Intelligent review method for software configuration file data of motor vehicle inspection and testing equipment

Abstract

The present application relates to the field of data review, and specifically discloses a method for intelligent review of the configuration file data of the motor vehicle inspection and testing equipment software, which uses natural language processing technology based on deep learning to process the configuration file of the motor vehicle inspection and testing equipment software, and extracts the semantic feature representation of each key data item in the configuration file through semantic embedding coding technology, and then, based on the contextual semantic association relationship between each key data item, updates the features and performs semantic enhancement processing on each key data item to achieve an in-depth understanding of the configuration file, thereby realizing intelligent identification and early warning prompts of sensitive information in the configuration file on this basis. In this way, the accuracy and efficiency of the configuration file review can be effectively improved, the workload of manual review can be reduced, and the safety risks caused by human negligence can be reduced.

Description

Technical Field

[0001] The present application relates to the field of data review, and more specifically, to a method for intelligently reviewing software configuration file data of motor vehicle inspection and testing equipment. Background Art

[0002] With the continuous advancement of motor vehicle inspection and testing technology, motor vehicle inspection and testing equipment plays a vital role in ensuring road traffic safety and promoting the improvement of vehicle technical conditions. Motor vehicle inspection and testing equipment usually relies on complex software systems to achieve automated and high-precision inspection processes. As the cornerstone of equipment operation, the correctness, security and compliance of software configuration files are directly related to the stability of the entire inspection system and the security of data.

[0003] Traditionally, configuration files for vehicle inspection and testing equipment software are manually written and maintained by technicians. During this process, configuration errors or security risks may occur due to human negligence. In particular, configuration files may contain hard-coded passwords, API keys, database connection information and other sensitive information. Once this information is leaked or maliciously used, it will pose a serious threat to the security of the vehicle inspection and testing system.

[0004] Traditional configuration file review methods mainly rely on manual inspection, which is time-consuming and labor-intensive and prone to omissions or errors. Especially when faced with a large number of configuration files and complex settings, it is inefficient and difficult to ensure the comprehensiveness and accuracy of the review. At the same time, some existing automated review tools are mainly at the level of simple rule matching or keyword search, unable to deeply understand the semantic content of configuration files, and have limitations when processing complex and changeable configuration files. It is difficult to accurately identify sensitive information hidden in the configuration files, resulting in insufficient reliability of the review results.

[0005] Therefore, an optimized method for intelligent review of software configuration file data of motor vehicle inspection and testing equipment is desired. Summary of the invention

[0006] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides a method for intelligent review of the configuration file data of the motor vehicle inspection and testing equipment software, which uses natural language processing technology based on deep learning to process the configuration file of the motor vehicle inspection and testing equipment software, and extracts the semantic feature representation of each key data item in the configuration file through semantic embedding coding technology, and then, based on the contextual semantic association relationship between each key data item, performs feature update and semantic enhancement processing on each key data item to achieve an in-depth understanding of the configuration file, thereby realizing intelligent identification and early warning prompts of sensitive information in the configuration file on this basis. In this way, the accuracy and efficiency of the configuration file review can be effectively improved, the workload of manual review can be reduced, and the safety risks caused by human negligence can be reduced.

[0007] According to one aspect of the present application, a method for intelligently reviewing software configuration file data of a motor vehicle inspection and testing equipment is provided, which includes:

[0008] Obtain software configuration files for motor vehicle inspection and testing equipment;

[0009] Performing semantic embedding coding on each key data item in the motor vehicle inspection and testing equipment software configuration file to obtain a set of semantic embedding coding vectors of the key data items in the configuration file;

[0010] Performing semantic enhancement on the set of semantic embedding coding vectors of the key data items of the configuration file to obtain a set of semantic embedding enhanced coding vectors of the key data items of the configuration file;

[0011] Based on the set of semantically embedded enhanced coding vectors of the key data items of the configuration file, it is determined whether to generate a configuration file security warning prompt.

[0012] In the above-mentioned intelligent review method for software configuration file data of motor vehicle inspection and testing equipment, semantic enhancement is performed on the set of semantic embedding coding vectors of key data items in the configuration file, including: based on the characteristic field effect distribution of the set of semantic embedding coding vectors of key data items in the configuration file, the set of semantic embedding coding vectors of key data items in the configuration file is updated to obtain a set of updated semantic embedding coding vectors of key data items in the configuration file; and bidirectional contextual semantic association enhancement is performed on the set of updated semantic embedding coding vectors of key data items in the configuration file to obtain a set of enhanced semantic embedding coding vectors of key data items in the configuration file.

[0013] In the above-mentioned intelligent review method for motor vehicle inspection and testing equipment software configuration file data, each key data item in the motor vehicle inspection and testing equipment software configuration file is semantically embedded and encoded to obtain a set of configuration file key data item semantic embedding encoding vectors, including: extracting a set of configuration file key data items from the motor vehicle inspection and testing equipment software configuration file; using a key item semantic embedding encoder based on the Bert model to semantically embed code each configuration file key data item in the set of configuration file key data items to obtain a set of configuration file key data item semantic embedding encoding vectors.

[0014] In the above-mentioned intelligent review method for software configuration file data of motor vehicle inspection and testing equipment, based on the feature field effect distribution of the set of semantic embedding coding vectors of configuration file key data items, the set of semantic embedding coding vectors of configuration file key data items is updated to obtain a set of updated configuration file key data item semantic embedding coding vectors, including: using a semantic reconstructor to perform feature reconstruction on each configuration file key data item semantic embedding coding vector in the set of configuration file key data item semantic embedding coding vectors to obtain a set of reconstructed configuration file key data item semantic embedding coding vectors; calculating the semantic depth implicit representation value of each reconstructed configuration file key data item semantic embedding coding vector in the set of reconstructed configuration file key data item semantic embedding coding vectors to obtain a set of key data item semantic depth implicit representation values; based on the sorting of the set of key data item semantic depth implicit representation values, reordering the set of configuration file key data item semantic embedding coding vectors to obtain the set of updated configuration file key data item semantic embedding coding vectors.

[0015] In the above-mentioned intelligent review method for software configuration file data of motor vehicle inspection and testing equipment, a semantic reconstructor is used to perform feature reconstruction on the semantic embedding coding vectors of each configuration file key data item in the set of semantic embedding coding vectors of the configuration file key data item to obtain a set of reconstructed semantic embedding coding vectors of the configuration file key data item, including: using a first weight parameter matrix and a second weight parameter matrix to perform continuous linear transformation on the semantic embedding coding vectors of the configuration file key data item to obtain the reconstructed semantic embedding coding vectors of the configuration file key data item.

[0016] In the above-mentioned intelligent review method for software configuration file data of motor vehicle inspection and testing equipment, the semantic depth implicit representation value of each reconstructed configuration file key data item semantic embedding coding vector in the set of the reconstructed configuration file key data item semantic embedding coding vector is calculated to obtain a set of semantic depth implicit representation values ​​of the key data item, including: calculating the square of the Euclidean norm of the semantic embedding coding vector of the reconstructed configuration file key data item to obtain the initial semantic depth implicit representation value of the key data item; calculating the ratio of the sum of the initial semantic depth implicit representation value of the key data item and the difference between the initial semantic depth implicit representation value of the key data item to obtain the semantic depth representation modulation coefficient; calculating the logarithmic function value of the semantic depth representation modulation coefficient with base 2 to obtain the semantic depth implicit representation value of the key data item.

[0017] In the above-mentioned intelligent review method for software configuration file data of motor vehicle inspection and testing equipment, the set of semantic embedding coding vectors of the updated configuration file key data items is subjected to bidirectional context semantic association enhancement to obtain the set of semantic embedding enhanced coding vectors of the configuration file key data items, including: inputting the set of semantic embedding coding vectors of the updated configuration file key data items into a first semantic encoder based on a forward LSTM model to obtain a set of forward key data item context semantic coding feature vectors; inputting the set of semantic embedding coding vectors of the updated configuration file key data items into a second semantic encoder based on a backward LSTM model to obtain a set of backward key data item context semantic coding feature vectors; and fusing the set of forward key data item context semantic coding feature vectors and the set of backward key data item context semantic coding feature vectors to obtain the set of configuration file key data item semantic embedding enhanced coding vectors.

[0018] In the above-mentioned intelligent review method for software configuration file data of motor vehicle inspection and testing equipment, the set of contextual semantic coding feature vectors of the forward key data items and the set of contextual semantic coding feature vectors of the backward key data items are fused to obtain the set of semantically embedded enhanced coding vectors of the configuration file key data items, including: using the first weighting parameter and the second weighting parameter to weighted modulate the set of contextual semantic coding feature vectors of the forward key data items and the set of contextual semantic coding feature vectors of the backward key data items respectively to obtain a set of weighted modulated contextual semantic coding feature vectors of the forward key data items and a set of weighted modulated contextual semantic coding feature vectors of the backward key data items; calculating the corresponding points between the set of weighted modulated contextual semantic coding feature vectors of the forward key data items, the set of weighted modulated contextual semantic coding feature vectors of the backward key data items and the bias item to obtain the set of semantically embedded enhanced coding vectors of the configuration file key data items.

[0019] In the above-mentioned intelligent review method for software configuration file data of motor vehicle inspection and testing equipment, whether to generate a configuration file security warning prompt is determined based on the set of semantically embedded enhanced coding vectors of the configuration file key data items, including: inputting each configuration file key data item semantically embedded enhanced coding vector in the set of semantically embedded enhanced coding vectors of the configuration file key data items into a classifier-based sensitive information identifier to obtain a recognition result, and the recognition result is used to indicate whether sensitive information exists; in response to the recognition result that sensitive information exists, generating a configuration file security warning prompt.

[0020] In the above-mentioned intelligent review method for software configuration file data of motor vehicle inspection and testing equipment, each configuration file key data item semantic embedding enhanced coding vector in the set of configuration file key data item semantic embedding enhanced coding vectors is respectively input into a classifier-based sensitive information identifier to obtain a recognition result, and the recognition result is used to indicate whether sensitive information exists, including: using the fully connected layer of the sensitive information identifier to fully connect the configuration file key data item semantic embedding enhanced coding vector to obtain a configuration file key data item semantic embedding fully connected coding feature vector; inputting the configuration file key data item semantic embedding fully connected coding feature vector into the Softmax classification function of the sensitive information identifier to obtain the probability value of the configuration file key data item semantic embedding enhanced coding vector belonging to each classification label, wherein the classification label includes the presence of sensitive information and the absence of sensitive information; and determining the classification label corresponding to the largest of the probability values ​​as the recognition result.

[0021] Compared with the prior art, the intelligent review method for the configuration file data of the motor vehicle inspection and testing equipment software provided by the present application uses natural language processing technology based on deep learning to process the configuration file of the motor vehicle inspection and testing equipment software, and extracts the semantic feature representation of each key data item in the configuration file through semantic embedding coding technology, and then, based on the contextual semantic association relationship between each key data item, updates the features of each key data item and performs semantic enhancement processing to achieve an in-depth understanding of the configuration file, thereby realizing intelligent identification and early warning prompts of sensitive information in the configuration file on this basis. In this way, the accuracy and efficiency of the configuration file review can be effectively improved, the workload of manual review can be reduced, and the safety risks caused by human negligence can be reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0023] Figure 1 The present invention is a flowchart of a method for intelligently reviewing software configuration file data of motor vehicle inspection and testing equipment according to an embodiment of the present application.

[0024] Figure 2 The present invention is a data flow diagram of a method for intelligently reviewing software configuration file data of motor vehicle inspection and testing equipment according to an embodiment of the present application.

[0025] Figure 3 It is a flowchart of sub-step S2 in the method for intelligent review of software configuration file data of motor vehicle inspection and testing equipment according to an embodiment of the present application.

[0026] Figure 4 This is a flowchart of sub-step S3 in the method for intelligent review of software configuration file data of motor vehicle inspection and testing equipment according to an embodiment of the present application.

[0027] Figure 5 It is a flowchart of sub-step S31 in the method for intelligent review of software configuration file data of motor vehicle inspection and testing equipment according to an embodiment of the present application. DETAILED DESCRIPTION

[0028] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0029] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0030] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. Meanwhile, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0031] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0032] In response to the technical problems described in the above background technology, this application proposes an optimized intelligent review method for the configuration file data of the motor vehicle inspection and testing equipment software, which uses natural language processing technology based on deep learning to process the configuration file of the motor vehicle inspection and testing equipment software, and extracts the semantic feature representation of each key data item in the configuration file through semantic embedding coding technology. Then, based on the contextual semantic association relationship between each key data item, each key data item is updated with features and semantic enhancement processing to achieve an in-depth understanding of the configuration file, thereby realizing intelligent identification and early warning prompts of sensitive information in the configuration file on this basis. In this way, the accuracy and efficiency of the configuration file review can be effectively improved, the workload of manual review can be reduced, and the safety risks caused by human negligence can be reduced.

[0033] Figure 1 The present invention is a flowchart of a method for intelligently reviewing software configuration file data of motor vehicle inspection and testing equipment according to an embodiment of the present application. Figure 2 The following is a data flow diagram of a method for intelligently reviewing software configuration file data of a motor vehicle inspection and testing device according to an embodiment of the present application. Figure 1 and Figure 2 As shown, the method for intelligent review of the software configuration file data of the motor vehicle inspection and testing equipment includes the following steps: S1, obtaining the software configuration file of the motor vehicle inspection and testing equipment; S2, performing semantic embedding coding on each key data item in the software configuration file of the motor vehicle inspection and testing equipment to obtain a set of semantic embedding coding vectors of the configuration file key data items; S3, performing semantic enhancement on the set of semantic embedding coding vectors of the configuration file key data items to obtain a set of semantic embedding enhanced coding vectors of the configuration file key data items; S4, determining whether to generate a configuration file security warning prompt based on the set of semantic embedding enhanced coding vectors of the configuration file key data items.

[0034] In the above-mentioned intelligent review method for the configuration file data of the motor vehicle inspection and testing equipment software, the step S1 is to obtain the configuration file of the motor vehicle inspection and testing equipment software. It should be understood that the normal operation of the motor vehicle inspection and testing equipment depends on various parameter settings in the configuration file, such as the device connection address, port number, authentication information, etc. Since many industry standards and rules have clear requirements for the security and compliance of configuration files. Therefore, in order to ensure that no hard-coded passwords or sensitive information are exposed in the configuration file, the present application further uses artificial intelligence technology to review sensitive information in the configuration file of the motor vehicle inspection and testing equipment software to ensure the security of the software system.

[0035] In the above-mentioned method for intelligent review of the configuration file data of the motor vehicle inspection and testing equipment software, the step S2 is to perform semantic embedding coding on each key data item in the configuration file of the motor vehicle inspection and testing equipment software to obtain a set of semantic embedding coding vectors of the key data items of the configuration file. Figure 3 Flow chart of sub-step S2 in the method for intelligently reviewing the software configuration file data of the motor vehicle inspection and testing equipment according to the embodiment of the present application. Figure 3 As shown, the step S2 includes the steps of: S21, extracting a set of configuration file key data items from the motor vehicle inspection and testing equipment software configuration file; S22, using a key item semantic embedding encoder based on a Bert model to perform semantic embedding encoding on each configuration file key data item in the set of configuration file key data items to obtain a set of semantic embedding encoding vectors of the configuration file key data items.

[0036] Specifically, the step S21 extracts a set of key data items of the configuration file from the software configuration file of the motor vehicle inspection and testing equipment. Specifically, considering that the software configuration file of the motor vehicle inspection and testing equipment includes multiple aspects such as device identification, user permissions, network configuration, etc. Therefore, in order to avoid information confusion in the subsequent file parsing process and thus omission of important information, the present application further extracts a set of key data items of the configuration file from the software configuration file of the motor vehicle inspection and testing equipment, and performs independent semantic understanding of each key data item of the configuration file to achieve a more sophisticated sensitive information review.

[0037] Specifically, the step S22 uses a key item semantic embedding encoder based on the Bert model to perform semantic embedding encoding on each configuration file key data item in the set of configuration file key data items to obtain a set of semantic embedding encoding vectors of the configuration file key data items. It should be understood that in order to accurately extract the semantic features of each configuration file key data item, the present application uses the Bert model with excellent performance in natural language processing tasks to perform semantic embedding encoding on each configuration file key data item in the set of configuration file key data items. Those of ordinary skill in the art should know that the Bert model is a pre-trained language model, which is based on the Transformer architecture design and captures contextual information in the text through bidirectional training. Unlike traditional unidirectional language models, the Bert model can simultaneously consider the left and right contexts of words, which makes it more effective in understanding sentence structure and semantic information, so that it can more accurately understand the semantic meaning of each key data item in the configuration file and generate a set of semantic embedding encoding vectors of the configuration file key data items.

[0038] In the above-mentioned intelligent review method for the software configuration file data of the motor vehicle inspection and testing equipment, the step S3 performs semantic enhancement on the set of semantic embedding coding vectors of the key data items of the configuration file to obtain a set of semantic embedding enhanced coding vectors of the key data items of the configuration file. It should be understood that, considering that each key data item in the software configuration file of the motor vehicle inspection and testing equipment has different degrees of influence on the sensitive information identification of the configuration file, for example, the device port configuration information may have lower sensitivity than the database connection string. Therefore, in order to further improve the recognition accuracy of sensitive information, the present application proposes a feature enhancement method based on information field effect distribution, which analyzes the information field effect distribution of each key data item in the software configuration file of the motor vehicle inspection and testing equipment to learn the semantic association effect between each key data item, thereby utilizing the contextual association relationship between each key data item to perform feature update and semantic enhancement processing on the set of semantic embedding coding vectors of the key data items of the configuration file, thereby enhancing the sensitive information recognition capability of the model.

[0039] Figure 4 Flow chart of sub-step S3 in the method for intelligent review of software configuration file data of motor vehicle inspection and testing equipment according to an embodiment of the present application. Figure 4As shown, the step S3 includes the steps of: S31, based on the characteristic field effect distribution of the set of semantic embedding coding vectors of the profile key data items, updating the set of semantic embedding coding vectors of the profile key data items to obtain a set of updated semantic embedding coding vectors of the profile key data items; S32, performing bidirectional contextual semantic association enhancement on the set of updated semantic embedding coding vectors of the profile key data items to obtain a set of semantic embedding enhanced coding vectors of the profile key data items.

[0040] Figure 5 Flow chart of sub-step S31 in the method for intelligently reviewing the software configuration file data of the motor vehicle inspection and testing equipment according to the embodiment of the present application. Figure 5 As shown, the step S31 includes the steps of: S311, using a semantic reconstructor to perform feature reconstruction on each semantic embedding coding vector of the profile key data item in the set of semantic embedding coding vectors of the profile key data item to obtain a set of reconstructed semantic embedding coding vectors of the profile key data item; S312, calculating the semantic depth implicit representation value of each reconstructed semantic embedding coding vector of the profile key data item in the set of reconstructed semantic embedding coding vectors of the profile key data item to obtain a set of semantic depth implicit representation values ​​of the key data item; S313, based on the sorting of the set of semantic depth implicit representation values ​​of the key data item, re-sorting the set of semantic embedding coding vectors of the profile key data item to obtain the set of updated semantic embedding coding vectors of the profile key data item.

[0041] More specifically, the step S311 further comprises: using the first weight parameter matrix and the second weight parameter matrix to continuously linearly transform the semantic embedding coding vector of the key data item of the configuration file to obtain the reconstructed semantic embedding coding vector of the key data item of the configuration file. That is, firstly, the semantic reconstructor is used to perform domain transformation on each of the semantic embedding coding vectors of the key data item of the configuration file in the set of the semantic embedding coding vectors of the key data item of the configuration file to unify the feature dimensions and enhance its representation ability in a specific context to generate a set of reconstructed semantic embedding coding vectors of the key data item of the configuration file.

[0042] More specifically, the step S312 further includes: calculating the square of the Euclidean norm of the semantic embedding coding vector of the key data item of the reconstructed configuration file to obtain the initial semantic depth implicit representation value of the key data item; calculating the ratio of the sum of the initial semantic depth implicit representation value of the key data item and the difference between the initial semantic depth implicit representation value of the key data item and the semantic depth representation modulation coefficient; calculating the logarithmic function value of the semantic depth representation modulation coefficient with base 2 to obtain the semantic depth implicit representation value of the key data item. That is, by calculating the semantic depth implicit representation value of each semantic embedding coding vector of the key data item of the reconstructed configuration file, the information density or complexity carried by each key data item in the overall information of the configuration file is measured, and the importance and influence of each key data item in the overall information of the configuration file are quantified.

[0043] More specifically, in step S313, based on the order of the set of semantic depth implicit representation values ​​of the key data items, the set of semantic embedding encoding vectors of the key data items of the configuration file is reordered to obtain the set of semantic embedding encoding vectors of the key data items of the updated configuration file. That is, according to the semantic depth implicit representation values ​​of each key data item, the feature order in the set is further adjusted, and the information structure of the configuration file is reorganized to ensure that the key information is processed preferentially in the subsequent context semantic association enhancement encoding process.

[0044] Specifically, the step S32 further includes: inputting the set of semantic embedding coding vectors of the updated configuration file key data items into the first semantic encoder based on the forward LSTM model to obtain a set of forward key data item contextual semantic coding feature vectors; inputting the set of semantic embedding coding vectors of the updated configuration file key data items into the second semantic encoder based on the backward LSTM model to obtain a set of backward key data item contextual semantic coding feature vectors; and fusing the set of forward key data item contextual semantic coding feature vectors and the set of backward key data item contextual semantic coding feature vectors to obtain a set of configuration file key data item semantic embedding enhanced coding vectors.

[0045] In a specific example of the present application, the set of contextual semantic coding feature vectors of the forward key data items and the set of contextual semantic coding feature vectors of the backward key data items are fused to obtain the set of semantically embedded enhanced coding vectors of the profile key data items, including: using a first weighting parameter and a second weighting parameter to weighted modulate the set of contextual semantic coding feature vectors of the forward key data items and the set of contextual semantic coding feature vectors of the backward key data items respectively to obtain a set of weighted modulated contextual semantic coding feature vectors of the forward key data items and a set of weighted modulated contextual semantic coding feature vectors of the backward key data items; calculating the corresponding points between the set of weighted modulated contextual semantic coding feature vectors of the forward key data items, the set of weighted modulated contextual semantic coding feature vectors of the backward key data items, and the bias item to obtain the set of semantically embedded enhanced coding vectors of the profile key data items.

[0046] Here, the present application adopts a bidirectional long short-term memory (LSTM) model to encode the set of semantic embedding encoding vectors of the updated configuration file key data items from the forward and backward directions respectively. Those skilled in the art should know that the forward LSTM model is a special type of recurrent neural network (RNN), which can fully capture and memorize the long-term dependencies in the time series data based on the gating mechanism, thereby effectively mining the contextual relationships between the key data items of each configuration file, and providing useful information support for subsequent data processing. The backward LSTM model is a variant of the long short-term memory network (LSTM), which is used together with the forward LSTM model when processing sequence data to form a bidirectional LSTM (BiLSTM). In the bidirectional LSTM, the forward LSTM captures the information flow pattern from front to back, while the backward LSTM operates in reverse to identify the dependencies from back to front. This dual-perspective encoding mechanism can capture the global dependencies of configuration file information at different levels, and use this to strengthen the semantic feature expression of each key data item. Then, the bidirectional contextual semantic information of each key data item is integrated by fusing the encoding results of the bidirectional LSTM model to form a richer and more comprehensive set of semantic embedding enhanced encoding vectors of the key data items in the configuration file, thereby achieving a deep understanding of the key information in the configuration file.

[0047] Accordingly, the step S3 includes: processing the set of semantic embedding coding vectors of the key data items of the configuration file with the following feature enhancement formula to obtain a set of semantic embedding enhanced coding vectors of the key data items of the configuration file, wherein the feature enhancement formula is:

[0048]

[0049] in, represents a set of semantic embedding encoding vectors of key data items of the configuration file, , , and They represent the first, second, and third semantic embedding encoding vectors of the key data items of the configuration file, respectively. and The semantic embedding encoding vector of the key data items in the configuration file, is the number of semantic embedding encoding vectors of the key data items of the configuration file, and denote the first weight parameter matrix and the second weight parameter matrix respectively, For the The semantic embedding encoding vector of the key data items of the reconstructed configuration file, represents the calculation of the Euclidean norm of a vector, represents the logarithmic function with base 2, Indicates The semantic depth of key data items implicitly represents the value, represents the set of semantic embedding encoding vectors of key data items in the updated configuration file, , , and respectively represent the first, second, and third in the set of semantic embedding coding vectors of the key data items of the update configuration file and Update the semantic embedding encoding vector of the key data items in the configuration file, represents the forward LSTM model, represents the backward LSTM model, represents the set of contextual semantic encoding feature vectors of forward key data items, Represents the set of contextual semantic encoding feature vectors of backward key data items, and denote the first weighting parameter and the second weighting parameter respectively, is the bias term, A set of semantically enhanced encoding vectors are embedded for key data items of the profile.

[0050] In the above-mentioned intelligent review method for the configuration file data of the motor vehicle inspection and testing equipment software, the step S4 determines whether to generate a configuration file security warning prompt based on the set of semantically embedded enhanced coding vectors of the configuration file key data items. In a specific example of the present application, the step S4 further includes: inputting the semantically embedded enhanced coding vectors of each configuration file key data item in the set of semantically embedded enhanced coding vectors of the configuration file key data items into a classifier-based sensitive information identifier to obtain a recognition result, and the recognition result is used to indicate whether sensitive information exists; in response to the recognition result that sensitive information exists, generating a configuration file security warning prompt.

[0051] That is, a classification algorithm is used to classify each key data item of the configuration file to determine whether there is sensitive information in it. Specifically, the classifier is trained using a large amount of labeled data to continuously adjust its internal parameters to minimize the difference between the predicted results and the actual annotations, so that the correlation and correspondence between the characteristics of the configuration file data items and sensitive information can be effectively learned. In practical applications, the decision rules learned during the training process are used to accurately classify the input configuration file key data item semantic embedding enhanced coding vectors, and the sensitive information identification results of each configuration file key data item are obtained, thereby realizing real-time monitoring and early warning of potential risks in the configuration file, so as to prompt relevant personnel to take corresponding risk control measures.

[0052] In a specific example of the present application, each of the semantic embedding enhanced coding vectors of the profile key data item in the set of the semantic embedding enhanced coding vectors of the profile key data item is respectively input into a classifier-based sensitive information identifier to obtain a recognition result, and the recognition result is used to indicate whether sensitive information exists, further comprising: using the fully connected layer of the sensitive information identifier to fully connect encode the semantic embedding enhanced coding vector of the profile key data item to obtain a semantic embedding fully connected coding feature vector of the profile key data item; inputting the semantic embedding fully connected coding feature vector of the profile key data item into the Softmax classification function of the sensitive information identifier to obtain the probability value of the semantic embedding enhanced coding vector of the profile key data item belonging to each classification label, wherein the classification label includes the presence of sensitive information and the absence of sensitive information; and determining the classification label corresponding to the largest of the probability values ​​as the recognition result.

[0053] In a preferred example of the present application, inputting the semantic embedding enhanced coding vector of the key data item of the profile into a sensitive information identifier based on a classifier to obtain an identification result includes: determining a profile key data item semantic embedding enhanced coding association response matrix and a profile key data item semantic embedding enhanced coding distance response matrix based on the association value and distance value of the eigenvalue of the semantic embedding enhanced coding vector of the key data item of the profile; matrix-multiplying the profile key data item semantic embedding enhanced coding vector with the profile key data item semantic embedding enhanced coding association response matrix to obtain the profile key data item semantic embedding enhanced coding association response vector; matrix-multiplying the profile key data item semantic embedding enhanced coding distance response matrix with the transposed vector of the profile key data item semantic embedding enhanced coding vector to obtain the profile key A data item semantic embedding enhanced coding distance response vector; after dot multiplication of the profile key data item semantic embedding enhanced coding association response matrix and the profile key data item semantic embedding enhanced coding distance response matrix, matrix multiplication with the transpose vector of the profile key data item semantic embedding enhanced coding vector is performed to obtain a profile key data item semantic embedding enhanced coding logic bias vector; the profile key data item semantic embedding enhanced coding association response vector, the profile key data item semantic embedding enhanced coding distance response vector and the profile key data item semantic embedding enhanced coding logic bias vector are dotted to obtain an optimized profile key data item semantic embedding enhanced coding vector; the optimized profile key data item semantic embedding enhanced coding vector is input into a classifier-based sensitive information identifier to obtain the identification result.

[0054] Accordingly, the optimization process of the semantic embedding enhanced coding vector of the key data items of the configuration file is represented by the following optimization formula:

[0055]

[0056] in, The semantic embedding enhancement encoding associated response matrix for the key data items of the profile, Encoding distance response matrix for semantic embedding enhancement of key data items in profiles, The key data items of the configuration file are semantically embedded in the enhanced coding association response matrix The eigenvalues ​​of the positions, The key data items of the configuration file are semantically embedded in the enhanced coding distance response matrix The eigenvalues ​​of the positions, and are the first and eigenvalues, The point-to-point vector between the profile key data item semantic embedding enhanced coding association response vector and the profile key data item semantic embedding enhanced coding distance response vector is obtained. Enhance the encoding logic bias vector for the semantic embedding of key data items in the configuration file, is the set of real numbers, The length of the encoding vector enhanced by the semantic embedding of the key data items in the configuration file, It means point multiplication by position. represents matrix multiplication, Indicates adding by position point. Enhanced encoding vectors for semantic embedding of key data items in configuration files, Enhanced encoding vectors for optimized semantic embedding of profile key data items.

[0057] That is, in the case where each of the semantic embedding coding vectors of the profile key data item in the set of semantic embedding coding vectors of the profile key data item respectively represents the textual semantic embedding coding features of each of the profile key data item in the set of profile key data items, when the set of semantic embedding coding vectors of the profile key data item is input into the semantic feature enhancement based on field effect distribution, the sparsity of the semantic distribution between each of the profile key data items in the set of profile key data items will make the semantic feature enhancement distribution sparse, resulting in large-scale simple repetition of the set of semantic embedding enhancement coding vectors of the profile key data item, affecting the logical dependency of the semantic enhancement, and reducing the accuracy of the recognition result based on the semantic association enhancement distribution.

[0058] Therefore, by taking the self-association matrix and self-distance matrix of the semantic embedding enhanced coding vector of the key data item in the profile as the statistical reference-free distribution response framework of the semantic embedding enhanced coding vector of the key data item in the profile, a retrieval-enhanced reverse response is constructed based on the semantic embedding enhanced coding vector of the key data item in the profile to avoid simple repetition of feature distribution, and by ensuring the intrinsic mapping logic of the semantic embedding enhanced coding vector of the key data item in the profile based on the retrieval-response context to avoid the surface combinatorial mapping of the semantic embedding enhanced coding vector of the key data item in the profile, so as to achieve the logical dependency mapping of the semantic embedding enhanced coding vector of the key data item in the profile to the classification target domain while maintaining the intuitive response relevance, thereby improving the accuracy of the recognition result obtained by inputting the semantic embedding enhanced coding vector of the key data item in the profile into the sensitive information identifier based on the classifier.

[0059] In summary, the intelligent review method for the configuration file data of the motor vehicle inspection and testing equipment software based on the embodiment of the present application is explained, which uses natural language processing technology based on deep learning to process the configuration file of the motor vehicle inspection and testing equipment software, and extracts the semantic feature representation of each key data item in the configuration file through semantic embedding coding technology, and then, based on the contextual semantic association relationship between each key data item, each key data item is updated with features and semantic enhancement processing to achieve an in-depth understanding of the configuration file, thereby realizing intelligent identification and early warning prompts of sensitive information in the configuration file on this basis. In this way, the accuracy and efficiency of the configuration file review can be effectively improved, the workload of manual review can be reduced, and the safety risks caused by human negligence can be reduced.

[0060] The basic principle of the present invention is described above in conjunction with specific embodiments. However, it should be pointed out that the advantages, strengths, effects, etc. mentioned in the present invention are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. must be possessed by each embodiment of the present invention. In addition, the specific details of the above embodiments are only for the purpose of illustration and facilitation of understanding, rather than limitation, and the above details do not limit the present invention to being implemented by adopting the above specific details.

[0061] In the above embodiments, the description of each embodiment has its own emphasis. For the parts that are not described or recorded in detail in a certain embodiment, please refer to the relevant description of other embodiments. In the several embodiments provided by the present invention, it should be understood that the disclosed system and method can be implemented in other ways. For example, the system embodiment described above is only schematic. For example, the unit division is only a logical function division, and there may be other division methods in actual implementation. The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0062] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference to a figure in a claim should not be considered as limiting the claim to which it relates.

[0063] In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. The multiple units stated in the system claims can also be implemented by one unit through software or hardware.

[0064] Finally, it should be noted that the above description has been given for the purpose of illustration and description. In addition, the above embodiments are only used to illustrate the technical solution of the present invention and are not limited thereto. Although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for intelligent review of software configuration file data of motor vehicle inspection and testing equipment, characterized in that: include: Obtain software configuration files for motor vehicle inspection and testing equipment; Performing semantic embedding coding on each key data item in the motor vehicle inspection and testing equipment software configuration file to obtain a set of semantic embedding coding vectors of the key data items in the configuration file; Performing semantic enhancement on the set of semantic embedding coding vectors of the key data items of the configuration file to obtain a set of semantic embedding enhanced coding vectors of the key data items of the configuration file; Determining whether to generate a configuration file security warning prompt based on the set of semantically embedded enhanced coding vectors of the configuration file key data items; Among them, semantic enhancement is performed on the set of semantic embedding coding vectors of the key data items of the configuration file, including: based on the characteristic field effect distribution of the set of semantic embedding coding vectors of the key data items of the configuration file, updating the set of semantic embedding coding vectors of the key data items of the configuration file to obtain an updated set of semantic embedding coding vectors of the key data items of the configuration file; performing bidirectional contextual semantic association enhancement on the set of updated semantic embedding coding vectors of the key data items of the configuration file to obtain the set of semantic embedding enhanced coding vectors of the key data items of the configuration file.

2. The method for intelligent review of software configuration file data of motor vehicle inspection and testing equipment according to claim 1 is characterized in that: The key data items in the motor vehicle inspection and testing equipment software configuration file are semantically embedded and encoded to obtain a set of semantically embedded encoding vectors of the key data items in the configuration file, including: Extracting a set of key data items of the configuration file from the motor vehicle inspection and testing equipment software configuration file; A key item semantic embedding encoder based on the Bert model is used to perform semantic embedding encoding on each configuration file key data item in the set of configuration file key data items to obtain a set of configuration file key data item semantic embedding encoding vectors.

3. The method for intelligent review of software configuration file data of motor vehicle inspection and testing equipment according to claim 2 is characterized in that: Based on the characteristic field effect distribution of the set of semantic embedding coding vectors of the key data items of the configuration file, the set of semantic embedding coding vectors of the key data items of the configuration file is updated to obtain a set of updated semantic embedding coding vectors of the key data items of the configuration file, including: Using a semantic reconstructor, feature-reconstructing each of the configuration file key data item semantic embedding coding vectors in the set of configuration file key data item semantic embedding coding vectors to obtain a set of reconstructed configuration file key data item semantic embedding coding vectors; Calculating the semantic depth implicit representation value of each reconstructed configuration file key data item semantic embedding coding vector in the set of the reconstructed configuration file key data item semantic embedding coding vectors to obtain a set of key data item semantic depth implicit representation values; Based on the sorting of the set of key data item semantic depth implicit representation values, the set of configuration file key data item semantic embedding coding vectors is re-sorted to obtain the set of updated configuration file key data item semantic embedding coding vectors.

4. The method for intelligent review of software configuration file data of motor vehicle inspection and testing equipment according to claim 3 is characterized in that: Using a semantic reconstructor to perform feature reconstruction on each of the configuration file key data item semantic embedding coding vectors in the set of configuration file key data item semantic embedding coding vectors to obtain a set of reconstructed configuration file key data item semantic embedding coding vectors, including: The configuration file key data item semantic embedding coding vector is continuously linearly transformed using the first weight parameter matrix and the second weight parameter matrix to obtain the reconstructed configuration file key data item semantic embedding coding vector.

5. The method for intelligent review of software configuration file data of motor vehicle inspection and testing equipment according to claim 4 is characterized in that: Calculating the semantic depth implicit representation value of each reconstructed configuration file key data item semantic embedding encoding vector in the set of the reconstructed configuration file key data item semantic embedding encoding vectors to obtain a set of key data item semantic depth implicit representation values, including: Calculating the square of the Euclidean norm of the semantic embedding encoding vector of the key data item of the reconstructed configuration file to obtain an initial semantic depth implicit representation value of the key data item; Calculating a ratio of a sum of an initial semantic depth implicit representation value of the key data item and a difference between an initial semantic depth implicit representation value of the key data item to obtain a semantic depth representation modulation coefficient; The logarithmic function value with base 2 of the semantic depth representation modulation coefficient is calculated to obtain the semantic depth implicit representation value of the key data item.

6. The method for intelligent review of software configuration file data of motor vehicle inspection and testing equipment according to claim 5 is characterized in that: Performing bidirectional context semantic association enhancement on the set of semantic embedding coding vectors of the key data items of the updated configuration file to obtain a set of semantic embedding enhanced coding vectors of the key data items of the configuration file, including: Inputting the set of semantic embedding encoding vectors of the key data items of the update configuration file into a first semantic encoder based on a forward LSTM model to obtain a set of forward key data item context semantic encoding feature vectors; Inputting the set of semantic embedding encoding vectors of the key data items of the updated configuration file into a second semantic encoder based on a backward LSTM model to obtain a set of backward key data item context semantic encoding feature vectors; The set of forward key data item context semantic encoding feature vectors and the set of backward key data item context semantic encoding feature vectors are fused to obtain the set of configuration file key data item semantic embedding enhanced encoding vectors.

7. The method for intelligent review of software configuration file data of motor vehicle inspection and testing equipment according to claim 6 is characterized in that: The set of forward key data item context semantic encoding feature vectors and the set of backward key data item context semantic encoding feature vectors are merged to obtain the set of configuration file key data item semantic embedding enhanced encoding vectors, including: Using a first weighting parameter and a second weighting parameter, respectively, the set of the forward key data item context semantic coding feature vectors and the set of the backward key data item context semantic coding feature vectors are weighted modulated to obtain a set of weighted modulated forward key data item context semantic coding feature vectors and a set of weighted modulated backward key data item context semantic coding feature vectors; The corresponding points between the set of weighted modulated forward key data item context semantic coding feature vectors, the set of weighted modulated backward key data item context semantic coding feature vectors and bias items are calculated to obtain the set of configuration file key data item semantic embedding enhanced coding vectors.

8. The method for intelligent review of software configuration file data of motor vehicle inspection and testing equipment according to claim 7 is characterized in that: Determining whether to generate a configuration file security warning prompt based on a set of semantically embedded enhanced coding vectors of key data items of the configuration file includes: Inputting each of the configuration file key data item semantic embedding enhanced coding vectors in the set of the configuration file key data item semantic embedding enhanced coding vectors into a classifier-based sensitive information identifier to obtain a recognition result, wherein the recognition result is used to indicate whether sensitive information exists; In response to the identification result that sensitive information exists, a configuration file security warning prompt is generated.

9. The method for intelligent review of software configuration file data of motor vehicle inspection and testing equipment according to claim 8, characterized in that: Inputting each configuration file key data item semantic embedding enhanced coding vector in the set of configuration file key data item semantic embedding enhanced coding vectors into the classifier-based sensitive information identifier to obtain a recognition result, wherein the recognition result is used to indicate whether sensitive information exists, including: Using the fully connected layer of the sensitive information identifier, the semantic embedding enhanced encoding vector of the key data item of the configuration file is fully connected encoded to obtain a semantic embedding fully connected encoding feature vector of the key data item of the configuration file; Inputting the semantic embedding fully connected encoding feature vector of the key data item of the configuration file into the Softmax classification function of the sensitive information identifier to obtain the probability value of the semantic embedding enhanced encoding vector of the key data item of the configuration file belonging to each classification label, wherein the classification label includes the presence of sensitive information and the absence of sensitive information; The classification label corresponding to the largest probability value among the probability values ​​is determined as the recognition result.

Images