Vehicle data auditing method and device, equipment and storage medium

Abstract

The invention provides a vehicle data auditing method and device, equipment and a storage medium. The method comprises the following steps: analyzing a data security regulation text through a natural language processing technology, and extracting a data security entity and a constraint relationship thereof to construct a dynamic knowledge graph; determining a corresponding structured fact according to the vehicle multi-source data; mapping the structured facts to corresponding nodes of the knowledge graph to establish semantic associations; and performing reasoning verification based on the associated knowledge graph, judging whether the data processing activity meets compliance requirements or not, and generating an audit report. It can be understood that according to the scheme, intelligent and automatic auditing of the whole life cycle of the vehicle data is achieved, the method can dynamically adapt to regulation changes, and the auditing efficiency and accuracy are remarkably improved.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, and more specifically to a method, apparatus, device, and storage medium for auditing vehicle data. Background Technology

[0002] With the increasing prevalence and sophistication of intelligent connected vehicles, these vehicles continuously generate massive amounts of data during daily operation. This data not only includes basic operational information such as vehicle status and control commands, but also highly sensitive personal privacy information such as user identity and location tracking. Therefore, conducting systematic, accurate, and efficient compliance audits of vehicle data has become a crucial step in ensuring data security and fulfilling legal obligations.

[0003] In existing technologies, compliance auditing of vehicle data is typically achieved using a matching method based on fixed rules. Specifically, this method relies on manual intervention, pre-converting safety regulations into specific keywords or regular expressions, and then auditing the acquired multi-source vehicle data files.

[0004] However, due to the highly specialized nature, complex semantics, and continuous updating of relevant safety regulations, it is difficult to keep audit rules up-to-date after the regulations are revised, which fails to meet the urgent need for efficient, accurate, and automated compliance supervision of vehicle data security.

[0005] It should be noted that the information disclosed in the background section of this application is intended only to enhance the understanding of the general background of this application, and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0006] In view of this, this application provides a method, apparatus, device and storage medium for auditing vehicle data, in order to solve the problems of existing technologies that rely on human experience, have lagging rule updates and are difficult to automate continuous auditing.

[0007] Firstly, this application provides a method for auditing vehicle data, including: Data security regulations are analyzed using natural language processing technology to extract data security entities and their constraints. A dynamic knowledge graph is then constructed based on these entities and constraints. The data security entities include data subjects, data processing activities, and security control measures. Based on multi-source vehicle data, determine the structured facts corresponding to the multi-source vehicle data; The structured facts are mapped to the corresponding nodes of the dynamic knowledge graph to establish a semantic association between the structured facts and the data security entity; Based on the dynamic knowledge graph that has established the semantic associations, inference verification is performed to determine whether the structured facts meet the compliance requirements of the data security regulations and generate a corresponding audit report.

[0008] In this embodiment, natural language processing (NLP) technology is used to automatically parse data security regulatory texts, transforming complex provisions into structured data security entities and their constraints. A dynamic knowledge graph is then constructed based on this structure, replacing manual interpretation and rule writing, thus enabling the understanding and calculation of regulations. By converting multi-source vehicle data into structured facts and mapping them to corresponding nodes in the dynamic knowledge graph, the system can automatically establish semantic relationships between actual data and regulatory requirements, thereby achieving continuous tracking of vehicle data. Based on the semantically linked dynamic knowledge graph, automated reasoning and verification are performed. The system can determine in real time whether data processing activities comply with regulatory requirements and automatically generate audit reports. It is understood that using NLP technology to automatically parse data security regulatory texts fundamentally solves the problems of reliance on human experience and delayed rule updates in traditional methods, achieving efficient, accurate, and adaptive automated compliance auditing that can adapt to regulatory changes.

[0009] One possible implementation includes: The dynamic knowledge graph is also used to automatically update the data security entity nodes and constraint relationships of the dynamic knowledge graph based on updates to the data security regulations text.

[0010] In this embodiment, by enabling the dynamic knowledge graph to automatically adjust its entity nodes and constraint relationships according to updates to data security regulations, the compliance judgment benchmark carried by the knowledge graph can evolve synchronously with the external legal environment. This automatic synchronization mechanism fundamentally overcomes the lag problem of traditional manual maintenance methods, significantly reduces the reliance on continuous manual intervention, and thus ensures that the audit system can maintain the timeliness and accuracy of its audit judgments autonomously and over the long term.

[0011] In one possible implementation, the step of parsing data security regulatory text using natural language processing technology to extract data security entities and their constraints includes: Identify data security regulatory texts and determine data security entities using named entity recognition technology; The constraint relationships between the data security entities are extracted using relation extraction techniques.

[0012] In this embodiment, named entity recognition technology is used to automatically identify key concepts as data security entities from data security regulatory texts. Relationship extraction technology is then used to accurately extract the logical relationships between these entities. This transforms unstructured regulatory provisions, which rely on manual interpretation, into structured, machine-processable data security entities and their constraints. This automation process significantly reduces reliance on the subjective experience of professionals, fundamentally improving the objectivity, consistency, and processing efficiency of building compliance knowledge models.

[0013] In one possible implementation, determining the structured facts corresponding to the vehicle multi-source data based on the vehicle multi-source data includes: Configuration parameters, operation records, and communication content are extracted from the vehicle's multi-source data, and the extracted configuration parameters, operation records, and communication content are transformed into structured facts in a preset format using information extraction technology.

[0014] In this embodiment, by systematically extracting key elements from multi-source vehicle data—including configuration parameters, operation records, and communication content—and using information extraction technology to uniformly transform them into structured facts in a preset format, standardized processing of multi-source data is achieved. This effectively overcomes the problem that the original data is messy and difficult to use directly for automated comparison, laying a solid data foundation for building accurate and reliable audit judgments.

[0015] In one possible implementation, mapping the structured facts to corresponding nodes in the dynamic knowledge graph and establishing a semantic association between the structured facts and the data security entity includes: Based on semantic similarity calculation, the text description in the structured facts is vectorized and matched with the data security entity corresponding to the dynamic knowledge graph; Based on the similarity between the text description in the structured facts and the data security entity corresponding to the dynamic knowledge graph, and a preset similarity threshold, a semantic association is established between the structured facts and the data security entity.

[0016] In this embodiment, by vectorizing the textual descriptions in structured facts and the data security entities in the dynamic knowledge graph, and automatically matching them based on semantic similarity calculation, the semantic strength of their association can be accurately quantified. By using a preset similarity threshold as the criterion, a standardized mapping from specific data to regulatory concepts is achieved, thereby reliably establishing semantic associations. This effectively replaces the subjective association method relying on human experience, significantly improving the objectivity, consistency, and processing efficiency of the mapping process.

[0017] In one possible implementation, the inference verification based on the dynamic knowledge graph with established semantic associations includes: In the dynamic knowledge graph where the semantic associations have been established, the data security entity corresponding to the structured fact is determined; Iterate through the data security entities to check whether they are linked to the corresponding security control measures.

[0018] In this embodiment, by automatically locating data security entity nodes corresponding to specific data within a dynamic knowledge graph with established semantic relationships, and systematically traversing and querying whether these nodes are linked to necessary security control measure nodes through the graph structure, automated verification of compliance status is achieved. It can be understood that the above process transforms the logical constraints of regulations into computable and traversable path checks, enabling efficient and comprehensive discovery of compliance gaps in data processing activities. This significantly improves the automation level and objectivity of audit results, providing core technical support for the formation of continuous and proactive monitoring capabilities.

[0019] One possible implementation also includes: Based on the compliance deviations identified in the audit report, executable data monitoring rules are dynamically generated. The aforementioned data monitoring rules are used to perform compliance checks on newly generated multi-source data from the vehicle.

[0020] In this embodiment, by transforming compliance deviation conclusions in the audit report into specific, executable data monitoring rules, and applying these rules to subsequently generated multi-source vehicle data, a closed loop from single audit to continuous compliance monitoring is achieved. It is understood that this embodiment enables the system to proactively prevent similar risks based on historical audit findings, thereby significantly reducing reliance on repetitive manual audits, enhancing the initiative and continuity of compliance assurance, and giving the entire audit system the ability to self-optimize and evolve.

[0021] Secondly, embodiments of this application provide a vehicle data auditing device, comprising: The dynamic knowledge graph construction module is used to parse data security regulations texts using natural language processing technology, extract data security entities and their constraints, and construct a dynamic knowledge graph based on the data security entities and constraints. The data security entities include data subjects, data processing activities, and security control measures. The structured fact determination module is used to determine the structured facts corresponding to the vehicle multi-source data based on the vehicle multi-source data. The semantic association establishment module is used to map the structured facts to the corresponding nodes of the dynamic knowledge graph and establish semantic associations between the structured facts and the data security entities. The audit report generation module is used to perform reasoning verification based on the dynamic knowledge graph, determine whether the structured facts meet the compliance requirements of the data security regulatory text, and generate a corresponding audit report.

[0022] Thirdly, embodiments of this application provide an electronic device, including: processor; Memory; And a computer program, wherein the computer program is stored in the memory, the computer program including instructions that, when executed by the processor, cause the electronic device to perform the method described in any one of the first aspects.

[0023] Fourthly, embodiments of this application provide a computer-readable storage medium including a stored program, wherein, when the program is executed, it controls the device where the computer-readable storage medium is located to perform the method described in any one of the first aspects.

[0024] It is understood that the image forming apparatus provided in the second aspect, the electronic device provided in the third aspect, and the computer-readable storage medium provided in the fourth aspect are all used to perform some or all of the methods provided in this application. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods, and will not be repeated here. Attached Figure Description

[0025] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the embodiments 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.

[0026] Figure 1 This is a schematic diagram of an application scenario provided by an embodiment of this application.

[0027] Figure 2 This is a flowchart illustrating a vehicle data auditing method provided in an embodiment of this application.

[0028] Figure 3 This is a schematic diagram of the structure of a vehicle data auditing platform provided in an embodiment of this application.

[0029] Figure 4 This is a schematic diagram of the structure of a vehicle data auditing device provided in an embodiment of this application.

[0030] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0031] To better understand the technical solution of this application, the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0032] It should be understood that the described embodiments are merely some, not all, of the embodiments in this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application.

[0033] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. The singular forms “a,” “the,” and “the” used in the embodiments of this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0034] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0035] Vehicle data auditing refers to the systematic, automated inspection and logical verification of data processing activities throughout the entire lifecycle of intelligent connected vehicles, from generation, collection, transmission, storage, use to final destruction. Its core objective is to ensure that these activities continuously comply with relevant safety laws and regulations, industry technical standards, and the company's own governance strategies.

[0036] To facilitate understanding, specific application scenarios will be illustrated below.

[0037] See Figure 1 This is a schematic diagram illustrating an application scenario provided by an embodiment of this application. For example... Figure 1 As shown, this application scenario includes a vehicle 101 and a server 102 located in the cloud or backend. During operation, the vehicle 101 continuously generates multi-source heterogeneous data, including vehicle status, environmental perception, user interaction, and system operation. This data is transmitted to the server 102 via the network through the vehicle communication unit. The server 102, as the core processing platform, hosts the audit system and is responsible for in-depth analysis of the received data.

[0038] It should be noted that, Figure 1The application scenario illustrated is merely an example. In practical applications, depending on system design, computing power allocation, and real-time requirements, audit-related algorithms can adopt a flexible functional deployment architecture. The vehicle-side can also perform some data processing and preliminary analysis, such as performing real-time compliance matching of key operation logs locally or performing anonymization preprocessing on sensitive data, before uploading the necessary information to the server for audit analysis. Therefore, the scope of protection of this application is not limited to... Figure 1 The interactive modes shown, and the data auditing solutions implemented in the vehicle, cloud, or vehicle-cloud collaborative environment, should all fall within the scope of this application.

[0039] In related technologies, a matching method based on fixed rules is typically used to achieve compliance auditing of vehicle data. Specifically, this method relies on manual intervention, pre-converting safety regulations into specific keywords or regular expressions and storing them on the server; then, the acquired multi-source vehicle data files are audited and checked.

[0040] However, due to the highly specialized nature, complex semantics, and continuous updating of relevant safety regulations, it is difficult to keep audit rules up-to-date after the regulations are revised, which fails to meet the urgent need for efficient, accurate, and automated compliance supervision of vehicle data security.

[0041] To address the aforementioned issues, this application embodiment utilizes natural language processing (NLP) technology to automatically parse data security regulatory texts, transforming complex provisions into structured data security entities and their constraints. Based on this, a dynamic knowledge graph is constructed. This process replaces manual interpretation and rule writing, enabling the understanding and calculation of regulations. By converting multi-source vehicle data into structured facts and mapping them to corresponding nodes in the dynamic knowledge graph, the system can automatically establish semantic relationships between actual data and regulatory requirements, thereby achieving continuous tracking of vehicle data. Based on the semantically linked dynamic knowledge graph, automated reasoning and verification are performed. The system can determine in real time whether data processing activities comply with regulatory requirements and automatically generate audit reports. It is understood that using NLP technology to automatically parse data security regulatory texts fundamentally solves the problems of reliance on human experience and delayed rule updates in traditional methods, achieving efficient, accurate, and adaptive automated compliance auditing that adapts to regulatory changes. Specifically, detailed descriptions are provided below in conjunction with the accompanying drawings and specific embodiments.

[0042] See Figure 2 This is a flowchart illustrating a vehicle data auditing method provided in an embodiment of this application. This method can be applied to the application scenarios shown above, such as... Figure 2 As shown, it mainly includes the following steps.

[0043] Step S201: Parse the data security regulations text using natural language processing technology, extract data security entities and their constraints, and construct a dynamic knowledge graph based on the data security entities and constraints.

[0044] In this embodiment, natural language processing (NLP) technology is first used to perform deep semantic analysis on the data security regulatory text. It is understood that NLP technology can parse text components, understand the contextual meaning of technical terms, and identify the core elements in the text that carry key compliance requirements.

[0045] Specifically, the system will automatically identify and extract data security entities from regulatory texts. These data security entities include "data subjects," "data processing activities," and "security control measures." "Data subjects" refer to the objects protected by regulations, such as specific data types like vehicle user personal identification information, real-time vehicle location trajectories, or biometric features. "Data processing activities" primarily describe the operational behaviors performed on data subjects, such as the collection, storage, transmission, use, or deletion of data—a series of lifecycle stages. "Security control measures" stipulate the management requirements that must be implemented in specific processing activities to ensure data security, such as encryption, access control, or fulfilling notification obligations.

[0046] After identifying the aforementioned data security entities, the system will further analyze the semantic logic of the text using natural language processing technology to automatically extract the specific constraints between these entities. For example, from a passage of regulation stating that "personal biometric information shall be stored in encrypted form," the system can not only identify the three entities "personal biometric information" (data subject), "storage" (data processing activity), and "encryption" (security control measures), but also accurately extract the constraint that the "storage" activity imposes on the "personal biometric information" data subject, requiring "encryption measures to be adopted."

[0047] Ultimately, all extracted data security entities and the constraints between them will be systematically organized and stored, forming a dynamic knowledge graph. In this dynamic knowledge graph, various entities serve as nodes, and the constraints between entities serve as edges connecting the nodes.

[0048] The dynamic knowledge graph is dynamic because its construction is not a one-time event, but rather it is dynamically updated as data security regulations are promulgated and revised. Specifically, in one possible implementation, the dynamic knowledge graph can automatically update its data security entity nodes and constraint relationships based on updates to data security regulations.

[0049] In a specific embodiment, when new data security regulations are promulgated or existing data security regulations are revised, the system can re-execute the above parsing and extraction process, automatically adding and updating the new or changed requirements in the form of entities and relationships to the knowledge graph, thereby ensuring that the entire dynamic knowledge graph can evolve in sync with the external legal environment, providing a continuously accurate and real-time updated rule benchmark for subsequent automated auditing.

[0050] Understandably, this automatic synchronization mechanism fundamentally overcomes the lag problem of traditional manual maintenance methods, significantly reduces the reliance on continuous manual intervention, and thus ensures that the audit system can maintain the timeliness and accuracy of its audit judgments in a long-term and autonomous manner.

[0051] As a further refinement of the aforementioned process of constructing a dynamic knowledge graph, this application also provides a specific technical path for achieving automated parsing of regulatory texts. Specifically, in one possible implementation, data security regulatory texts are identified using named entity recognition technology to determine data security entities; and constraint relationships between data security entities are extracted using relation extraction technology.

[0052] The system employs named entity recognition (NAME) technology to scan and analyze data security regulations, automatically identifying and classifying all keywords or phrases with specific compliance semantics into data security entities. For example, when processing the clause "Vehicle trajectory data must be anonymized when transmitted outside the vehicle," the NAME recognition module accurately identifies "vehicle trajectory data" as the data subject, "transmission" as a data processing activity, and "anonymization" as a security control measure. In this way, key elements originally hidden within paragraphs are automatically and accurately located and extracted, laying the foundation for subsequent logical analysis.

[0053] After successfully identifying all relevant entities, the relation extraction technology is activated to analyze the context of the entities and accurately determine the constraints between them. Continuing with the above example, the relation extraction module analyzes the syntactic and semantic relationships between the three entities "vehicle trajectory data," "transmission," and "anonymization," and then extracts the constraint relationship that "the data processing activity of transmission" imposes "anonymization" on the data subject "vehicle trajectory data."

[0054] It should be noted that the aforementioned sequential processing of named entity recognition and relation extraction is merely one preferred technical path for parsing data security regulatory texts. Alternative or parallel implementations can also employ an end-to-end joint model to extract entities and relations in one step, or introduce deeper semantic understanding techniques (such as dependency parsing) to assist in determining complex logical relationships within long sentences. The selection and combination of these specific technologies all aim to achieve the same core objective: to automatically transform unstructured regulatory texts into a set of structured data security entities and their constraints that can be directly used by a dynamic knowledge graph, without human intervention. This automated process is the cornerstone for ensuring that the entire auditing system can intelligently understand regulations and evolve dynamically with them.

[0055] In this embodiment, named entity recognition technology is used to automatically identify key concepts from data security regulatory texts as data security entities, and relation extraction technology is used to accurately extract the logical relationships between these entities. This transforms the originally unstructured regulatory provisions, which rely on human interpretation, into structured data security entities and their constraints that can be processed automatically by machines. This significantly reduces the reliance on the subjective experience of professionals and improves the objectivity, consistency, and processing efficiency of building compliance knowledge models from the source.

[0056] Step S202: Based on the vehicle multi-source data, determine the structured facts corresponding to the vehicle multi-source data.

[0057] In this embodiment of the application, after completing the structured modeling of the regulatory text, the structured facts corresponding to the vehicle multi-source data are determined based on the vehicle multi-source data.

[0058] As a complex cyber-physical system, a vehicle continuously generates a large amount of raw data from numerous electronic control units, sensors, application software, and network interfaces during operation. This data comes from diverse sources, has varying formats, and implicit semantics, collectively constituting the multi-source vehicle data required for auditing. The purpose of this step is to transform this raw, unstructured, or semi-structured multi-source data into a standardized representation with a unified format, clear semantics, and the ability to be directly recognized and processed by subsequent knowledge graphs. This representation is called structured facts.

[0059] Understandably, the process of determining structured facts is essentially a process of information extraction and semantic standardization. The system needs to parse the input multi-source vehicle data, identify the segments with compliance significance, and reorganize them into machine-understandable statements. The key to this process is that the extracted structured facts must be able to establish a clear correspondence with the data security entities and data processing activities in the dynamic knowledge graph constructed in step S201.

[0060] For example, from a network traffic record of in-vehicle communication, the system may need to identify the core event "Application module A is reading vehicle location information". These identified core events and states will be constructed as "Application module A" performing a "read" operation, with the object being "vehicle location information".

[0061] Understandably, through the above steps, the originally messy multi-source vehicle data is transformed into a series of semantically clear structured facts. Each structured fact clearly describes a specific state or operation that occurs within the vehicle, directly corresponding to the instantiation of one or more entities in the knowledge graph.

[0062] In one possible implementation, configuration parameters, operation records, and communication content are extracted from multi-source vehicle data, and the extracted configuration parameters, operation records, and communication content are transformed into structured facts in a preset format using information extraction technology.

[0063] First, it's important to clarify that configuration parameters refer to the static policies and rules defined in the vehicle software and system. Their data primarily originates from system configuration files such as access control lists and standardized data dictionaries used to describe data semantics. Operation logs refer to the dynamic sequence of events generated during system runtime. Their data mainly comes from system log files generated by various electronic control units and applications. Communication content refers to the data and control information exchanged during internal vehicle network or external vehicle communication. This data is obtained through deep analysis of captured network traffic packets.

[0064] Information extraction technology refers to the technical means of identifying and extracting key information items that conform to a predetermined pattern from raw data based on predefined rules or models. For example, from an access control list, the configuration parameter "ECU_A is allowed to read the sensor data area" can be extracted. Information extraction technology identifies the three key information items "ECU_A", "read", and "sensor data area" from this configuration statement according to predetermined syntax or semantic rules.

[0065] It should also be noted that the aforementioned "preset format" is a preset output specification, the purpose of which is to organize the extracted key information items into a standardized, machine-readable data structure. For example, the three information items extracted above can be organized according to the logical statement structure of "the subject has operation rights to the data object", thereby generating a structured fact with a uniform format such as "ECU_A has read rights to the sensor data area".

[0066] In this embodiment, by systematically extracting key elements from multi-source vehicle data—including configuration parameters, operation records, and communication content—and using information extraction technology to uniformly transform them into structured facts in a preset format, standardized processing of multi-source data is achieved. This effectively overcomes the problem that the original data is messy and difficult to use directly for automated comparison, laying a solid data foundation for building accurate and reliable audit judgments.

[0067] Step S203: Map structured facts to corresponding nodes in the dynamic knowledge graph to establish semantic associations between structured facts and data security entities.

[0068] In this embodiment, structured facts extracted from multi-source vehicle data are precisely matched, or mapped, with a dynamic knowledge graph constructed based on data security regulations. For each structured fact describing the specific operating status of a vehicle, its semantic home is found in the dynamic knowledge graph expressing compliance requirements, thereby establishing a computable semantic relationship between real "data" and standardized "regulations".

[0069] As we can understand, structured facts are statements of specific events, while nodes in a dynamic knowledge graph are definitions of categories. Mapping is the process of classifying or linking a specific fact to its corresponding category (graph node).

[0070] For example, a structured fact might be "The infotainment system transmitted the vehicle's geographic location coordinates to an external network server." The corresponding nodes in the dynamic knowledge graph are located as follows: "Infotainment system" and "vehicle geographic location coordinates" are mapped to the corresponding nodes under the "Data Subject" category; the action of "transmission" is mapped to the corresponding nodes under the "Data Processing Activity" category; simultaneously, it is necessary to identify how this structured fact as a whole relates to the regulatory logic subgraph related to "sensitive location information processing" within the graph.

[0071] Through this precise mapping, structured facts that originally only described "what happened" are linked to a dynamic knowledge graph framework composed of regulatory logic that explains "what is permitted or required." Each structured fact establishes a clear semantic association with one or more data security entity nodes through mapping. This means that every specific data operation that occurs on the vehicle is given compliance-level meaning and coordinates in the knowledge graph. After this step is completed, the dynamic knowledge graph is no longer just a static regulatory library, but transforms into a "compliance scenario model" injected with the current real-time data state of the vehicle, preparing for the next step of directly performing logical verification and audit judgment on this model.

[0072] Furthermore, in one possible implementation, based on semantic similarity calculation, the textual descriptions in the structured facts are vectorized and matched with the data security entities corresponding to the dynamic knowledge graph; based on the similarity between the textual descriptions in the structured facts and the data security entities corresponding to the dynamic knowledge graph, and a preset similarity threshold, a semantic association between the structured facts and the data security entities is established.

[0073] Specifically, firstly, the system utilizes word embedding or sentence embedding techniques from natural language processing to transform textual descriptions in structured facts (e.g., "infotainment system transmits vehicle location data") and data security entities in dynamic knowledge graphs (e.g., "geographical location information," "data transmission") into numerical vectors in a high-dimensional space. These transformed vectors capture the deep semantic features of the text, ensuring that semantically similar texts are also positioned close to each other in the vector space. For example, the vector representations of "vehicle location data" and "geographical location information" will be very similar, while their vector representations will differ significantly from those of "engine speed."

[0074] Then, this proximity is quantified by calculating the semantic similarity between the two vectors. The system calculates the similarity between the structured fact description vector and each candidate data security entity vector in the knowledge graph. Finally, a judgment is made based on a preset similarity threshold: when a certain similarity value exceeds the threshold, the system considers the structured fact to be semantically sufficiently relevant to the corresponding data security entity, thus establishing a formal semantic association between the two.

[0075] It should be noted that other methods can also be used to establish the association, and this application does not impose specific restrictions on this. In this embodiment, by vectorizing the textual descriptions in structured facts and the data security entities in the dynamic knowledge graph, and automatically matching them based on semantic similarity calculation, the semantic strength of their association can be accurately quantified. By using a preset similarity threshold as the criterion, a standardized mapping from specific data to regulatory concepts is achieved, thereby reliably establishing semantic associations. This effectively replaces the subjective association method relying on human experience, significantly improving the objectivity, consistency, and processing efficiency of the mapping process.

[0076] Step S204: Perform reasoning verification based on the dynamic knowledge graph with established semantic associations, determine whether the structured facts meet the compliance requirements of the data security regulations, and generate the corresponding audit report.

[0077] In this embodiment, reasoning verification is performed based on a dynamic knowledge graph with established semantic relationships to determine whether structured facts meet the compliance requirements of data security regulations and generate a corresponding audit report. It can be understood that "reasoning verification" refers to systematically examining whether the actual data operations constituted by all mapped structured facts fully comply with the regulatory logic requirements corresponding to the dynamic knowledge graph.

[0078] It should be noted that the object operated on by "reasoning verification" refers to a dynamic knowledge graph with established semantic relationships. In this dynamic knowledge graph model, structured facts serve as instance nodes, which are connected to data security entity nodes representing regulatory concepts through semantic relationships, forming a semantic network that contains both abstract rules and concrete behaviors, and can be traversed and queried by the program. Reasoning verification involves running a series of logical checking rules on this network.

[0079] In one possible implementation, a data security entity corresponding to a structured fact is identified within a dynamic knowledge graph that has established semantic associations; the data security entity is then traversed and queried to determine whether it is linked to the corresponding security control measures.

[0080] For example, a regulation might be expressed in the graph as: "Personal location information must be explicitly authorized by the user before it can be transmitted outside the vehicle." When the system verifies an associated fact—such as "the vehicle APP_A transmitted the vehicle's real-time location"—it locates the corresponding node and edge in the graph and checks if there exists a compliant path from the "vehicle's real-time location" node, through the "transmission" edge, and finally linked to the "explicitly authorized by the user" node. If the path exists, the operation is deemed compliant; if the path is missing (e.g., only the transmission fact is found, but no authorization record is associated with it), the system identifies a compliance gap, i.e., a violation.

[0081] Understandably, transforming the logical constraints of regulations into computable and traversable path checks can efficiently and comprehensively identify compliance gaps in data processing activities, thereby significantly improving the automation level and objectivity of audit results, and providing core technical support for forming continuous and proactive monitoring capabilities.

[0082] Finally, the system summarizes, analyzes, and formats all these verification results, automatically generating a structured audit report. This report not only lists the conclusions but also traces and points out the specific basis for each judgment based on the correlation between the knowledge graph and the facts (such as which regulations were violated or which control link was missing), thus forming a traceable chain of audit evidence.

[0083] In this embodiment, natural language processing (NLP) technology is used to automatically parse data security regulatory texts, transforming complex provisions into structured data security entities and their constraints. A dynamic knowledge graph is then constructed based on this structure, replacing manual interpretation and rule writing, thus enabling the understanding and calculation of regulations. By converting multi-source vehicle data into structured facts and mapping them to corresponding nodes in the dynamic knowledge graph, the system can automatically establish semantic relationships between actual data and regulatory requirements, thereby achieving continuous tracking of vehicle data. Based on the semantically linked dynamic knowledge graph, automated reasoning and verification are performed. The system can determine in real time whether data processing activities comply with regulatory requirements and automatically generate audit reports. It is understood that using NLP technology to automatically parse data security regulatory texts fundamentally solves the problems of reliance on human experience and delayed rule updates in traditional methods, achieving efficient, accurate, and adaptive automated compliance auditing that can adapt to regulatory changes.

[0084] Building upon the completion of a single audit and the generation of a report, a closed loop from single audit to continuous compliance monitoring is achieved by transforming compliance deviation conclusions from the audit report into specific, actionable data monitoring rules and applying these rules to subsequently generated multi-source vehicle data. Here, compliance deviations refer to specific non-compliance items between data processing activities and regulatory requirements identified during reasoning and verification.

[0085] Specifically, the system analyzes the root causes of these compliance deviations and then dynamically generates executable data monitoring rules. For example, if an audit finds that an operation of "transmitting vehicle trajectory data to the cloud" is deemed a violation due to the lack of a corresponding "user authorization record," the system can automatically generate a corresponding data monitoring rule. The core logic of this rule is: when a "transmission of vehicle trajectory data" event is detected, a valid "user authorization" associated event must exist simultaneously. This rule is formalized into machine-understandable instructions or configurations.

[0086] Subsequently, the system will utilize data monitoring rules to perform compliance checks on newly generated multi-source data from the vehicle. When the vehicle generates new multi-source data during subsequent operation, the monitoring module will apply these rules for real-time or near real-time scanning and filtering. Once data processing behavior that matches the violation pattern in the rules is detected (for example, the transmission of trajectory data without associated authorized records reappears), the system can immediately trigger an alarm, attempt to block it, or record more detailed evidence, thereby preventing or recording violations at the first opportunity, rather than waiting for the next periodic audit.

[0087] In this embodiment, the system can proactively prevent similar risks based on historical audit findings, thereby significantly reducing reliance on repetitive manual audits, enhancing the initiative and continuity of compliance assurance, and enabling the entire audit system to have the ability to self-optimize and evolve.

[0088] Corresponding to the above embodiments, this application also provides a vehicle data auditing platform. Specifically, see [link to relevant documentation]. Figure 3 This figure illustrates the structure of a vehicle data auditing platform provided in this application embodiment. As shown, the platform comprises a knowledge layer, a data layer, an audit engine, and an application layer. The algorithms in the knowledge layer primarily use natural language processing to parse data security regulations, extract data security entities and their relationships, and construct a dynamic knowledge graph based on these entities and relationships. The algorithms in the data layer primarily acquire multi-source vehicle data and determine the structured facts corresponding to the multi-source vehicle data. The algorithms in the audit engine primarily map the structured facts to corresponding nodes in the dynamic knowledge graph, establishing semantic associations between the structured facts and data security entities; and perform reasoning verification based on the semantically associated dynamic knowledge graph to determine whether the structured facts meet the compliance requirements of the data security regulations. The algorithms in the application layer primarily generate corresponding audit reports and display the data's lifecycle and compliance status within the vehicle on a visual dashboard.

[0089] For details, please refer to the embodiments described above. For the sake of brevity, this application will not repeat them here.

[0090] Corresponding to the above embodiments, this application also provides a vehicle data auditing device. Specifically, see... Figure 4This figure shows a schematic diagram of a vehicle data auditing device provided in an embodiment of this application. The vehicle data auditing device 400 specifically includes: a dynamic knowledge graph construction module 401, a structured fact determination module 402, a semantic association establishment module 403, and an audit report generation module 404. Specifically, the dynamic knowledge graph construction module 401 is used to parse data security regulatory texts using natural language processing technology, extract data security entities and their constraints, and construct a dynamic knowledge graph based on the data security entities and constraints; the structured fact determination module 402 is used to determine structured facts corresponding to the multi-source vehicle data; the semantic association establishment module 403 is used to map the structured facts to corresponding nodes in the dynamic knowledge graph, establishing semantic associations between the structured facts and the data security entities; the audit report generation module 404 is used to perform reasoning verification based on the dynamic knowledge graph, determine whether the structured facts meet the compliance requirements of the data security regulatory text, and generate a corresponding audit report.

[0091] For details, please refer to the embodiments described above. For the sake of brevity, this application will not repeat them here.

[0092] Corresponding to the above embodiments, this application also provides a schematic diagram of the structure of an electronic device. See also... Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device 500 may include a processor 501, a memory 502, and a communication unit 503. These components communicate through one or more buses. Those skilled in the art will understand that the structure of the electronic device shown in the figure does not constitute a limitation on the embodiments of the present invention. It may be a bus-shaped structure or a star-shaped structure, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0093] The communication unit 503 is used to establish a communication channel, enabling the electronic device to communicate with other devices. It receives user data from other devices or sends user data to other devices.

[0094] The processor 501 serves as the control center of the electronic device, connecting various parts of the device via interfaces and lines. It executes software programs, instructions, and / or modules stored in the memory 502, and calls data stored in the memory to perform various functions and / or process data. The processor may be composed of integrated circuits (ICs), such as a single packaged IC or multiple packaged ICs with the same or different functions connected together. For example, the processor 501 may consist only of a central processing unit (CPU). In this embodiment, the CPU may have a single processing core or include multiple processing cores.

[0095] The memory 502 is used to store the execution instructions of the processor 501. The memory 502 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0096] When the execution instructions in memory 502 are executed by processor 501, the electronic device 500 is able to perform operations. Figure 2 Some or all of the steps in the illustrated embodiments.

[0097] In a specific implementation, this application also provides a computer storage medium, wherein the computer storage medium may store a program, and when the program is executed, it may include some or all of the steps in the various embodiments of the simulation scene generation method provided by this invention. The storage medium may be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0098] In this application embodiment, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent the existence of A alone, the simultaneous existence of A and B, or the existence of B alone. A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects have an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, and c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0099] Those skilled in the art will recognize that the units and algorithm steps described in the embodiments disclosed herein can be implemented using electronic hardware, computer software, or a combination of electronic hardware and software. 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 implementation should not be considered beyond the scope of this application.

[0100] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0101] In the several embodiments provided in this application, any function, if implemented as a software functional unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0102] The same or similar parts between the various embodiments in this specification can be referred to mutually. In particular, the device embodiments and terminal embodiments are basically similar to the method embodiments, so the description is relatively simple, and the relevant parts can be referred to the description in the method embodiments.

Claims

1. A method for auditing vehicle data, characterized in that, include: Data security regulations are analyzed using natural language processing technology to extract data security entities and their constraints. A dynamic knowledge graph is then constructed based on these entities and constraints. The data security entities include data subjects, data processing activities, and security control measures. Based on multi-source vehicle data, determine the structured facts corresponding to the multi-source vehicle data; The structured facts are mapped to the corresponding nodes of the dynamic knowledge graph to establish a semantic association between the structured facts and the data security entity; Based on the dynamic knowledge graph that has established the semantic associations, inference verification is performed to determine whether the structured facts meet the compliance requirements of the data security regulations and generate a corresponding audit report.

2. The method according to claim 1, characterized in that, include: The dynamic knowledge graph is also used to automatically update the data security entity nodes and constraint relationships of the dynamic knowledge graph based on updates to the data security regulations text.

3. The method according to claim 1, characterized in that, The process of parsing data security regulations using natural language processing technology to extract data security entities and their constraints includes: Identify data security regulatory texts and determine data security entities using named entity recognition technology; The constraint relationships between the data security entities are extracted using relation extraction techniques.

4. The method according to claim 1, characterized in that, The step of determining the structured facts corresponding to the multi-source vehicle data includes: Configuration parameters, operation records, and communication content are extracted from the vehicle's multi-source data, and the extracted configuration parameters, operation records, and communication content are transformed into structured facts in a preset format using information extraction technology.

5. The method according to claim 4, characterized in that, The step of mapping the structured facts to corresponding nodes in the dynamic knowledge graph and establishing a semantic association between the structured facts and the data security entity includes: Based on semantic similarity calculation, the text description in the structured facts is vectorized and matched with the data security entity corresponding to the dynamic knowledge graph; Based on the similarity between the text description in the structured facts and the data security entity corresponding to the dynamic knowledge graph, and a preset similarity threshold, a semantic association is established between the structured facts and the data security entity.

6. The method according to claim 5, characterized in that, The inference verification based on the dynamic knowledge graph with established semantic associations includes: In the dynamic knowledge graph where the semantic associations have been established, the data security entity corresponding to the structured fact is determined; Iterate through the data security entities to check whether they are linked to the corresponding security control measures.

7. The method according to claim 1, characterized in that, Also includes: Based on the compliance deviations identified in the audit report, executable data monitoring rules are dynamically generated. The aforementioned data monitoring rules are used to perform compliance checks on newly generated multi-source data from the vehicle.

8. A vehicle data auditing device, characterized in that, include: The dynamic knowledge graph construction module is used to parse data security regulations texts using natural language processing technology, extract data security entities and their constraints, and construct a dynamic knowledge graph based on the data security entities and constraints. The data security entities include data subjects, data processing activities, and security control measures. The structured fact determination module is used to determine the structured facts corresponding to the vehicle multi-source data based on the vehicle multi-source data. The semantic association establishment module is used to map the structured facts to the corresponding nodes of the dynamic knowledge graph and establish semantic associations between the structured facts and the data security entities. The audit report generation module is used to perform reasoning verification based on the dynamic knowledge graph, determine whether the structured facts meet the compliance requirements of the data security regulatory text, and generate a corresponding audit report.

9. An electronic device, characterized in that, include: processor; Memory; And a computer program, wherein the computer program is stored in the memory, the computer program including instructions that, when executed by the processor, cause the electronic device to perform the method of any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device on which the computer-readable storage medium is located to perform the method according to any one of claims 1 to 7.

Images