Intelligent matching method for compliance of motor vehicle inspection and detection equipment software

Abstract

The invention relates to the technical field of intelligent detection, and particularly discloses a method for intelligently matching software compliance of motor vehicle inspection and detection equipment, which comprises the following steps of: firstly, constructing a software compliance inspection database based on a standard file; carrying out semantic analysis on each compliance test item in the database and the software information of the motor vehicle inspection and detection equipment by adopting a natural language processing technology based on deep learning, and respectively extracting semantic feature representations of the equipment software information and each compliance test item; and furthermore, the equipment software information is inquired and matched with each compliance test item, so that compliance detection of the motor vehicle inspection and detection equipment software is intelligently realized. Therefore, the efficiency and the accuracy of compliance inspection can be greatly improved, the subjectivity and the error of manual auditing are reduced, the updating and the change of a software version are better adapted, and the software of the motor vehicle inspection and detection equipment is ensured to accord with the latest specified standard.

Description

Technical Field

[0001] The present application relates to the field of intelligent detection technology, and more specifically, to a method for intelligently matching the software compliance of motor vehicle inspection and testing equipment. Background Art

[0002] With the rapid growth of the number of motor vehicles, the safety and environmental performance of motor vehicles have become a matter of widespread concern in society. In order to ensure that vehicles comply with relevant national and local laws and regulations, motor vehicle inspection and testing has become one of the important means to ensure traffic safety and environmental protection. In the process of motor vehicle inspection and testing, the accuracy and reliability of the inspection and testing equipment are the key to ensuring the validity of the test results. Among them, the software system of the inspection and testing equipment is an important factor in ensuring the normal operation of its functions.

[0003] Traditional compliance checks on motor vehicle inspection and testing equipment software mainly rely on manual review, that is, professional technicians evaluate the functions and performance indicators of the equipment software one by one according to relevant standards and specifications. However, this method is not only time-consuming and laborious, but also easily affected by personal experience and technical level, and it is difficult to ensure the consistency and accuracy of the review. Especially when faced with a large number of complex and changing software versions and updates, the traditional manual review method is inefficient and cannot meet the rapidly developing market needs.

[0004] Therefore, an intelligent matching method for the software compliance of motor vehicle inspection and testing equipment is expected. Summary of the invention

[0005] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides a method for intelligent matching of the compliance of the software of motor vehicle inspection and testing equipment, which first builds a software compliance inspection database based on standard files, and uses natural language processing technology based on deep learning to perform semantic analysis on each compliance inspection item in the database and the software information of the motor vehicle inspection and testing equipment, and extracts the semantic feature representation of the equipment software information and each compliance inspection item respectively, and then, by querying and matching the equipment software information with each compliance inspection item, the compliance inspection of the motor vehicle inspection and testing equipment software is intelligently realized. In this way, the efficiency and accuracy of compliance inspection can be greatly improved, the subjectivity and errors of manual review can be reduced, the software version can be better adapted to the update and change, and the motor vehicle inspection and testing equipment software can be ensured to meet the latest prescribed standards.

[0006] Accordingly, according to one aspect of the present application, a method for intelligently matching software compliance of motor vehicle inspection and testing equipment is provided, which includes: Based on the standard file, construct a software compliance verification database, wherein the software compliance verification database includes a natural language description of multiple compliance verification items; Extracting information description of motor vehicle inspection and testing equipment software, including functions, safety and data processing capabilities; Semantically encoding the information description of the motor vehicle inspection and testing equipment software and the natural language description of each compliance inspection item in the software compliance inspection database to obtain a set of motor vehicle inspection and testing equipment software information description semantic encoding vectors and compliance inspection item semantic embedding encoding vectors; Performing query matching on the set of the motor vehicle inspection and testing equipment software information description semantic coding vector and the compliance inspection item semantic embedding coding vector to obtain the to-be-inspected equipment software information-compliance inspection item semantic query response coding vector; Based on the software information of the equipment to be inspected-compliance inspection item semantic query response coding vector, determine whether the motor vehicle inspection and testing equipment software is compliant.

[0007] In the above-mentioned intelligent matching method for compliance of software of motor vehicle inspection and testing equipment, query matching is performed on the set of semantic coding vectors describing the software information of the motor vehicle inspection and testing equipment and the set of semantic embedded coding vectors of the compliance inspection items, including: extracting the significant semantic features of the set of semantic embedded coding vectors of the compliance inspection items as a prompt template; based on the prompt template, cross-domain optimized query coding based on significant semantic guidance is performed on the semantic coding vectors describing the software information of the motor vehicle inspection and testing equipment and the set of semantic embedded coding vectors of the compliance inspection items to obtain the semantic query response coding vector of the equipment software information to be inspected-compliance inspection items.

[0008] In the above-mentioned intelligent matching method for compliance of motor vehicle inspection and testing equipment software, the information description of the motor vehicle inspection and testing equipment software and the natural language description of each compliance inspection item in the software compliance inspection database are semantically encoded respectively to obtain a set of motor vehicle inspection and testing equipment software information description semantic encoding vectors and compliance inspection item semantic embedding encoding vectors, including: the information description of the motor vehicle inspection and testing equipment software and the natural language description of each compliance inspection item in the software compliance inspection database are respectively input into a semantic encoder based on the Bert model to obtain a set of motor vehicle inspection and testing equipment software information description semantic encoding vectors and compliance inspection item semantic embedding encoding vectors.

[0009] In the above-mentioned intelligent matching method for software compliance of motor vehicle inspection and testing equipment, the significant semantic features of the set of semantic embedding coding vectors of the compliance inspection items are extracted as the prompt template, including: constructing a key matrix based on the set of semantic embedding coding vectors of the compliance inspection items; extracting the maximum value of each key vector in the key matrix to obtain the significant semantic feature vector of the compliance inspection item as the prompt template.

[0010] In the above-mentioned intelligent matching method for software compliance of motor vehicle inspection and testing equipment, a key matrix is ​​constructed based on the set of semantic embedding coding vectors of compliance inspection items, including: using the key embedding matrix to linearly embed code each compliance inspection item semantic embedding coding vector in the set of semantic embedding coding vectors of compliance inspection items to obtain a set of semantic embedding coding vectors of compliance inspection items after linear transformation; using the semantic embedding coding vectors of compliance inspection items after linear transformation as the key vectors, and arranging the set of semantic embedding coding vectors of compliance inspection items after linear transformation into a matrix to obtain the key matrix.

[0011] In the above-mentioned intelligent matching method for compliance of software of motor vehicle inspection and testing equipment, based on the prompt template, the set of semantic coding vectors of the motor vehicle inspection and testing equipment software information description and the semantic embedding coding vectors of the compliance inspection items are subjected to cross-domain optimized query encoding based on significant semantic guidance to obtain the semantic query response coding vector of the software information of the equipment to be inspected-compliance inspection items, including: using the query embedding matrix and the value embedding matrix to perform linear embedding encoding on the semantic coding vectors of the software information description of the motor vehicle inspection and testing equipment respectively to obtain the value vector and the query vector; the query vector, the value vector, each key vector in the key matrix and the prompt template are input into the heterogeneous converter structure based on template prompt optimization to obtain the set of cross-domain query coding feature vectors of the software information of the equipment to be inspected-compliance inspection items; the set of cross-domain query coding feature vectors of the software information of the equipment to be inspected-compliance inspection items are subjected to query feature aggregation to obtain the semantic query response coding vector of the software information of the equipment to be inspected-compliance inspection items.

[0012] In the above-mentioned intelligent matching method for software compliance of motor vehicle inspection and testing equipment, the query vector, the value vector, each key vector in the key matrix and the prompt template are respectively input into a heterogeneous converter structure based on template prompt optimization to obtain a set of cross-domain query encoding feature vectors of the software information of the equipment to be inspected-compliance inspection items, including: multiplying the query vector by the transposed vector of the key vector and then dividing it by the bi-norm of the prompt template to obtain an attention score matrix; multiplying the attention score matrix by the prompt template after passing it through a softmax function to obtain a template prompt optimized attention weight vector; calculating the position point multiplication between the template prompt optimized attention weight vector and the value vector to obtain the cross-domain query encoding feature vector of the software information of the equipment to be inspected-compliance inspection items.

[0013] In the above-mentioned intelligent matching method for software compliance of motor vehicle inspection and testing equipment, query feature aggregation is performed on the set of cross-domain query coding feature vectors of the software information of the equipment to be inspected-compliance inspection items to obtain the semantic query response coding vector of the software information of the equipment to be inspected-compliance inspection items, including: calculating the positional mean vector of the set of cross-domain query coding feature vectors of the software information of the equipment to be inspected-compliance inspection items to obtain the semantic query response coding vector of the software information of the equipment to be inspected-compliance inspection items.

[0014] In the above-mentioned intelligent matching method for compliance of motor vehicle inspection and testing equipment software, whether the motor vehicle inspection and testing equipment software is compliant is determined based on the semantic query response coding vector of the equipment software information to be inspected-compliance inspection item, including: passing the semantic query response coding vector of the equipment software information to be inspected-compliance inspection item through a classifier-based compliance intelligent matching module to determine whether the motor vehicle inspection and testing equipment software is compliant.

[0015] Compared with the prior art, the intelligent matching method for compliance of motor vehicle inspection and testing equipment software provided by the present application first constructs a software compliance inspection database based on standard files, and uses natural language processing technology based on deep learning to perform semantic analysis on each compliance inspection item in the database and the motor vehicle inspection and testing equipment software information, and extracts the semantic feature representation of the equipment software information and each compliance inspection item respectively, and then, by querying and matching the equipment software information with each compliance inspection item, the compliance inspection of the motor vehicle inspection and testing equipment software is intelligently realized. In this way, the efficiency and accuracy of compliance inspection can be greatly improved, the subjectivity and errors of manual review can be reduced, the software version can be better adapted to updates and changes, and the motor vehicle inspection and testing equipment software can be ensured to meet the latest prescribed standards. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] 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.

[0017] Figure 1 It is a flowchart of a method for intelligently matching software compliance of motor vehicle inspection and testing equipment according to an embodiment of the present application.

[0018] Figure 2 This is a data flow diagram of a method for intelligently matching software compliance of motor vehicle inspection and testing equipment according to an embodiment of the present application.

[0019] Figure 3 This is a flowchart of step S140 in the method for intelligent matching of software compliance of motor vehicle inspection and testing equipment according to an embodiment of the present application.

[0020] Figure 4 This is a flowchart of sub-step S142 in the method for intelligent matching of software compliance of motor vehicle inspection and testing equipment according to an embodiment of the present application. DETAILED DESCRIPTION

[0021] Below, the embodiments of the present application will be described in more detail in conjunction with the accompanying drawings, and the above and other purposes, features and advantages of the present application will become more apparent. 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 example embodiments described herein.

[0022] As mentioned in the above background technology, the compliance check of traditional motor vehicle inspection and testing equipment software mainly relies on manual review, which is not only time-consuming and laborious, but also easily affected by personal experience and technical level, and it is difficult to ensure the consistency and accuracy of the review. Especially in the face of a large number of complex and changeable software versions and updates, the traditional manual review method is not only inefficient, but also unable to meet the rapidly developing market demand. In view of the above technical problems, the technical concept of this application is: first, a software compliance inspection database is constructed based on standard files, and the natural language processing technology based on deep learning is used to perform semantic analysis on each compliance inspection item and motor vehicle inspection and testing equipment software information in the database, and the semantic feature representation of the device software information and each compliance inspection item is extracted respectively, and then, by querying and matching the device software information with each compliance inspection item, the compliance inspection of the motor vehicle inspection and testing equipment software is intelligently realized. In this way, the efficiency and accuracy of compliance inspection can be greatly improved, the subjectivity and errors of manual review can be reduced, the software version can be better adapted to the update and change, and the motor vehicle inspection and testing equipment software can be ensured to meet the latest prescribed standards.

[0023] Figure 1 It is a flowchart of a method for intelligently matching software compliance 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 matching the compliance of software of a motor vehicle inspection and testing device according to an embodiment of the present application. Figure 1 and Figure 2As shown, according to the embodiment of the present application, the method for intelligent matching of software compliance of motor vehicle inspection and testing equipment includes the following steps: S110, constructing a software compliance verification database based on standard files, wherein the software compliance verification database includes natural language descriptions of multiple compliance verification items; S120, extracting information descriptions of motor vehicle inspection and testing equipment software, wherein the information descriptions include functions, security, and data processing capabilities; S130, semantically encoding the information descriptions of the motor vehicle inspection and testing equipment software and the natural language descriptions of each compliance verification item in the software compliance verification database to obtain a set of motor vehicle inspection and testing equipment software information description semantic coding vectors and compliance verification item semantic embedded coding vectors; S140, query matching the motor vehicle inspection and testing equipment software information description semantic coding vectors and the compliance verification item semantic embedded coding vectors to obtain a device software information to be inspected-compliance verification item semantic query response coding vector; S150, determining whether the motor vehicle inspection and testing equipment software is compliant based on the device software information to be inspected-compliance verification item semantic query response coding vector.

[0024] In the above-mentioned intelligent matching method for compliance of software of motor vehicle inspection and testing equipment, the step S110 constructs a software compliance inspection database based on the standard file, wherein the software compliance inspection database includes a natural language description of multiple compliance inspection items. It should be understood that the standard file is formulated by an authoritative agency or industry organization, represents the best practices or minimum requirements recognized in the industry, and provides a clear benchmark for compliance inspection of motor vehicle inspection and testing equipment software. By constructing a software compliance inspection database based on the standard file, it can be ensured that all compliance inspection items comply with the authoritative standards, which helps to ensure the quality and safety of motor vehicle inspection and testing equipment.

[0025] In the above-mentioned intelligent matching method for compliance of motor vehicle inspection and testing equipment software, the step S120 extracts the information description of the motor vehicle inspection and testing equipment software, and the information description includes functions, security and data processing capabilities. That is, in order to comprehensively and accurately evaluate whether the equipment software meets the compliance requirements, the present application further extracts the information description of the motor vehicle inspection and testing equipment software to obtain the functions, security and data processing capabilities of the equipment software, for example, whether the equipment software has the necessary encryption measures to protect data security, what specific types of data can be processed, etc., so as to achieve a comprehensive understanding of the performance of the equipment software and provide detailed data support for subsequent compliance matching detection.

[0026] In the above-mentioned intelligent matching method for compliance of motor vehicle inspection and testing equipment software, the step S130, the information description of the motor vehicle inspection and testing equipment software and the natural language description of each compliance inspection item in the software compliance inspection database are semantically encoded to obtain a set of motor vehicle inspection and testing equipment software information description semantic encoding vectors and compliance inspection item semantic embedding encoding vectors. In a specific example of the present application, the step S130 includes: the information description of the motor vehicle inspection and testing equipment software and the natural language description of each compliance inspection item in the software compliance inspection database are respectively input into a semantic encoder based on the Bert model to obtain a set of motor vehicle inspection and testing equipment software information description semantic encoding vectors and compliance inspection item semantic embedding encoding vectors. It should be understood that by matching and analyzing the information description of the device software and each compliance inspection item, it can be determined whether the device software meets the compliance requirements. However, considering that the traditional keyword-based matching method may produce misjudgments due to vocabulary differences or inconsistent expressions. Therefore, the present application adopts a semantic encoder based on the Bert model to semantically embed the information description of the motor vehicle inspection and testing equipment software and the natural language description of each compliance inspection item. Specifically, the Bert model is based on a bidirectional Transformer architecture, which can deeply understand the contextual information of each word in the text, effectively capture the contextual semantic features in the text, and convert device software information and compliance check items into vector representations in a high-dimensional space, so as to facilitate query matching between the two at the semantic level, thereby reducing misjudgments caused by vocabulary differences and improving the accuracy and robustness of compliance detection.

[0027] In the above-mentioned intelligent matching method for compliance of motor vehicle inspection and testing equipment software, the step S140 performs query matching on the set of the motor vehicle inspection and testing equipment software information description semantic coding vector and the compliance inspection item semantic embedding coding vector to obtain the device software information to be inspected-compliance inspection item semantic query response coding vector. In the technical solution of the present application, in order to further improve the semantic query matching accuracy between the device software information and each compliance inspection item, a query matching method based on significant semantic guidance is proposed, which optimizes the query encoding process by utilizing the overall critical semantic information of the set of compliance inspection item semantic embedding coding vectors, thereby enhancing the pertinence and accuracy of semantic query matching.

[0028] Figure 3 FIG. 1 is a flowchart of step S140 in the method for intelligently matching the compliance of software of a motor vehicle inspection and testing device according to an embodiment of the present application. Figure 3As shown, the step S140 includes: S141, extracting the significant semantic features of the set of semantic embedded coding vectors of the compliance check items as a prompt template; S142, based on the prompt template, performing cross-domain optimized query coding based on significant semantic guidance on the semantic coding vector of the software information description of the motor vehicle inspection and testing equipment and the set of semantic embedded coding vectors of the compliance check items to obtain the semantic query response coding vector of the software information of the equipment to be inspected-compliance check items.

[0029] Specifically, the step S141 includes: constructing a key matrix based on the set of semantic embedding coding vectors of the compliance check items; extracting the maximum value of each key vector in the key matrix to obtain a significant semantic feature vector of the compliance check item as the prompt template. Wherein, constructing a key matrix based on the set of semantic embedding coding vectors of the compliance check items includes: using the key embedding matrix to linearly embed code each semantic embedding coding vector of the compliance check items in the set of semantic embedding coding vectors of the compliance check items to obtain a set of semantic embedding coding vectors of the compliance check items after linear transformation; using the semantic embedding coding vectors of the compliance check items after linear transformation as the key vectors, and performing matrix arrangement on the set of semantic embedding coding vectors of the compliance check items after linear transformation to obtain the key matrix.

[0030] Figure 4 FIG. 1 is a flowchart of sub-step S142 in the method for intelligently matching the compliance of software of a motor vehicle inspection and testing device according to an embodiment of the present application. Figure 4 As shown, the step S142 includes: S1421, using the query embedding matrix and the value embedding matrix to perform linear embedding encoding on the semantic coding vector of the motor vehicle inspection and testing equipment software information description respectively to obtain a value vector and a query vector; S1422, inputting the query vector, the value vector, each key vector in the key matrix and the prompt template into a heterogeneous converter structure optimized based on template prompts to obtain a set of cross-domain query coding feature vectors of the software information of the equipment to be inspected-compliance inspection items; S1423, performing query feature aggregation on the set of cross-domain query coding feature vectors of the software information of the equipment to be inspected-compliance inspection items to obtain the semantic query response coding vector of the software information of the equipment to be inspected-compliance inspection items.

[0031] More specifically, the step S1422 includes: multiplying the query vector by the transposed vector of the key vector and then dividing it by the binary norm of the prompt template to obtain an attention score matrix; passing the attention score matrix through a softmax function and then multiplying it by the prompt template to obtain a template prompt optimized attention weight vector; calculating the position point multiplication between the template prompt optimized attention weight vector and the value vector to obtain the cross-domain query encoding feature vector of the device software information to be inspected-compliance inspection item.

[0032] More specifically, the step S1423 includes: calculating the positional mean vector of the set of cross-domain query encoding feature vectors of the device software information to be inspected-compliance inspection items to obtain the semantic query response encoding vector of the device software information to be inspected-compliance inspection items.

[0033] That is, the set of the compliance inspection item semantic embedding coding vectors and the motor vehicle inspection and testing equipment software information description semantic coding vector are processed by the following query matching formula to obtain the to-be-inspected equipment software information-compliance inspection item semantic query response coding vector, wherein the query matching formula is:

[0034] in, represents a set of semantic embedding encoding vectors of the compliance check items, , , and are respectively the first, second, and third in the set of semantic embedding encoding vectors of the compliance check item. and Semantic embedding encoding vectors of compliance check items, The value of is the number of semantic embedding encoding vectors of the compliance check item, , and denote the key embedding matrix, query embedding matrix and value embedding matrix respectively, , and Represent different bias terms, represents the key matrix, , , and are the first, second, and and key vector, i.e., the semantic embedding encoding vector of the compliance check item after the linear transformation, represents the maximum value function, represents the prompt template, represents the semantic coding vector of the software information description of the motor vehicle inspection and testing equipment, and denote the query vector and value vector respectively, represents the transpose of a vector, represents the two-norm of a vector, is the normalized exponential function, represents the matrix multiplication operation, It means point multiplication by position. The first one in the set of cross-domain query encoding feature vectors representing the device software information to be checked-compliance check item The software information of the device to be tested - the cross-domain query encoding feature vector of the compliance test items, Represents the semantic query response encoding vector of the device software information to be inspected-compliance inspection item.

[0035] Specifically, first, each compliance check item semantic embedding coding vector in the set of compliance check item semantic embedding coding vectors is linearly transformed to map it to a new latent feature space so that it is suitable for the subsequent query matching process. Then, each compliance check item semantic embedding coding vector after linear transformation is used as a key vector and arranged into a key matrix to facilitate subsequent parallel processing. Then, the maximum value of each key vector is selected from the key matrix to combine into a compliance check item significant semantic feature vector to capture important compliance check information, and use it as a prompt template to guide the subsequent query matching process to focus more on the key semantic feature part of each compliance check item to reduce noise interference. Then, the query embedding matrix and the value embedding matrix are respectively applied to the semantic encoding vector of the motor vehicle inspection and testing equipment software information description to generate the corresponding query vector and value vector, and the query vector, the value vector, each key vector in the key matrix and the prompt template are interactively encoded through a specially designed heterogeneous converter structure. By introducing additional prompt templates in the traditional attention mechanism to automatically adjust the weight distribution strategy, not only the semantic relevance between each compliance inspection item and the equipment software information is considered, but also the global contextual semantic structure of each compliance inspection item in the database is considered, which helps to generate more accurate and relevant query outputs. Finally, the cross-domain optimized query matching results of the equipment software information and each compliance inspection item are fused by position mean to integrate the query matching information of all compliance inspection items, and generate the equipment software information to be inspected-compliance inspection item semantic query response encoding vector, thereby realizing a comprehensive matching evaluation of the compliance of the motor vehicle inspection and testing equipment software.

[0036] In the above-mentioned intelligent matching method for compliance of motor vehicle inspection and testing equipment software, the step S150 determines whether the motor vehicle inspection and testing equipment software is compliant based on the semantic query response coding vector of the device software information to be inspected-compliance inspection item. In a specific example of the present application, the step S150 includes: passing the device software information to be inspected-compliance inspection item semantic query response coding vector through a compliance intelligent matching module based on a classifier to determine whether the motor vehicle inspection and testing equipment software is compliant. It should be understood that after the above processing, the device software information to be inspected-compliance inspection item semantic query response coding vector comprehensively reflects the degree of semantic association between the device software and each compliance inspection item. Here, the classifier performs feature learning on the device software information to be inspected-compliance inspection item semantic query response coding vector, and can judge whether the device software meets the compliance requirements according to the semantic query matching features contained therein, thereby mapping it to the corresponding category label and outputting the matching result. In this way, instant feedback can be provided for the compliance of the motor vehicle inspection and testing equipment software, greatly shortening the detection cycle and improving work efficiency.

[0037] Taking into account that the set of the motor vehicle inspection and testing equipment software information description semantic coding vector and the compliance inspection item semantic embedding coding vector respectively represent the text semantic coding features of the information description of the motor vehicle inspection and testing equipment software and the text semantic coding features of each compliance inspection item in the software compliance inspection database, when performing query encoding with the key matrix significant features as the prompt template, the difference in text semantic encoding dimensions caused by the source content difference will make the semantic query high-dimensional representation of the semantic query response coding vector of the equipment software information to be inspected-compliance inspection item have a complex spatial structure. Therefore, when the equipment software information to be inspected-compliance inspection item semantic query response coding vector is input into the compliance intelligent matching module based on the classifier for classification probability regression, it will affect the accuracy of the classification result.

[0038] Preferably, in an example of the present application, when the device software information to be inspected-compliance inspection item semantic query response encoding vector passes through a classifier-based compliance intelligent matching module, the device software information to be inspected-compliance inspection item semantic query response encoding vector is optimized, and the optimization process specifically includes: Determine the mean and variance of the feature set composed of the feature values ​​of the semantic query response encoding vector of the software information of the device to be inspected-compliance inspection item, and divide the mean by the variance to obtain the software information-compliance inspection item semantic query response encoding probability weight value:

[0039] in, and respectively represent the mean and variance of the feature set composed of the feature values ​​of the semantic query response encoding vector of the device software information to be inspected-compliance inspection item, Represents the software information-compliance inspection item semantic query response encoding probability weight value; The software information-compliance check item semantic query response encoding vector of the device to be inspected is multiplied by the inverse of the maximum eigenvalue of the semantic query response encoding vector of the device to be inspected software information-compliance check item to obtain the software information-compliance check item semantic query response encoding probability constraint vector:

[0040] in, represents the semantic query response encoding vector of the device software information to be inspected-compliance inspection item, represents the maximum eigenvalue in the semantic query response encoding vector of the device software information to be inspected-compliance inspection item, It means point multiplication by position. Represents software information-compliance check item semantic query response encoding probability constraint vector; The software information-compliance check item semantic query response encoding probability constraint vector is added to the software information-compliance check item semantic query response encoding probability weight value, and the base 2 logarithm is calculated to obtain the software information-compliance check item semantic query response encoding information interaction vector:

[0041] in, Represents software information-compliance check item semantic query response encoding probability constraint vector, Represents the probability weight value of the software information-compliance inspection item semantic query response encoding, Indicates adding by position point. Represents software information-compliance check item semantic query response encoding information interaction vector; After subtracting the software information-compliance check item semantic query response encoding probability constraint vector from the software information-compliance check item semantic query response encoding probability weight value, the inverse of each eigenvalue is calculated to obtain the software information-compliance check item semantic query response encoding sequence constraint vector:

[0042] in, Represents software information-compliance check item semantic query response encoding probability constraint vector, Represents the probability weight value of the software information-compliance inspection item semantic query response encoding, It means to subtract by position point. Represents software information-compliance check item semantic query response encoding sequence constraint vector; Finally, the software information-compliance check item semantic query response coding information interaction vector is added to the software information-compliance check item semantic query response coding sequence constraint vector to obtain an optimized device software information-compliance check item semantic query response coding vector. Finally, the optimized device software information-compliance check item semantic query response coding vector is passed through a classifier-based compliance intelligent matching module to determine whether the motor vehicle inspection and testing equipment software is compliant.

[0043] That is, considering that the weight matrix during classification probability regression acts on the semantic query response coding vector of the device software information-compliance check item to be tested, the distribution diversity of the semantic query response coding vector of the device software information-compliance check item to be tested causes the uncertainty of the weight stimulus parameters of the weight matrix, resulting in the lack of posterior probability density inference of the semantic query response coding vector of the device software information-compliance check item to be tested through the weight matrix, which will affect the accuracy of the probability regression results.

[0044] The above-mentioned optimization process simulates the mesoscale interaction structure under probability constraints between the eigenvalue scale and the eigenvector scale of the semantic query response coding vector of the software information of the device to be tested-compliance check items through the probability statistical characteristics of the semantic query response coding vector of the software information of the device to be tested-compliance check items, so as to construct a bidirectional latent variable motif based on the mesoscale short sequence relative to the probability statistical value to perform mesoscale bidirectional migration, and perform a posteriori recovery based on the sequence constraints on the interactive information, thereby improving the convergence effect of the semantic query response coding vector of the software information of the device to be tested-compliance check items in the probability density domain, and improving the accuracy of the matching result obtained by its input into the compliance intelligent matching module based on the classifier.

[0045] In summary, the intelligent matching method for the compliance of the motor vehicle inspection and testing equipment software according to the embodiment of the present application is explained, which first builds a software compliance inspection database based on standard files, and uses natural language processing technology based on deep learning to perform semantic analysis on each compliance inspection item in the database and the motor vehicle inspection and testing equipment software information, and extracts the semantic feature representation of the equipment software information and each compliance inspection item respectively, and then, by querying and matching the equipment software information with each compliance inspection item, the compliance inspection of the motor vehicle inspection and testing equipment software is intelligently realized. In this way, the efficiency and accuracy of compliance inspection can be greatly improved, the subjectivity and errors of manual review can be reduced, the software version can be better adapted to the update and change, and the motor vehicle inspection and testing equipment software can be ensured to meet the latest prescribed standards.

[0046] 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.

[0047] In the above embodiments, the description of each embodiment has its own emphasis. For 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 module division is only a logical function division, and there may be other division methods in actual implementation. The modules described as separate components may or may not be physically separated, and the components displayed as modules 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 modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0048] 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.

[0049] 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.

[0050] 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 intended to limit it. Although the present invention has been 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 intelligently matching the compliance of software of motor vehicle inspection and testing equipment, characterized in that: include: Based on the standard file, construct a software compliance verification database, wherein the software compliance verification database includes a natural language description of multiple compliance verification items; Extracting information description of motor vehicle inspection and testing equipment software, including functions, safety and data processing capabilities; Semantically encoding the information description of the motor vehicle inspection and testing equipment software and the natural language description of each compliance inspection item in the software compliance inspection database to obtain a set of motor vehicle inspection and testing equipment software information description semantic encoding vectors and compliance inspection item semantic embedding encoding vectors; Performing query matching on the set of the motor vehicle inspection and testing equipment software information description semantic coding vector and the compliance inspection item semantic embedding coding vector to obtain the to-be-inspected equipment software information-compliance inspection item semantic query response coding vector; Determining whether the motor vehicle inspection and testing equipment software is compliant based on the to-be-inspected equipment software information-compliance inspection item semantic query response encoding vector; Among them, query matching is performed on the set of semantic coding vectors of the software information description of the motor vehicle inspection and testing equipment and the set of semantic embedded coding vectors of the compliance inspection items, including: extracting significant semantic features of the set of semantic embedded coding vectors of the compliance inspection items as a prompt template; based on the prompt template, cross-domain optimized query coding based on significant semantics guidance is performed on the semantic coding vectors of the software information description of the motor vehicle inspection and testing equipment and the set of semantic embedded coding vectors of the compliance inspection items to obtain the semantic query response coding vector of the equipment software information to be inspected-compliance inspection items.

2. The method for intelligently matching the compliance of software of motor vehicle inspection and testing equipment according to claim 1 is characterized in that: Semantically encoding the information description of the motor vehicle inspection and testing equipment software and the natural language description of each compliance inspection item in the software compliance inspection database to obtain a set of motor vehicle inspection and testing equipment software information description semantic encoding vectors and compliance inspection item semantic embedding encoding vectors, including: The information description of the motor vehicle inspection and testing equipment software and the natural language description of each compliance inspection item in the software compliance inspection database are respectively input into a semantic encoder based on the Bert model to obtain a set of semantic encoding vectors of the motor vehicle inspection and testing equipment software information description and semantic embedded encoding vectors of the compliance inspection items.

3. The method for intelligently matching the compliance of software of motor vehicle inspection and testing equipment according to claim 2 is characterized in that: Extracting significant semantic features of the set of semantic embedding encoding vectors of the compliance check items as a prompt template includes: Constructing a key matrix based on a set of semantic embedding encoding vectors of the compliance check items; The maximum value of each key vector in the key matrix is ​​extracted to obtain a compliance check item significant semantic feature vector as the prompt template.

4. The method for intelligently matching the compliance of software of motor vehicle inspection and testing equipment according to claim 3 is characterized in that: Constructing a key matrix based on the set of semantic embedding encoding vectors of the compliance check items, including: Using a key embedding matrix, linearly embedding-encode each compliance check item semantic embedding encoding vector in the set of compliance check item semantic embedding encoding vectors to obtain a set of compliance check item semantic embedding encoding vectors after linear transformation; The semantic embedding coding vector of the compliance check item after the linear transformation is used as the key vector, and the set of the semantic embedding coding vectors of the compliance check item after the linear transformation is arranged in a matrix to obtain the key matrix.

5. The method for intelligently matching the compliance of software of motor vehicle inspection and testing equipment according to claim 4 is characterized in that: Based on the prompt template, the motor vehicle inspection and testing equipment software information description semantic coding vector and the set of the compliance inspection item semantic embedding coding vector are subjected to cross-domain optimized query coding based on significant semantic guidance to obtain the to-be-inspected equipment software information-compliance inspection item semantic query response coding vector, including: Using a query embedding matrix and a value embedding matrix, linear embedding coding is performed on the semantic coding vector of the motor vehicle inspection and testing equipment software information description to obtain a value vector and a query vector respectively; Inputting the query vector, the value vector, each key vector in the key matrix and the prompt template into a heterogeneous converter structure optimized based on template prompts to obtain a set of cross-domain query encoding feature vectors of the device software information to be inspected-compliance inspection item; The query feature aggregation is performed on the set of cross-domain query encoding feature vectors of the device software information to be inspected-compliance inspection items to obtain the semantic query response encoding vector of the device software information to be inspected-compliance inspection items.

6. The method for intelligently matching the compliance of software of motor vehicle inspection and testing equipment according to claim 5 is characterized in that: The query vector, the value vector, each key vector in the key matrix and the prompt template are respectively input into a heterogeneous converter structure optimized based on template prompts to obtain a set of cross-domain query encoding feature vectors of the device software information to be inspected-compliance inspection items, including: Multiplying the query vector by the transposed vector of the key vector and dividing by the two-norm of the prompt template to obtain an attention score matrix; The attention score matrix is ​​multiplied by the prompt template after passing it through a softmax function to obtain a template prompt optimized attention weight vector; The template prompt optimized attention weight vector and the value vector are multiplied by the position point to obtain the cross-domain query encoding feature vector of the device software information to be inspected-compliance inspection item.

7. The method for intelligently matching the compliance of software of motor vehicle inspection and testing equipment according to claim 6 is characterized in that: The query feature aggregation is performed on the set of cross-domain query encoding feature vectors of the device software information to be inspected-compliance inspection item to obtain the semantic query response encoding vector of the device software information to be inspected-compliance inspection item, including: The position-wise mean vector of the set of cross-domain query encoding feature vectors of the to-be-inspected device software information-compliance inspection item is calculated to obtain the to-be-inspected device software information-compliance inspection item semantic query response encoding vector.

8. The method for intelligently matching the compliance of software of motor vehicle inspection and testing equipment according to claim 7 is characterized in that: Determining whether the motor vehicle inspection and testing equipment software is compliant based on the to-be-inspected equipment software information-compliance inspection item semantic query response encoding vector includes: The software information of the equipment to be inspected - compliance inspection item semantic query response encoding vector is passed through a classifier-based compliance intelligent matching module to determine whether the motor vehicle inspection and testing equipment software is compliant.

Images