Cross-department false information processing method based on urban social governance

Abstract

The invention provides a cross-department false information processing method based on urban social governance, and belongs to the technical field of false information processing, and the method comprises the steps: obtaining to-be-recognized first information; the first information is information obtained by standardizing the multi-source data; extracting risk features of the first information based on a predetermined standard feature library; the risk features comprise content features, main body features, propagation features, time features and regional features; inputting the risk feature into a pre-trained long-short-term memory network to obtain a risk level of the first information; and in response to the fact that the risk level is greater than a predetermined risk level, generating a cross-department disposal work order, and sending the cross-department disposal work order to a corresponding government affair department based on the department responsibility graph to realize cross-department false information processing. According to the invention, the processing efficiency of cross-department false information can be improved.

Description

Technical Field

[0001] This application belongs to the field of false information processing technology, and more specifically, it relates to a cross-departmental false information processing method based on urban social governance. Background Technology

[0002] In the information age, the spread of misinformation poses a serious threat to social stability and economic development. Currently, combating misinformation has become an important part of urban social governance.

[0003] In existing technologies, the handling of false information mainly relies on independent efforts by various departments. For example, the Cyberspace Administration of China (CAC) is responsible for monitoring online false information, the market supervision department handles false advertising in the commercial sector, and the public security department combats illegal and criminal activities related to false information. Each department typically establishes independent information collection systems, analysis models, and handling procedures. For instance, the CAC uses web crawler technology to capture suspected false information and makes judgments through keyword matching and manual review; the market supervision department collects clues about false advertising based on consumer complaints and market inspections and uses industry standards for evaluation and handling.

[0004] However, this decentralized approach to handling misinformation has significant shortcomings. The independent procedures of each department prevent the formation of a synergistic effect. For complex misinformation incidents involving multiple sectors, multiple transfers of leads are often necessary, delaying the response and resulting in low efficiency in dealing with misinformation. Summary of the Invention

[0005] The purpose of this application is to provide a cross-departmental method for handling false information based on urban social governance, so as to improve the efficiency of handling false information across departments.

[0006] A first aspect of this application provides a cross-departmental method for processing false information based on urban social governance, comprising:

[0007] Obtain the first information to be identified; the first information is the information after standardization of multi-source data;

[0008] Risk features are extracted from the first information based on a predetermined standard feature library; risk features include content features, subject features, dissemination features, time features, and geographical features.

[0009] The risk characteristics are input into a pre-trained long short-term memory network to obtain the risk level of the first information;

[0010] In response to a risk level exceeding the predetermined risk level, a cross-departmental handling work order is generated. Based on the departmental responsibility map, the cross-departmental handling work order is sent to the corresponding government department to achieve cross-departmental handling of false information.

[0011] Cross-departmental work orders include initial information, risk level, characteristic analysis of the initial information, and handling recommendations for the initial information.

[0012] In one embodiment of this application, the standard feature library includes a keyword library, a behavior pattern library, and a propagation path library; risk features of the first information are extracted based on the predetermined standard feature library, including:

[0013] Extract keywords, behavioral patterns, and propagation paths from the primary information;

[0014] By using a pre-defined standard feature library, the keywords, behavioral patterns, and propagation paths of the first information are standardized and transformed to obtain the risk characteristics of the first information.

[0015] In one embodiment of this application, cross-departmental processing work orders are sent to the corresponding government departments based on a departmental responsibility map to achieve cross-departmental handling of false information, including:

[0016] Based on the departmental responsibility map and the feature analysis of the first information in the cross-departmental handling work order, the candidate government departments corresponding to the first information are determined.

[0017] Based on the historical processing efficiency of each processing department, the candidate processing departments corresponding to the first information are filtered to obtain the target government department corresponding to the first information;

[0018] The cross-departmental processing work order will be sent to the target government department corresponding to the first piece of information.

[0019] In one embodiment of this application, the cross-departmental method for handling false information based on urban social governance further includes:

[0020] Using the gradient boosting algorithm, a multi-dimensional credit indicator system is used to evaluate the credit of the entity corresponding to the first piece of information, and obtain the credit rating of the entity corresponding to the first piece of information. The multi-dimensional credit indicator system includes indicators for evaluating violations, indicators for evaluating entity attributes, indicators for evaluating rectification attitude, indicators for evaluating historical credit, and indicators for evaluating social impact.

[0021] Based on the credit rating of the entity corresponding to the first piece of information, and the pre-defined rule base for mapping punitive measures to credit ratings, a disposal suggestion for the entity corresponding to the first piece of information is generated.

[0022] In one embodiment of this application, the cross-departmental method for handling false information based on urban social governance further includes:

[0023] The compliance of the handling recommendations is determined by using a predefined paradigm knowledge base to check whether the subject corresponding to the first piece of information meets the specified characteristics.

[0024] The specified characteristics include: compliance characteristics, privacy protection characteristics, procedural legitimacy characteristics, proportionality principle characteristics, ethical remedies characteristics, and subject eligibility characteristics.

[0025] In one embodiment of this application, the cross-departmental method for handling false information based on urban social governance further includes:

[0026] Obtain the second information to be identified;

[0027] Extracting the temporal and spatial features of the secondary information;

[0028] By correlating the temporal and spatial features of the second information, the spatiotemporal correlation features of the second information are obtained.

[0029] Spatiotemporal correlation features are input into a pre-trained ensemble learning model to obtain the burst probability of the second information, and cross-departmental information processing is performed on the second information based on the burst probability.

[0030] In one embodiment of this application, the cross-departmental method for handling false information based on urban social governance further includes:

[0031] The cross-departmental method for handling misinformation is based on consortium blockchains deployed on various government departments and smart contracts deployed on these consortium blockchains.

[0032] A second aspect of this application provides a cross-departmental false information processing device based on urban social governance, comprising:

[0033] The data acquisition unit is used to acquire the first information to be identified; the first information is the information after standardization of multi-source data.

[0034] The feature extraction unit is used to extract risk features of the first information based on a predetermined standard feature library; the risk features include content features, subject features, dissemination features, time features, and geographical features;

[0035] The risk prediction unit is used to input risk features into a pre-trained long short-term memory network to obtain the risk level of the first information.

[0036] The first information processing unit is used to generate a cross-departmental handling work order in response to a risk level greater than the predetermined risk level, and send the cross-departmental handling work order to the corresponding government department based on the departmental responsibility map to realize cross-departmental false information processing.

[0037] Cross-departmental work orders include initial information, risk level, characteristic analysis of the initial information, and handling suggestions for the initial information.

[0038] In one embodiment of this application, the standard feature library includes a keyword library, a behavior pattern library, and a propagation path library; the risk prediction unit is specifically used for:

[0039] Extract keywords, behavioral patterns, and propagation paths from the primary information;

[0040] By using a pre-defined standard feature library, the keywords, behavioral patterns, and propagation paths of the first information are standardized and transformed to obtain the risk characteristics of the first information.

[0041] In one embodiment of this application, the first information processing unit is specifically used for:

[0042] Based on the departmental responsibility map and the feature analysis of the first information in the cross-departmental handling work order, the candidate government departments corresponding to the first information are determined.

[0043] Based on the historical processing efficiency of each processing department, the candidate processing departments corresponding to the first information are filtered to obtain the target government department corresponding to the first information;

[0044] The cross-departmental processing work order will be sent to the target government department corresponding to the first piece of information.

[0045] In one embodiment of this application, the cross-departmental false information processing device 20 based on urban social governance further includes a second information processing unit;

[0046] The second information processing unit is used for:

[0047] Using the gradient boosting algorithm, a multi-dimensional credit indicator system is used to evaluate the credit of the entity corresponding to the first piece of information, and obtain the credit rating of the entity corresponding to the first piece of information. The multi-dimensional credit indicator system includes indicators for evaluating violations, indicators for evaluating entity attributes, indicators for evaluating rectification attitude, indicators for evaluating historical credit, and indicators for evaluating social impact.

[0048] Based on the credit rating of the entity corresponding to the first piece of information, and the pre-defined rule base for mapping punitive measures to credit ratings, a disposal suggestion for the entity corresponding to the first piece of information is generated.

[0049] In one embodiment of this application, the cross-departmental false information processing device 20 based on urban social governance further includes a judgment unit;

[0050] The decision unit is used for:

[0051] The compliance of the handling recommendations is determined by using a predefined paradigm knowledge base to check whether the subject corresponding to the first piece of information meets the specified characteristics.

[0052] The specified characteristics include: compliance characteristics, privacy protection characteristics, procedural legitimacy characteristics, proportionality principle characteristics, ethical remedies characteristics, and subject eligibility characteristics.

[0053] In one embodiment of this application, the cross-departmental false information processing device 20 based on urban social governance further includes a third information processing unit;

[0054] The third information processing unit is used to: acquire the second information to be identified;

[0055] Extracting the temporal and spatial features of the secondary information;

[0056] By correlating the temporal and spatial features of the second information, the spatiotemporal correlation features of the second information are obtained.

[0057] Spatiotemporal correlation features are input into a pre-trained ensemble learning model to obtain the burst probability of the second information, and cross-departmental information processing is performed on the second information based on the burst probability.

[0058] In one embodiment of this application, the cross-departmental misinformation processing method is implemented based on consortium blockchains deployed on various government departments and smart contracts deployed on the consortium blockchains.

[0059] A third aspect of this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of the above-described method for processing cross-departmental false information based on urban social governance.

[0060] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method for processing cross-departmental false information based on urban social governance.

[0061] The beneficial effects of the cross-departmental false information processing method based on urban social governance provided in this application embodiment are as follows:

[0062] This application's embodiments standardize multi-source data, facilitating the extraction and analysis of different types of features and resolving the issues of inconsistent data formats and chaotic feature recognition standards across departments. Furthermore, this application's embodiments input the risk characteristics of the first piece of information into a pre-trained Long Short-Term Memory (LSTM) network to obtain the risk level of the first piece of information, providing a basis for tiered handling. When the risk level exceeds a predetermined risk level, this application's embodiments send work orders to each department based on a departmental responsibility map, clarifying each department's tasks and deadlines. This solves the problem of traditional cross-departmental clue transfer relying on manual emails or offline coordination, which easily leads to missing the optimal handling opportunity. It enhances cross-departmental collaborative work efficiency and improves the handling efficiency of false information. Attached Figure Description

[0063] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0064] Figure 1 A flowchart illustrating a cross-departmental method for processing false information based on urban social governance, provided as an embodiment of this application;

[0065] Figure 2 A structural block diagram of a cross-departmental false information processing device based on urban social governance provided in an embodiment of this application;

[0066] Figure 3 This is a schematic block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0067] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0068] To make the objectives, technical solutions, and advantages of this application clearer, the following description will be provided in conjunction with the accompanying drawings and specific embodiments.

[0069] Please refer to Figure 1 , Figure 1 The flowchart of a cross-departmental false information processing method based on urban social governance provided in an embodiment of this application can be executed by an electronic device, and the method may include: S101-S104.

[0070] S101: Obtain the first information to be identified.

[0071] In this embodiment, the first information is the standardized information obtained from the multi-source data. This embodiment can obtain the multi-source data in the following way:

[0072] Data is collected from a first data source using a distributed crawler component, from a second data source using an ETL tool, and from a third data source using an API interface. In this embodiment, the multiple data sources can be government departments, such as market supervision departments, cyberspace administration departments, and public security departments.

[0073] The equipment categorizes and stores the collected multi-source data based on administrative region labels in a data storage space for subsequent standardization processing. These administrative region labels can be four-level classification labels, such as "province-city-district-grid". The multi-source data includes both structured and unstructured data, which can be processed separately. For example, natural language processing models are used to perform word segmentation, part-of-speech tagging, and information recognition on the unstructured data to extract multiple keywords. Database to Resource Description Framework (D2R) conversion technology is used to transform the structured data into a knowledge graph architecture; for example, a penalty decision document is converted into a "subject-behavior-spatiotemporal" knowledge graph triple. Finally, the processed data is standardized and mapped to obtain the first piece of information to be identified.

[0074] S102: Extract risk features based on a pre-defined standard feature library.

[0075] Risk characteristics include content characteristics, subject characteristics, dissemination characteristics, time characteristics, and geographical characteristics.

[0076] In this embodiment, the predetermined standard feature library is a fake information feature library, which includes three types: a keyword library, a behavior pattern library, and a propagation path library. The keyword library can be obtained by extracting and filtering from historical data of government departments and online data. The keyword library can be dynamically updated based on the Term Frequency-Inverse Document Frequency (TF-IDF) algorithm. The behavior pattern library can be constructed based on the typical behaviors of risk subjects in government scenarios, and the data format in the behavior pattern library can be "account registration - content publishing - bulk forwarding". The propagation path library can be obtained based on a fake information propagation network constructed from a graph database. The propagation path library stores information such as the nodes in the fake information propagation network, the forwarding relationships between nodes, and the propagation timeline.

[0077] The risk characteristics of primary information include content characteristics, subject characteristics, dissemination characteristics, time characteristics, and geographical characteristics. Content characteristics refer to the content of the primary information, such as events and behaviors related to the subject. Subject characteristics include the subject's historical credit records and account registration information. Dissemination characteristics include the number of reposts, comments, and dissemination speed; for example, a risk warning signal may be triggered when the number of reposts exceeds 500 within one hour. Time characteristics represent sensitive time points corresponding to events and behaviors related to the subject; for example, the probability of rumor dissemination increases by 300% during natural disasters. Geographical characteristics represent information related to regional risk levels; for example, information from areas with a high incidence of financial fraud automatically has a higher risk weight.

[0078] In one embodiment, risk features of the first information are extracted based on a predetermined standard feature library, including:

[0079] Extract keywords, behavioral patterns, and propagation paths from the primary information;

[0080] By using a pre-defined standard feature library, the keywords, behavioral patterns, and propagation paths of the first information are standardized and transformed to obtain the risk characteristics of the first information.

[0081] In this embodiment, core keywords with risk associations are extracted from the text content of the first information. Examples include "subject (Company XX), region (City XXXX District), content (fake health product advertisement), and dissemination carrier (WeChat Moments)." For the subject's actions and associated attributes in the first information, sequential and related behavioral logic is extracted through rule matching or pattern recognition. For example, the extracted behavioral pattern could be "Account A - posts content selling expired products - batch forwarding 300 times." For the information dissemination content in the first information, such as public opinion dissemination records and the trajectory of illegal information dissemination, key links and flow of dissemination are reconstructed through time-series analysis. For example, the extracted original dissemination path is: starting point - key node - carrier - region - time sequence.

[0082] After extracting the keywords, behavioral patterns, and propagation paths of the primary information, a pre-defined keyword library, behavioral pattern library, and propagation path library can be used to standardize and transform these elements to obtain the risk characteristics of the primary information. For example, if the original propagation path mentioned above is origin-key node-carrier-region-time sequence, it indicates that the propagation path implicitly contains risk dimensions such as propagation characteristics, regional characteristics, and time characteristics.

[0083] S103: Input the risk features into a pre-trained long short-term memory network to obtain the risk level of the first information.

[0084] In this embodiment, the risk features of the first information are input into a pre-trained Long Short-Term Memory (LSTM) network for risk level assessment. This pre-trained LSM network is a three-layer neural network model, consisting of an input layer, a hidden layer, and an output layer. Five types of risk features of the first information are input into the input layer. Upon receiving these features, the input layer transmits them to the hidden layer. The hidden layer processes the time-series features through LSM network units, focuses on key features using an attention mechanism, and analyzes these key features. The analysis results are then transmitted to the output layer, which generates a risk level and confidence score based on the analysis results. For example, the risk level of the first information can be divided into five levels, L1-L5, with L5 representing the highest risk level.

[0085] S104: In response to a risk level exceeding the predetermined risk level, a cross-departmental handling work order is generated. Based on the departmental responsibility map, the cross-departmental handling work order is sent to the corresponding government department to achieve cross-departmental handling of false information.

[0086] Among them, cross-departmental handling work orders include first information, risk level, characteristic analysis of first information, and handling suggestions for first information.

[0087] In this embodiment, the predetermined risk level is assumed to be L3. If the risk level of the first piece of information is L5, then its risk level is greater than the predetermined risk level. In this case, the device can automatically generate a cross-departmental handling work order. The content of this cross-departmental handling work order includes the original information, risk level, feature analysis report, and handling suggestions. For example, "XX school engages in fraudulent enrollment, with a risk level of L4, involving XXXX District of XX City. The XX City Education Bureau has investigated its qualifications and suspended enrollment."

[0088] After generating a cross-departmental handling work order, the work order is sent to the corresponding government departments based on the departmental responsibility map to achieve cross-departmental handling of false information. The departmental responsibility map is a map of the areas that each government department should be responsible for. For example, false information related to education will be pushed to the Cyberspace Administration, the Education Department, and the Public Security Department.

[0089] As can be seen from the above, this application embodiment, by standardizing multi-source data, facilitates the extraction and analysis of different types of features, solving the problems of inconsistent data formats and chaotic feature recognition standards across departments. This application embodiment also inputs the risk features of the first information into a pre-trained long short-term memory network to obtain the risk level of the first information, providing a basis for tiered handling. When the risk level is greater than a predetermined risk level, this application embodiment sends work orders to each department based on a departmental responsibility map, clarifying the tasks and deadlines of each department. This solves the problem of traditional cross-departmental clue transfer relying on manual emails or offline coordination, which easily leads to missing the best handling opportunity. It can enhance the efficiency of cross-departmental collaborative work and improve the efficiency of handling false information.

[0090] In one embodiment of this application, cross-departmental processing work orders are sent to the corresponding government departments based on a departmental responsibility map to achieve cross-departmental handling of false information, including:

[0091] Based on the departmental responsibility map and the feature analysis of the first information in the cross-departmental handling work order, the candidate government departments corresponding to the first information are determined.

[0092] Based on the historical processing efficiency of each processing department, the candidate processing departments corresponding to the first information are filtered to obtain the target government department corresponding to the first information;

[0093] The cross-departmental processing work order will be sent to the target government department corresponding to the first piece of information.

[0094] In this embodiment, the departmental responsibility map is a structured database of government departmental responsibilities, including the statutory responsibilities, boundaries of authority, common collaboration scenarios, and hierarchical relationships between departments, used to quickly match the responsible entities for information handling. Common collaboration scenarios include, for example, the Market Supervision Bureau being responsible for investigating and handling false advertising, and the Cyberspace Administration being responsible for controlling online information; hierarchical relationships between departments include, for example, the jurisdiction of municipal and district-level departments. Based on the feature analysis of the first information and the departmental responsibility map, multiple government departments that may have the authority to handle such cases are initially selected. For example, candidate departments involved in online false information may include the Cyberspace Administration, the Public Security Bureau, and the Market Supervision Bureau, and these departments are selected as candidate government departments.

[0095] Historical performance indicators for each processing department in handling similar information include average processing time, completion rate, public satisfaction, and collaboration efficiency with other departments. Based on these historical performance indicators, the historical processing efficiency of each department can be obtained. By filtering candidate processing departments corresponding to the first piece of information based on their historical processing efficiency, the target government department corresponding to the first piece of information can be determined, ensuring efficient execution of work orders and improving event handling efficiency. For example, if the processing efficiency of the first piece of information sent to department AA is 20% higher than that of department BB, similar work orders can be prioritized and sent to department AA.

[0096] In one embodiment of this application, the cross-departmental misinformation processing method further includes:

[0097] Using the gradient boosting algorithm, a multi-dimensional credit indicator system is used to evaluate the credit of the entity corresponding to the first piece of information, and obtain the credit rating of the entity corresponding to the first piece of information. The multi-dimensional credit indicator system includes indicators for evaluating violations, indicators for evaluating entity attributes, indicators for evaluating rectification attitude, indicators for evaluating historical credit, and indicators for evaluating social impact.

[0098] Based on the credit rating of the entity corresponding to the first piece of information, and the pre-defined rule base for mapping punitive measures to credit ratings, a disposal suggestion for the entity corresponding to the first piece of information is generated.

[0099] In this embodiment, the gradient boosting algorithm can be the XGBoost algorithm. The multi-dimensional credit indicator system includes evaluations of violation behavior, subject attribute indicators, rectification attitude indicators, historical credit indicators, and social impact indicators. Violation behavior indicators can include the number of violations, the degree of harm, and the scope of dissemination; subject attribute indicators include subject type and industry risk level, and the subject type can be an individual or a company. Rectification attitude indicators include the timeliness of correction and the degree of cooperation. Historical credit indicators include the number of breaches of trust and credit repair records over a historical period. Social impact indicators include public opinion intensity and the number of public complaints.

[0100] In this embodiment, the credit rating of the subject corresponding to the first piece of information can be divided into 6 levels, for example:

[0101] AAA level (900-1000 points): Exemplary in trustworthiness, enjoying green channel access;

[0102] Grade A (700-899 points): Trustworthy entities, under normal management;

[0103] Grade B (500-699 points): General credit, subject to enhanced monitoring;

[0104] Level C (300-499 points): Warning of dishonesty, some rights restricted;

[0105] Level D (100-299 points): Serious breach of trust, subject to joint disciplinary action;

[0106] Level E (0-99 points): Blacklisted, completely restricted.

[0107] Using the gradient boosting algorithm, a credit rating is performed on the entity corresponding to the first piece of information based on a multi-dimensional credit indicator system, resulting in the entity's credit rating, including:

[0108] In this embodiment, after determining the credit rating of the entity corresponding to the first information, a proposed course of action for the entity can be generated by combining a preset mapping rule base of punitive measures and credit ratings. The preset mapping rule base of punitive measures and credit ratings includes multiple mapping rules, such as:

[0109] The punitive measures for Class D entities may include: market regulators listing them in the list of entities with abnormal operations, financial institutions reducing their credit limits by 50%, and restricting their participation in government procurement, etc.

[0110] The punitive measures for entities classified as Level E may include: being listed in the business anomaly directory by market regulators, having financial institutions reduce credit lines by 50%, being restricted from participating in government procurement, being restricted from high-end consumption, and having restrictions on their qualifications for office.

[0111] The pre-defined rule base for mapping disciplinary measures and credit ratings supports custom extensions, such as a special rule that directly downgrades those who publish false epidemic information during an epidemic to level D.

[0112] In this embodiment, after a government department has processed the entity corresponding to the first piece of information, it can synchronize the information to other government departments through a data interface. For example, it can interface with the State Administration for Market Regulation system to synchronize the list of entities with abnormal business operations; interface with the People's Bank of China's credit reporting system to update enterprise credit reports; and interface with the government service platform to restrict administrative approval authority. In this embodiment, information synchronization uses a message queue to handle high-concurrency synchronization requests, allowing information to be synchronized to different government departments simultaneously.

[0113] As can be seen from the above, this embodiment comprehensively reflects the credit status of an entity through a multi-dimensional credit indicator system, avoiding the one-sidedness of evaluation based on a single indicator; the gradient boosting algorithm effectively captures the non-linear relationships between indicators, improving the accuracy of credit rating assessment. Furthermore, this embodiment combines a preset mapping rule base to generate disposal suggestions for the entity corresponding to the first piece of information, effectively matching punitive measures with the entity's credit, ensuring both unified cross-departmental disposal standards and differentiated punishment.

[0114] In one embodiment of this application, the cross-departmental misinformation processing method further includes:

[0115] The compliance of the handling recommendations is determined by using a predefined paradigm knowledge base to check whether the subject corresponding to the first piece of information meets the specified characteristics.

[0116] The specified characteristics include: compliance characteristics, privacy protection characteristics, procedural legitimacy characteristics, proportionality principle characteristics, ethical remedies characteristics, and subject eligibility characteristics.

[0117] In this embodiment, the process of constructing the paradigm knowledge base includes: case feature extraction and classification, and construction of the paradigm knowledge base.

[0118] Case feature extraction and classification includes extracting six categories of features from historical cases, namely designated features: compliance features, privacy protection features, procedural legitimacy features, proportionality features, ethical remedies features, and subject eligibility features. Specifically, compliance features indicate whether relevant provisions of the Cybersecurity Law were violated; privacy protection features indicate whether sensitive personal information was leaked; procedural legitimacy features indicate whether the handling process complied with legal procedures; proportionality features indicate whether the punitive measures are commensurate with the severity of the violation; ethical remedies features indicate whether channels for appeals and their processing timelines were provided; and subject eligibility features indicate whether the handling entity possesses the legal authority. Based on these six categories of features, clustering algorithms can be used to classify historical cases, resulting in various categories of compliance cases, such as privacy breach cases and procedural violation cases.

[0119] The construction of the paradigm knowledge base includes building a three-layer knowledge base architecture, namely the rule layer, the case layer, and the reasoning layer.

[0120] The rule layer stores compliance rules such as legal provisions and departmental regulations; the case layer stores various categories of compliance cases, violation cases, and feature tags obtained above; and the reasoning layer stores case matching algorithms, such as graph similarity algorithms, as well as rules for generating correction strategies.

[0121] Based on the constructed paradigm knowledge base, it can be used to detect whether the handling suggestions of the subject corresponding to the first information meet the specified characteristics in order to determine the compliance of the handling suggestions. That is, the characteristics of the subject corresponding to the first information are matched with the violation cases in the paradigm knowledge base using graph similarity. If the similarity is greater than the similarity threshold, an early warning is triggered. For example, if it is found that the feature matching degree of a certain handling measure with the "excessive punishment" case reaches 85%, a compliance warning is immediately triggered.

[0122] As can be seen from the above, this embodiment can improve the accuracy of compliance judgment by constructing a paradigm knowledge base and using the paradigm knowledge base to detect the compliance of the handling suggestions of the subject corresponding to the first information.

[0123] In one embodiment of this application, the cross-departmental misinformation processing method further includes:

[0124] When the handling recommendation of the entity corresponding to the first piece of information is deemed non-compliant, correction is carried out based on a three-level correction mechanism. Level 1 correction involves the system automatically identifying the violation and recommending alternative measures; Level 2 correction involves the system automatically reporting to the department head for review if the handling department does not adopt the Level 1 correction recommendation; Level 3 correction involves the system recording the violation details and initiating an internal audit process if the department head still upholds the original handling.

[0125] The entire three-tiered correction mechanism described above can be stored on blockchain, ensuring information traceability.

[0126] In one embodiment of this application, the cross-departmental misinformation processing method further includes:

[0127] Obtain the second information to be identified;

[0128] Extracting the temporal and spatial features of the secondary information;

[0129] By correlating the temporal and spatial features of the second information, the spatiotemporal correlation features of the second information are obtained.

[0130] Spatiotemporal correlation features are input into a pre-trained ensemble learning model to obtain the burst probability of the second information, and cross-departmental information processing is performed on the second information based on the burst probability.

[0131] In this embodiment, the second information to be identified is obtained. The content of this second information may be the same as or different from the first information mentioned above, but the acquisition method is the same and will not be elaborated here. After obtaining the second information to be identified, the temporal and spatial features of the second information are extracted. The temporal features refer to using Long Short-Term Memory (LSTM) networks to analyze time series data and identify abnormal propagation patterns. The spatial features refer to dividing the data acquisition area into a 1km×1km grid and calculating the propagation node density of each grid. For example, when the number of propagation nodes in some grids is ≥10, a spatial risk warning can be triggered. This embodiment uses spatiotemporal indexing technology to associate spatial features and temporal features to obtain the spatiotemporal correlation features of the second information, which is beneficial for discovering high-risk combinations of "specific area + specific time". For example, the risk of fraudulent information propagation during weekday lunch breaks in financial districts increases by 200%.

[0132] In this embodiment, the pre-trained ensemble learning model can be trained based on historical fake information data. By inputting the spatiotemporal correlation features of the second piece of information into the pre-trained ensemble learning model, the outbreak probability of the second piece of information can be obtained, which is the outbreak probability within 72 hours. When the outbreak probability of the second piece of information exceeds a preset outbreak probability threshold, an early warning can be triggered. The pre-trained ensemble learning model in this embodiment can be automatically updated based on new propagation patterns, thereby improving prediction accuracy.

[0133] In one embodiment of this application, the cross-departmental misinformation processing method further includes:

[0134] The cross-departmental method for handling misinformation is based on consortium blockchains deployed on various government departments and smart contracts deployed on these consortium blockchains.

[0135] In this embodiment, the cross-departmental false information processing method is based on consortium blockchains deployed on various government departments and smart contracts deployed on the consortium blockchains. This facilitates data traceability, ensures data immutability, and provides conclusive evidence for dispute resolution.

[0136] Corresponding to the cross-departmental method for handling false information based on urban social governance in the above embodiment, Figure 2 This is a structural block diagram of a cross-departmental false information processing device based on urban social governance, provided as an embodiment of this application. For ease of explanation, only the parts relevant to the embodiment of this application are shown. References Figure 2 The cross-departmental false information processing device 20 based on urban social governance includes: a data acquisition unit 21, a feature extraction unit 22, a risk prediction unit 23, and a first information processing unit 24.

[0137] Among them, the data acquisition unit 21 is used to acquire the first information to be identified; the first information is the information after standardization of multi-source data;

[0138] The feature extraction unit 22 is used to extract risk features of the first information based on a predetermined standard feature library; the risk features include content features, subject features, dissemination features, time features, and geographical features;

[0139] Risk prediction unit 23 is used to input risk features into a pre-trained long short-term memory network to obtain the risk level of the first information;

[0140] The first information processing unit 24 is used to generate a cross-departmental handling work order in response to a risk level greater than a predetermined risk level, and send the cross-departmental handling work order to the corresponding government department based on the departmental responsibility map to achieve cross-departmental false information processing.

[0141] Cross-departmental work orders include initial information, risk level, characteristic analysis of the initial information, and handling recommendations for the initial information.

[0142] In one embodiment of this application, the standard feature library includes a keyword library, a behavior pattern library, and a propagation path library; the risk prediction unit 23 is specifically used for:

[0143] Extract keywords, behavioral patterns, and propagation paths from the primary information;

[0144] By using a pre-defined standard feature library, the keywords, behavioral patterns, and propagation paths of the first information are standardized and transformed to obtain the risk characteristics of the first information.

[0145] In one embodiment of this application, the first information processing unit 24 is specifically used for:

[0146] Based on the departmental responsibility map and the feature analysis of the first information in the cross-departmental handling work order, the candidate government departments corresponding to the first information are determined.

[0147] Based on the historical processing efficiency of each processing department, the candidate processing departments corresponding to the first information are filtered to obtain the target government department corresponding to the first information;

[0148] The cross-departmental processing work order will be sent to the target government department corresponding to the first piece of information.

[0149] In one embodiment of this application, the cross-departmental false information processing device 20 based on urban social governance further includes a second information processing unit 25;

[0150] The second information processing unit 25 is used for:

[0151] Using the gradient boosting algorithm, a multi-dimensional credit indicator system is used to evaluate the credit of the entity corresponding to the first piece of information, and obtain the credit rating of the entity corresponding to the first piece of information. The multi-dimensional credit indicator system includes indicators for evaluating violations, indicators for evaluating entity attributes, indicators for evaluating rectification attitude, indicators for evaluating historical credit, and indicators for evaluating social impact.

[0152] Based on the credit rating of the entity corresponding to the first piece of information, and the pre-defined rule base for mapping punitive measures to credit ratings, a disposal suggestion for the entity corresponding to the first piece of information is generated.

[0153] In one embodiment of this application, the cross-departmental false information processing device 20 based on urban social governance further includes a judgment unit 26;

[0154] Judgment unit 26 is used for:

[0155] The compliance of the handling recommendations is determined by using a predefined paradigm knowledge base to check whether the subject corresponding to the first piece of information meets the specified characteristics.

[0156] The specified characteristics include: compliance characteristics, privacy protection characteristics, procedural legitimacy characteristics, proportionality principle characteristics, ethical remedies characteristics, and subject eligibility characteristics.

[0157] In one embodiment of this application, the cross-departmental false information processing device 20 based on urban social governance further includes a third information processing unit 27;

[0158] The third information processing unit 27 is used to: acquire the second information to be identified;

[0159] Extracting the temporal and spatial features of the secondary information;

[0160] By correlating the temporal and spatial features of the second information, the spatiotemporal correlation features of the second information are obtained.

[0161] Spatiotemporal correlation features are input into a pre-trained ensemble learning model to obtain the burst probability of the second information, and cross-departmental information processing is performed on the second information based on the burst probability.

[0162] In one embodiment of this application, the cross-departmental misinformation processing method is implemented based on consortium blockchains deployed on various government departments and smart contracts deployed on the consortium blockchains.

[0163] See Figure 3 , Figure 3 This is a schematic block diagram of an electronic device provided according to an embodiment of this application. Figure 3The electronic device 300 in this embodiment may include one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The processors 301, input devices 302, output devices 303, and memories 304 communicate with each other via a communication bus 305. The memories 304 store computer programs, including program instructions. The processors 301 execute the program instructions stored in the memories 304. Specifically, the processors 301 are configured to invoke the program instructions to perform the functions of the units in the above-described device embodiments, for example... Figure 2 The functions of the data acquisition unit 21, feature extraction unit 22, risk prediction unit 23, and first information processing unit 24 are shown.

[0164] It should be understood that, in the embodiments of this application, the processor 301 may be a central processing unit (CPU), but it may also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0165] Input device 302 may include a touchpad, a fingerprint sensor (for collecting the user's fingerprint information and fingerprint orientation information), a microphone, etc., and output device 303 may include a display (LCD, etc.), a speaker, etc.

[0166] The memory 304 may include read-only memory and random access memory, and provides instructions and data to the processor 301. A portion of the memory 304 may also include non-volatile random access memory.

[0167] In specific implementations, the processor 301, input device 302, and output device 303 described in the embodiments of this application can execute the implementation methods described in the cross-departmental false information processing method based on urban social governance provided in the embodiments of this application, or they can execute the implementation methods of the electronic devices described in the embodiments of this application, which will not be repeated here.

[0168] In another embodiment of this application, a computer-readable storage medium is provided. This computer-readable storage medium stores a computer program, which includes program instructions. When executed by a processor, the program instructions implement all or part of the processes in the methods described above. Alternatively, the computer program can instruct related hardware to complete the process. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include any entity or device capable of carrying computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0169] The computer-readable storage medium can be an internal storage unit of the electronic device in any of the foregoing embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device. Furthermore, the computer-readable storage medium can include both internal and external storage units of the electronic device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0170] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0171] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the electronic devices and units described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0172] In the several embodiments provided in this application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces or units, or it may be an electrical, mechanical, or other form of connection.

[0173] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of this application, depending on actual needs.

[0174] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0175] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A cross-departmental method for processing false information based on urban social governance, characterized in that, include: Obtain the first information to be identified; The first piece of information is information obtained by standardizing multi-source data; Risk features of the first information are extracted based on a predetermined standard feature library; the risk features include content features, subject features, dissemination features, time features, and geographical features. The risk features are input into a pre-trained long short-term memory network to obtain the risk level of the first information; In response to a risk level exceeding a predetermined risk level, a cross-departmental handling work order is generated, and the cross-departmental handling work order is sent to the corresponding government department based on the departmental responsibility map to achieve cross-departmental handling of false information. The cross-departmental handling work order includes the first information, the risk level, the feature analysis of the first information, and the handling suggestions for the first information.

2. The method as described in claim 1, characterized in that, The standard feature library includes a keyword library, a behavior pattern library, and a propagation path library; the risk features extracted from the first information based on the predetermined standard feature library include: Extract the keywords, behavioral patterns, and propagation paths of the first information; The risk characteristics of the first information are obtained by standardizing and transforming the keywords, behavioral patterns, and propagation paths of the first information using a predetermined standard feature library.

3. The method as described in claim 1, characterized in that, The step of sending the cross-departmental processing work order to the corresponding government department based on the departmental responsibility map to achieve cross-departmental false information processing includes: Based on the departmental responsibility map and the feature analysis of the first information in the cross-departmental processing work order, the candidate government department corresponding to the first information is determined. Based on the historical processing efficiency of each processing department, the candidate processing departments corresponding to the first information are filtered to obtain the target government department corresponding to the first information. The cross-departmental processing work order is sent to the target government department corresponding to the first information.

4. The method as described in claim 1, characterized in that, Also includes: Using the gradient boosting algorithm, a credit evaluation is performed on the subject corresponding to the first information based on a multi-dimensional credit indicator system to obtain the credit rating of the subject corresponding to the first information. The multi-dimensional credit indicator system includes indicators for evaluating violations, indicators for evaluating the subject's attributes, indicators for evaluating the attitude towards rectification, indicators for evaluating historical credit, and indicators for evaluating social impact. Based on the credit rating of the entity corresponding to the first information and the preset rule base for mapping punitive measures to credit ratings, a disposal suggestion is generated for the entity corresponding to the first information.

5. The method as described in claim 1, characterized in that, Also includes: The compliance of the handling recommendations is determined by using a predetermined paradigm knowledge base to detect whether the handling recommendations of the subject corresponding to the first information meet the specified characteristics. The specified features include: compliance features, privacy protection features, procedural legitimacy features, proportionality features, ethical remedies features, and subject eligibility features.

6. The method as described in claim 1, characterized in that, Also includes: Obtain the second information to be identified; Extract the temporal and spatial features of the second information; The temporal and spatial features of the second information are correlated to obtain the spatiotemporal correlation features of the second information; The spatiotemporal correlation features are input into a pre-trained ensemble learning model to obtain the burst probability of the second information, and cross-departmental information processing is performed on the second information based on the burst probability.

7. The method as described in claim 1, characterized in that, Also includes: The cross-departmental method for handling misinformation is based on consortium blockchains deployed on various government departments and smart contracts deployed on these consortium blockchains.

8. A cross-departmental false information processing device based on urban social governance, characterized in that, include: The data acquisition unit is used to acquire the first information to be identified; The first piece of information is information obtained by standardizing multi-source data; The feature extraction unit is used to extract risk features of the first information based on a predetermined standard feature library; the risk features include content features, subject features, dissemination features, time features, and geographical features. A risk prediction unit is used to input the risk features into a pre-trained long short-term memory network to obtain the risk level of the first information; The first information processing unit is used to generate a cross-departmental handling work order in response to the risk level being greater than a predetermined risk level, and to send the cross-departmental handling work order to the corresponding government department based on the departmental responsibility map to achieve cross-departmental false information processing. The cross-departmental handling work order includes the first information, the risk level, the feature analysis of the first information, and the handling suggestions for the first information.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 7.

Images