and a causal reasoning report is obtained based on the inference results and the auxiliary decision-making report. Calculations are performed based on regional type data and complaint assessment data, and the congestion prediction values ​​are corrected based on the calculation results to obtain revised prediction values. The key datasets are stored in a database, and the causal reasoning report is displayed through a visualization panel. By identifying and processing the identification results, the system can quantify and predict traffic pressure based on multi-source data, providing valuable core data support for staff to solve traffic congestion. In addition, this invention provides corresponding decision-making suggestions through further in-depth mining of congestion prediction values, thereby greatly reducing the workload of staff and effectively reducing the time cost required for decision-making. At the same time, through causal reasoning explanations of congestion, it can intuitively provide staff with the main reasons for the occurrence of user situations, assisting staff to resolve congestion situations as quickly as possible. Furthermore, by tilting data towards peripheral areas, it effectively reduces policy discrimination caused by traditional "data bias," thereby effectively balancing the management differences between central and non-central areas. Overall, this invention has significant advantages such as high accuracy in traffic condition prediction, strong data mining capabilities, and a significant role in balancing management. Example 2

[0141] Please see Figure 2 As shown, the parts not described in detail in this embodiment are described in Embodiment 1. An e-government big data processing method is provided, the method includes: S1: collecting traffic dataset, the traffic dataset including traffic flow data, average vehicle speed data, road occupancy data and event occurrence data;

[0142] S2: Collect deep datasets, which include policy impact data, remote sensing congestion data, and complaint assessment data;

[0143] S3: Preprocess the traffic dataset and deep dataset, and extract feature data to obtain feature vectors;

[0144] S4: Analyze the feature vectors to obtain congestion prediction values;

[0145] S5: Analyze the congestion forecast values ​​and generate a decision support report based on the analysis results;

[0146] S6: Make causal inferences on the auxiliary decision-making report, and obtain a causal reasoning report based on the inference results and the auxiliary decision-making report;

[0147] S7: Calculate based on regional type data and complaint assessment data, and correct the congestion prediction value according to the calculation results to obtain the predicted correction value;

[0148] S8: Store key datasets in the database, display causal reasoning reports through a visualization panel, identify the causal reasoning reports, and process them based on the identification results. Example 3

[0149] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0150] Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be embraced within the invention. No appended diagram markings in the claims should be construed as limiting the scope of the claims.

[0151] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0152] Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices recited in a system claim may also be implemented by a single unit or device through software or hardware. The terms "first," "second," etc., are used to indicate names and do not indicate any specific order.

[0153] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An e-government big data processing system, characterized in that, The system includes: a data interaction and extraction module, a traffic quantitative prediction module, an auxiliary decision generation module, a causal reasoning and explanation module, and a module that prioritizes disadvantaged groups, wherein: The data interaction extraction module is used to preprocess the traffic dataset and the deep dataset, and to extract feature data to obtain feature vectors. The steps for preprocessing traffic and deep datasets and extracting features include: Q1: Clean the data by removing outliers and normalize all sub-data items in the basic dataset to the range of [0, 1] according to the normalization formula; Q2: Congestion risk is calculated based on average vehicle speed data and road occupancy data. The specific formula for the calculation is as follows: ; Obtain congestion risk characteristic data ,in, This is average vehicle speed data. For the maximum possible speed, This refers to road occupancy data; Q3: The potential impact is calculated using traffic flow data and event occurrence data. The specific formula for the calculation is as follows: ; Obtain potential impact characteristic data ,in, For traffic flow data, Data on the occurrence of the event; The AI ​​event detection camera collects abnormal events within a designated area to obtain event occurrence data. Abnormal events include vehicle accidents and construction fence situations. When an abnormal event is detected, the output value of the event sending data is 1. Q4: By combining policy impact data and complaint assessment data, social factor characteristics are extracted. The specific calculation formula for the extraction is as follows: ; Obtain social factor characteristic data ,in, For policy impact data, For complaint assessment data; Through the government database API interface, policy text sentiment values ​​are collected to obtain policy impact data. The policy text sentiment value is obtained by analyzing keywords using the BERT model. When the keyword is related to traffic restrictions, it is marked as a negative keyword and the policy impact data value is 0. When the keyword is related to travel subsidies, it is marked as a positive keyword and the policy impact data value is 1. Through the intelligent voice analysis platform, the sentiment tendency of citizens' voices in hotline recordings is collected, and the scores are output based on the ASR to text tool and the LSTM sentiment analysis model to obtain complaint assessment data. The scale of the score is -1 to 1. Q5: Calculate remote sensing congestion data Subtract the absolute value of the event occurrence data to obtain the verification feature data. ; Q6: The specific formula for calculating traffic pressure based on congestion risk characteristic data, social factor characteristic data, and remote sensing congestion data is as follows: ; Obtain traffic pressure characteristic data ; Q7: Package congestion risk characteristic data, potential impact characteristic data, social factor characteristic data, validation characteristic data, and traffic pressure characteristic data to obtain a feature vector; The traffic quantification prediction module is used to analyze the feature vectors to obtain congestion prediction values; The auxiliary decision generation module is used to analyze the congestion prediction values ​​and generate an auxiliary decision report based on the analysis results. The causal reasoning explanation module is used to make causal inferences on the auxiliary decision-making report and obtain a causal reasoning report based on the inference results and the auxiliary decision-making report. The disadvantaged group tilt module is used to calculate based on regional type data and complaint assessment data, and correct the congestion prediction value according to the calculation results to obtain the prediction correction value. Methods for calculating congestion predictions based on regional type data and complaint assessment data, and then revising the predictions based on the calculation results, include: Substitute the regional type data and complaint assessment data into the calculation formula: ; Obtain the predicted correction value ,in, This is the congestion forecast value. For correction factor, Complaint assessment data for remote areas, Complaint assessment data for the central area; Regional type data is used to divide the data collection area into urban central areas and remote urban areas; when When the value is less than 0.8, the predicted correction value will be returned to the auxiliary decision generation module to replace the congestion prediction value.

2. The e-government big data processing system according to claim 1, characterized in that, The system also includes: a traffic data acquisition module, a depth data acquisition module, and a data differentiation and processing module, wherein: The traffic data acquisition module is used to collect traffic datasets, which include traffic flow data, average vehicle speed data, road occupancy data, and event occurrence data. The deep data acquisition module is used to collect deep datasets, which include policy impact data, remote sensing congestion data, and complaint assessment data. The data differentiation and processing module is used to store key datasets in a database, display causal reasoning reports through a visualization panel, identify the causal reasoning reports, and process them based on the identification results.

3. The e-government big data processing system according to claim 1, characterized in that, The steps for analyzing eigenvectors include: Step 1: Obtain a set of historical feature vectors stored in the database, and compare them with the current time based on the timestamp. Group the comparison results from smallest to largest and label them accordingly. The labeling results are L1, L2, L3, ..., Ln. Use the labeling results as the sample set. Step 2: Divide the sample set into 70% training set, 15% test set and 15% validation set, and build the first traffic congestion prediction model based on the sample set. Step 3: Input the historical feature vector and the feature vector into the traffic congestion prediction model for calculation. The specific formula for calculation is as follows: ; Obtain the first congestion prediction value ,in, To integrate the number of models, It is the Sigmoid activation function. The number of decision trees in a single ensemble model. For the first The dynamic weighting factors of each decision tree. For the first In the group integration model, the first Gradient boosting decision trees For feature vectors, As a weighting factor for the attention mechanism, For multi-head self-attention functions, For the first Historical feature vectors of the ensemble model; Step 4: Calculate the error value based on the first congestion prediction value from Step 3. The specific formula for calculation is as follows: ; Obtain the error value ,in, For the sample size, This represents the actual congestion level. Step 5: Based on the error value in Step 4, judge according to the error threshold. When the error value is less than or equal to the error threshold, output the traffic congestion prediction model. When the error value is greater than the error threshold, return to Step 3 to retrain the first traffic congestion prediction model. Step 6: Input the feature vector into the traffic congestion prediction model and output the congestion prediction value; Step 7: Output the congestion prediction value to the auxiliary decision generation module.

4. The e-government big data processing system according to claim 1, characterized in that, Methods for analyzing congestion forecasts and generating decision support reports based on the analysis results include: Based on traffic threshold intervals (R1, R2); When the congestion prediction value is less than R1, a congestion report and emergency instructions are generated; when the congestion prediction value is greater than or equal to R1 and less than R2, a medium report and optimization instructions are generated; when the congestion prediction value is greater than or equal to R2, a normal report is generated. A normal report includes a statement indicating that traffic conditions are predicted to be good, and staff should monitor the situation according to the pre-set workflow. The medium-level report includes optimization instructions and states that the predicted traffic conditions are moderate, and staff are requested to monitor whether the traffic congestion situation escalates. The congestion report includes emergency instructions and a description of the predicted traffic congestion, requesting staff to immediately arrange for management personnel to go to the scene to direct traffic; The optimization instruction contains a set of characters representing the trigger for semaphore optimization; Emergency commands contain a set of characters representing the generation of the best route to the congestion site; Package the normal report, moderate report, and congestion report to obtain a decision support report.

5. The e-government big data processing system according to claim 1, characterized in that, Methods for making causal inferences about decision support reports and combining the inference results with the decision support reports include: When the auxiliary decision-making report is a moderate report or a congestion report, the causal reasoning mechanism is triggered; The causal inference mechanism is based on a causal inference rule base. It compares the input feature vector, traffic dataset, and deep dataset, and outputs the inference result. The inference results are integrated into the corresponding auxiliary decision-making report to obtain the causal reasoning report.

6. The e-government big data processing system according to claim 2, characterized in that, Methods for identifying causal reasoning reports and processing the results include: When the causal reasoning report is rated as medium, an optimization instruction will be sent to the intelligent controller; When the causal reasoning report is a congestion report, the emergency instructions are sent to the intelligent route generation tool to obtain the emergency route, and the emergency route is output to the staff receiving terminal; Key datasets include traffic datasets, deep datasets, feature vectors, decision support reports, causal inference reports, and prediction corrections.

7. A method for processing e-government big data, implemented according to any one of claims 1-6, characterized in that, The work includes the following steps: S1: Collect traffic datasets, which include traffic flow data, average vehicle speed data, road occupancy data, and event occurrence data; S2: Collect deep datasets, which include policy impact data, remote sensing congestion data, and complaint assessment data; S3: Preprocess the traffic dataset and deep dataset, and extract feature data to obtain feature vectors; S4: Analyze the feature vectors to obtain congestion prediction values; S5: Analyze the congestion forecast values ​​and generate a decision support report based on the analysis results; S6: Make causal inferences on the auxiliary decision-making report, and obtain a causal reasoning report based on the inference results and the auxiliary decision-making report; S7: Calculate based on regional type data and complaint assessment data, and correct the congestion prediction value according to the calculation results to obtain the predicted correction value; S8: Store key datasets in the database, display causal reasoning reports through a visualization panel, identify the causal reasoning reports, and process them based on the identification results.

Images