Architecture model construction and review method for automatic review of technical system

Abstract

The present application relates to a method for constructing an architecture model and reviewing it for automatic review of a technical system, wherein the architecture model construction includes obtaining predefined review perspectives, including capability perspective, business perspective, system perspective, information activity perspective, and standard perspective; extracting corresponding model data based on the review perspective; constructing a target perspective review model based on the model data, following the review perspective item by item; and obtaining multi-perspective review results of a target document based on multiple target perspective review models. The present application can ensure the effect of technical system review, achieve relationship consistency and product consistency, lay the foundation for the feasibility, credibility, and scientific nature of the review results, and effectively improve the efficiency of automatic review of technical systems and the accuracy of automatic review results.

Description

Technical Field

[0001] The present application relates to the field of document review technology, and in particular to a method for constructing an architecture model and reviewing it for automatic review of technical systems. Background Art

[0002] The main object of technical system review is documents, and the commonly used methods are manual review and expert experience judgment. The Fifth Institute of Electronics of the Ministry of Industry and Information Technology proposed a set of document review methods in 2021. It has the functions of formulating document review forms, reviewing documents and filling in records, filling in software document review problem reports, research and development units modifying documents and filling in document review problem handling reports, document review regression testing, and software document evaluation and summary. It also provides review methods and common typical problems from four aspects: document completeness, standardization, consistency, and accuracy. In 2019, the School of Computer Science and Technology of Xidian University proposed an electronic design document review system based on OpenXML Word documents. It uses a lightweight interface to access Word objects to obtain all format attribute information, and uses an optimized algorithm to judge and compare the irregular format attribute information, which is used to assist in format and content modification, and realizes batch review and download of compressed files under high concurrency for use by document staff.

[0003] Currently, tools are mainly used to check the format of supporting documents of target projects to be reviewed and the traceability relationship between supporting documents, while the specific content review of supporting documents is still mainly done manually. This is time-consuming, labor-intensive, and costly, and is easily limited by the experts' field knowledge. It is also difficult to quickly locate problems in target projects to be reviewed, and there is a lack of methods and tools for automated technical system review. Summary of the invention

[0004] This application intends to provide an architecture model construction and review method for automatic review of technical systems, as well as corresponding devices, electronic servers, and storage media. The technical problems to be solved are achieved through the following technical solutions.

[0005] In a first aspect of this embodiment, a method for constructing an architecture model for automatic review of a technical system is proposed, including:

[0006] Obtaining predefined review perspectives, including capability perspective, business perspective, system perspective, information activity perspective, and standard perspective;

[0007] Extracting corresponding model data according to the review perspective;

[0008] Based on the model data, follow the review perspective item by item to build a target perspective review model;

[0009] Based on the multiple target perspective review models, a multi-perspective review result of the target document is obtained.

[0010] Furthermore, following the review perspectives item by item, a target perspective review model is constructed, including:

[0011] Corresponding to the capability perspective, constructing the target perspective review model including a capability decomposition model and an effect attribute model;

[0012] Corresponding to the business perspective, constructing the target perspective review model including a business activity process model and a business resource flow model;

[0013] Corresponding to the system perspective, constructing the target perspective review model including a system composition model, a system function description model and a system interaction model;

[0014] Corresponding to the information activity perspective, constructing the target perspective review model including an information activity process model and an information classification model;

[0015] Corresponding to the standard perspective, constructing the target perspective review model includes a standard system model.

[0016] Furthermore, corresponding to the system perspective, it also includes constructing a mapping relationship between system and capability, a mapping relationship between system function and activity, a mapping relationship between system and information activity, and a mapping relationship between system and business activity.

[0017] Furthermore, the target perspective review model has description, paradigm and review granularity requirements that meet consistency requirements.

[0018] In a second aspect of this embodiment, a system architecture model review method for automatic review of technical systems is proposed, including:

[0019] Adopting the architecture model construction method for automatic technical system review described in the first aspect, a multi-perspective review model of the target project to be reviewed is obtained;

[0020] Using a model node comparison algorithm, obtaining a comparison result between the multi-view review model and a pre-built reference review model;

[0021] Build corresponding quantitative review system processes based on predefined review content;

[0022] Based on the quantitative review system process and the comparison results, the technical system review conclusion of the target project to be reviewed is obtained.

[0023] Furthermore, the model node comparison algorithm obtains the comparison result by calculating the model node similarity between the multi-view review model and the reference review model, including:

[0024] A named entity recognition algorithm combining rules and conditional random fields is used to extract attribute information of multiple model nodes;

[0025] Using a subject content similarity algorithm, respectively obtain the similarity of multiple model node attribute information between the multi-view review model and the reference review model;

[0026] Based on the similarity of attribute information of multiple model nodes, the model node similarity between the multi-view review model and the reference review model is obtained.

[0027] Furthermore, the model node attribute information includes a node name and a node description; and / or, based on the similarity of multiple model node attribute information, the node similarity between the multi-perspective review model and the reference review model is obtained, including: calculating the final model node similarity based on the weights of the node name and the node description.

[0028] Furthermore, before constructing the corresponding quantitative review system process, it also includes: constructing a corresponding relationship between the review perspective and the review content, and the review content includes the capacity satisfaction, process compliance, system support, information compliance and standard coverage; wherein,

[0029] The capability perspective corresponds to the capability satisfaction situation;

[0030] The business perspective corresponds to the process compliance;

[0031] The system perspective corresponds to the capability satisfaction, the process compliance and the system support respectively;

[0032] The information activity perspective corresponds to the information compliance situation; the standard perspective corresponds to the standard coverage situation.

[0033] Furthermore, based on the predefined review content, a corresponding quantitative review system process is constructed, including:

[0034] For each of the review contents, construct the corresponding review content attributes and the calculation indicator system corresponding to each review content attribute;

[0035] Based on the attributes of the review content and the calculation indicators, a quantitative review process corresponding to the review content is constructed.

[0036] Furthermore, according to the quantitative review system process and the comparison results, the technical system review conclusion of the target project to be reviewed is obtained, including:

[0037] Obtain the capability satisfaction, process compliance, system support, information compliance and standard coverage of the target project to be reviewed to form a five-tuple, and input the five-tuple into the XGBoost model to generate the technical system review conclusion of the target project to be reviewed;

[0038] Based on the technical system review conclusion, the parameters of the XGBoost model are trained to obtain an updated XGBoost model.

[0039] In a third aspect of this embodiment, a system architecture model construction device for automatic review of technical systems is proposed, including:

[0040] An audit perspective predefinition module is configured to obtain predefined audit perspectives, wherein the audit perspectives include capability perspective, business perspective, system perspective, information activity perspective, and standard perspective;

[0041] A model data extraction module is configured to extract corresponding model data according to the review perspective;

[0042] A review model building module is configured to build a target perspective review model according to the model data and the review perspective item by item;

[0043] The review result acquisition module is configured to obtain multi-perspective review results of the target document based on multiple target perspective review models.

[0044] In a fourth aspect of this embodiment, a system architecture model review device for automatic review of a technical system is provided, comprising:

[0045] A review model acquisition module is configured to adopt the architecture model construction method for automatic review of technical systems described in the first aspect to obtain a multi-perspective review model of the target project to be reviewed;

[0046] A comparison result acquisition module is configured to use a model node comparison algorithm to obtain a comparison result between the multi-view review model and a pre-built reference review model;

[0047] The quantitative review system process building module is configured to build the corresponding quantitative review system process according to the predefined review content;

[0048] The review conclusion acquisition module is configured to obtain the technical system review conclusion of the target project to be reviewed based on the quantitative review system process and the comparison result.

[0049] In a fifth aspect of the present embodiment, an electronic server is proposed, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the processor implements the steps of the method for constructing an architecture model for automatic review of a technical system as described in the first aspect or the method for reviewing an architecture model for automatic review of a technical system as described in the second aspect.

[0050] In the sixth aspect of this embodiment, a computer-readable storage medium is proposed, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for constructing an architecture model for automatic review of a technical system as described in the first aspect or the method for reviewing an architecture model for automatic review of a technical system as described in the second aspect are implemented.

[0051] The embodiments of the present application include the following advantages:

[0052] 1. Adopt a multi-angle unified description framework to construct a model construction method for technical system review. In order to improve the efficiency, accuracy and comprehensiveness of the review, it is necessary to impose normative constraints on the design of the model to be reviewed so that it can be effectively used for the technical document modeling of the project to be reviewed and the automatic comparison and display in the review stage. A system modeling process for technical system review is proposed, and a description method, paradigm and review requirements for the model to be reviewed are provided to unify the form of results, ensure the effectiveness of the technical system review, achieve relationship consistency and product consistency, and lay the foundation for improving the feasibility, credibility and scientificity of the review results.

[0053] 2. The model review method of the embodiment of the present application first conducts automated comparative analysis based on the architecture model data, including extracting keywords in the model node description based on named entity recognition technology; obtaining the similarity of node names and node descriptions based on the subject similarity calculation technology; obtaining the total similarity between nodes based on the user-set weights; and displaying similar or violated nodes and relationships in a visual manner. Second, a method for conducting technical system review and analysis based on multiple angles, including capability satisfaction analysis, process compliance analysis, system support analysis, information compliance analysis, and standard coverage analysis. Third, a method for summarizing and generating technical system review results based on multi-angle review and analysis results and using the XGBoost model. First, based on historical project data, the XGBoost model is trained; second, the initial review conclusion is given based on the multi-angle review and analysis results of the project model to be reviewed newly uploaded by the user and the XGBoost model; then, the user can manually adjust the review conclusion and automatically incorporate the new review data into the training set of the XGBoost model; finally, the XGBoost model is iteratively optimized for the next review result summary. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is one of the flowcharts of the method for constructing an architecture model for automatic review of a technical system according to an embodiment of the present application;

[0055] Figure 2 This is the second flow chart of the method for constructing an architecture model for automatic review of a technical system according to an embodiment of the present application;

[0056] Figure 3 This is one of the flowcharts of the system architecture model review method for automatic review of the technical system according to the embodiment of the present application;

[0057] Figure 4 This is the second flow chart of the system architecture model review method for automatic review of technical systems in the embodiment of the present application;

[0058] Figure 5 This is the third flow chart of the architecture model review method for automatic review of technical systems in the embodiment of the present application;

[0059] Figure 6 This is the fourth flowchart of the architecture model review method for automatic review of technical systems in an embodiment of the present application;

[0060] Figure 7 It is a schematic diagram of the structure of an architecture model building device for automatic review of technical systems according to an embodiment of the present application;

[0061] Figure 8 It is a schematic diagram of the structure of the system architecture model review device for automatic review of the technical system according to an embodiment of the present application;

[0062] Fig. 9 It is a structural diagram of an electronic server provided in an embodiment of the present application. DETAILED DESCRIPTION

[0063] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0064] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0065] Embodiment 1:

[0066] like Figure 1 FIG. 1 is a flow chart of a method for constructing an architecture model for automatic review of a technical system according to an embodiment of the present application. Figure 1 , the method comprising:

[0067] S101: Obtain predefined review perspectives, where the review perspectives include capability perspective, business perspective, system perspective, information activity perspective, and standard perspective.

[0068] S102: Extracting corresponding model data according to the review perspective.

[0069] S103: Based on the model data, follow the review perspective item by item to build a target perspective review model.

[0070] S104: Based on the multiple target perspective review models, obtain a multi-perspective review result of the target document.

[0071] Specifically, in order to support the development of technical system review and form digital architecture assets, for the review projects with existing relevant documents, it is necessary to build corresponding models based on the review content, realize the digitization of review content and the graphical display of project, and support the technical system compliance review. First, analyze the content of existing documents that support the technical system compliance review, and determine the system architecture design model that needs to be built according to the review objectives and specific content corresponding to the system architecture design model; analyze the relationship between existing documents and corresponding models, extract capability data, system data, activity data, system capability data, etc., and conduct demand analysis and capability design; carry out the design of models related to the review key perspective capability perspective, business perspective, system perspective, information activity perspective and standard perspective, sort out and improve the overall design of the model, and finally form a multi-perspective review model.

[0072] In some embodiments, the review perspective is followed item by item to construct a target perspective review model, such as Figure 2 As shown, including:

[0073] S211: Corresponding to the capability perspective, constructing the target perspective review model including a capability decomposition model and an effect attribute model.

[0074] S212: Corresponding to the business perspective, constructing the target perspective review model including a business activity process model and a business resource flow model.

[0075] S213: Corresponding to the system perspective, constructing the target perspective review model including a system composition model, a system function description model and a system interaction model.

[0076] S214: Corresponding to the information activity perspective, constructing the target perspective review model including an information activity process model and an information classification model.

[0077] S215: Corresponding to the standard perspective, construct the target perspective review model including the standard system model.

[0078] Specifically, according to the review content of the system review, in order to support the needs of automated review technology and achieve the availability of results, the review model needs to cover the capability perspective, business perspective, system perspective, information activity perspective, and standard perspective. There are multiple different types of models under different perspectives, which together constitute the design content of the target project to be reviewed. For the capability perspective, the system's capabilities are described, requirements for the development of system capabilities are proposed, and the system's capability classification and composition, capability effects, and capability relationships are determined to support the planning of system capabilities, mainly including capability decomposition models and effect attribute models. For the business perspective, the process and constraints of task execution, as well as the organizational grouping and collaborative relationships of task completion are described, and task process optimization and organizational relationship settings are supported, mainly including business activity process models and business resource flow models. For the system perspective, the system composition and structure that supports the system to complete tasks, the functions of the system and their evolution process are described, mainly including system composition models and system interaction models. For the information activity perspective, the information flow is the core, describing the composition of information activities, the input and output interfaces of information activities, the functions and information systems of information conversion, and the collaborative relationships of information activities, mainly including information activity process models and information classification models. The purpose of describing the construction and application-related standards, specifications, regulations, rules, policies and regulations from a standard perspective is to provide a basis for system planning, construction, evaluation, management and application, mainly including the standard system model.

[0079] In some embodiments, corresponding to the system perspective, it also includes constructing a system-capability mapping relationship, a system-function-activity mapping relationship, a system-information activity mapping relationship, and a system-business activity mapping relationship.

[0080] Specifically, for the main models under review, different models and model data are constructed from different perspectives. Attention should be paid to the relationship between model data when constructing the model, mainly including the modeling process of different models from the capability perspective, business perspective, system perspective, information activity perspective and standard perspective. Tables 1 to 5 show the modeling process of the capabilities perspective, business perspective, system perspective, information activity perspective and standard perspective respectively.

[0081] Table 1 Modeling process of capability perspective model

[0082]

[0083] Table 2 Business perspective modeling process

[0084]

[0085] Table 3 Modeling process of system perspective model

[0086]

[0087]

[0088] Table 4 Modeling process of information activity perspective model

[0089]

[0090] Table 5 Modeling process of standard perspective model

[0091]

[0092] In some embodiments, the target perspective review model has a description, paradigm and review granularity requirements that meet consistency requirements. Specifically, as shown in Tables 6 to 10, they are the model review granularity requirements for capability perspective, business perspective, system perspective, information activity perspective and standard perspective.

[0093] Table 6 Capability perspective model review granularity requirements

[0094]

[0095] Table 7 Business perspective model review granularity requirements

[0096]

[0097] Table 8 System perspective model review granularity requirements

[0098]

[0099] Table 9 Information activity perspective model review granularity requirements

[0100]

[0101] Table 10 Standard viewpoint model review granularity requirements

[0102]

[0103] The embodiment of the present application adopts a unified description framework from multiple angles to construct an architecture model construction method for automatic review of technical systems. In order to improve the efficiency, accuracy and comprehensiveness of the review, it is necessary to impose normative constraints on the design of the target model to be reviewed, so that it can be effectively used for technical document modeling of the target project to be reviewed and automatic comparison and review conclusion display during the review stage. The system modeling process for technical system review provides description methods, paradigms and review requirements for the target model to be reviewed, unifies the form of results, ensures the effectiveness of technical system review, achieves relationship consistency and product consistency, and lays the foundation for improving the feasibility, credibility and scientificity of the review results.

[0104] Embodiment 2:

[0105] like Figure 3 FIG. 1 is a flow chart of a method for reviewing an architecture model for automatic review of a technical system according to an embodiment of the present application. The method comprises:

[0106] S301: Using the method for constructing an architecture model for automatic technical system review described in Example 1, a multi-perspective review model of the target project to be reviewed is obtained.

[0107] S302: Using a model node comparison algorithm, obtain a comparison result between the multi-view review model and a pre-built reference review model.

[0108] S303: Construct corresponding quantitative review system processes based on predefined review contents.

[0109] S304: Obtain the technical system review conclusion of the target project to be reviewed based on the quantitative review system process and the comparison result.

[0110] In some embodiments, before constructing the corresponding quantitative review system process, it also includes: constructing a corresponding relationship between the review perspective and the review content, and the review content includes capability satisfaction, process compliance, system support, information compliance and standard coverage; wherein,

[0111] The capability perspective corresponds to the capability satisfaction situation;

[0112] The business perspective corresponds to the process compliance;

[0113] The system perspective corresponds to the capability satisfaction, the process compliance and the system support respectively;

[0114] The information activity perspective corresponds to the information compliance situation; the standard perspective corresponds to the standard coverage situation.

[0115] As shown in Table 11, it is a correspondence table between review content, review perspective and review model.

[0116] Table 11 Correspondence between review content, review perspective and review model

[0117]

[0118] Specifically, the technical system review mainly focuses on the ability to meet the situation, process compliance, system support, information compliance, standard coverage, etc. Based on the review objectives, a method is proposed to use a unified description framework from multiple perspectives to construct corresponding perspective models such as capability perspective, business perspective, system perspective, information activity perspective, and standard perspective. Based on the formed perspective model, the automated review of the technical system is carried out. First, the user uploads the perspective model constructed based on the target project to be reviewed; second, the reference review model in the pre-built project is checked, and the model node range in the corresponding pre-built project is defined according to the content of the target project to be reviewed, and the review setting work is carried out; then, the model of the target project to be reviewed is automatically analyzed and compared with the model within the defined scope of the pre-built project, and the similarity of the model nodes is calculated; second, according to the comparison results, the ability to meet the situation, process compliance, system support, information compliance, standard coverage, etc. are calculated; finally, based on the compliance of the various indicators obtained, the final conclusion of the technical system review of the target project to be reviewed is drawn.

[0119] In some embodiments, the model node comparison algorithm obtains the comparison result by calculating the model node similarity between the multi-view review model and the reference review model, such as Figure 4 As shown, including:

[0120] S411: Using a named entity recognition algorithm that combines rules with conditional random fields to extract attribute information of multiple model nodes.

[0121] S412: Using a subject content similarity algorithm, respectively obtain the similarity of attribute information of multiple model nodes between the multi-perspective review model and the reference review model.

[0122] S413: Based on the similarity of the attribute information of the multiple model nodes, obtain the model node similarity between the multi-view review model and the reference review model.

[0123] Specifically, the pre-built reference review model and the target model to be reviewed constructed by the user are mainly composed of nodes and the relationships between nodes. Nodes contain attributes such as node name and node description. Based on the name, description and other attribute contents of the nodes, the pre-built reference review model and the target model to be reviewed are automatically compared and analyzed to obtain the node similarity between the target model to be reviewed and the pre-built reference review model. First, the named entity recognition technology based on the combination of rules and conditional random fields is used to extract the key information in the description; then the subject content similarity calculation method is used to obtain the similarity in the node name and description key information between the pre-built reference review model and the target model to be reviewed, and the node similarity between the target model to be reviewed and the pre-built reference review model is obtained comprehensively.

[0124] At present, there is still a blank in the research on named entity recognition of technical system review both internationally and domestically. Compared with named entity recognition in English and named entity recognition of Chinese names, place names, etc., the difficulties of named entity recognition of technical system review are mainly in the following aspects: (1) Chinese text is different from English. There are no natural separators between words like spaces in English text. Therefore, the premise of Chinese named entity recognition is to first determine the boundaries between Chinese words, that is, Chinese word segmentation; (2) Chinese word segmentation and named entity recognition affect each other. The accuracy of word segmentation directly affects the accuracy of name entity recognition and extraction to a certain extent; (3) The composition structure of Chinese name entities is extremely complex and there is no limit to the degree. At present, there is no recognition model for name entities in technical system review. Different name entity types have different internal and external characteristics. It is difficult or even impossible to use a simple and unified model to describe the internal and external characteristics of all entities. There is no model that can solve all problems once and for all. The model is closely related to the problem to be solved.

[0125] In order to solve the above difficulties, this embodiment adopts a named entity recognition algorithm based on a combination of rules and conditional random fields. The probability of a labeled sequence Y for a specific sequence X is , where x is the sequence to be labeled, Z(x) is the normalization factor, represents the state transition characteristics of the observation sequence x and the corresponding annotation sequence at position i-1 and position i, Represents the entire observation sequence And its corresponding annotation sequence is The state characteristics of a location. and They are and The weight of . Using conditional random fields to study named entity recognition and integrating the reverse maximum probability word segmentation method into the conditional random field model can effectively improve the accuracy and recall rate of named entity recognition.

[0126] The phrase elements of the target model to be reviewed and the pre-built reference review model are represented by MWT (Multi-WordsTerms). Assuming that MWT is composed of three parts, the formula is described as follows: ; Where T: represents the head of MWT; M: represents the middle of MWT; E: represents the tail of MWT. And define the following synonym recognition rules:

[0127] : When the T of two MWTs is the same and M does not exist, if E is synonymous, then the two MWTs are synonyms;

[0128] : When the E of two MWTs is the same and M does not exist, if T is synonymous, the two MWTs are synonyms;

[0129] :when E and When T is the same and M does not exist, if T and If the two MWTs are synonyms, then they are synonyms;

[0130] : When the E and T of two MWTs are the same and M is synonymous, the two MWTs are synonyms;

[0131] : When the T and M of two MWTs are the same and E is synonymous, the two MWTs are synonyms;

[0132] : When the E and M of two MWTs are the same and T is synonymous, the two MWTs are synonyms;

[0133] In the embodiment of the present application, the synonym extraction technology mainly includes: first, according to the difference between Chinese and English, it is necessary to first segment the Chinese words and filter out the English characters. Secondly, the skip-gram method in Word2Vec is used to embed the words in the text into the word vector space according to the co-occurrence information of the words in the context, and the cosine similarity calculation in the subsequent steps is used to obtain the similarity between the two words. Finally, the MWT that needs to be calculated for similarity is split into three parts (i.e., the head, the middle, and the tail). When the two terms are partially the same, the cosine similarity of the different parts is calculated according to the word vector obtained in the second step, and the semantically identical or similar parts are selected according to the preset threshold. If the different parts of the two terms are similar in meaning, the two terms are synonymous.

[0134] In some embodiments, the model node attribute information includes a node name and a node description; and / or, based on the similarity of multiple model node attribute information, the node similarity between the multi-perspective review model and the reference review model is obtained, including: calculating the final model node similarity based on the weights of the node name and the node description.

[0135] Specifically, based on the above two technologies, the similarities of node names and node descriptions between the target model to be reviewed and the pre-built reference review model are obtained. According to the weights of the names and descriptions set by the user, the final node similarity is obtained. According to the node similarity, subsequent indicator analysis work is carried out.

[0136] In some embodiments, a corresponding quantitative review system process is constructed based on the predefined review content, such as Figure 5 As shown, including:

[0137] S511: For each of the review contents, construct corresponding review content attributes and a calculation indicator system corresponding to each review content attribute.

[0138] S512: Based on the review content attributes and the calculation indicators, construct a quantitative review process corresponding to the review content.

[0139] The following is a detailed description of the automatic analysis scheme based on the comparison results of model nodes to conduct multi-angle analysis such as capability satisfaction, process compliance, system support, information compliance, and standard coverage.

[0140] 1. Capacity Satisfaction

[0141] The content of the capability satisfaction review includes analyzing the capability satisfaction through quantitative review of the capability, decomposition relationship, correspondence with the system and effect attributes. The corresponding calculation indicators are shown in Table 12.

[0142] Table 12 Capability attribute measurement index table

[0143]

[0144] The process of reviewing the ability to meet the requirements is as follows:

[0145] First, based on the knowledge base capability range defined by the experts, the review system automatically calculates indicators and conducts review analysis to obtain preliminary review results. Then, the experts confirm the results of the review analysis and calculate indicators such as capability matching, coverage, decomposition relationship matching, and the proportion of capabilities supported by the system. Finally, the capability satisfaction is obtained.

[0146] (1) Experts define the scope of knowledge base capabilities: In order to more accurately calculate capability indicators, experts are first required to define the capabilities that need to be implemented or refined in the knowledge base model, focus on the key capabilities of different projects, exclude other capabilities with lower correlation, and improve the scientificity and accuracy of the calculation results;

[0147] (2) Automatic review and analysis: Automatically compare model nodes and display the review results in a digital and graphical manner. Different types of visual display methods can assist experts in reviewing the models of pending projects;

[0148] (3) Experts confirm the review and analysis results: During the review and analysis process, different experts may have different understandings of the review results and there may be limitations in automatic review. After the model is automatically reviewed and analyzed, the review experts can manually adjust the review results such as node correspondence, line correspondence, and violation correspondence;

[0149] (4) Calculation of capability indicators: Based on the conformity evaluation calculation model established in accordance with the technical system review specifications, quantitative calculations are performed based on the comparative analysis results of the project model to be reviewed and the knowledge base architecture model to obtain the conformity evaluation calculation results of the review indicators. The evaluation calculations are used to display the coverage rate, matching degree, connection relationship compliance, connection relationship violation, data support in the model to be reviewed, and other review indicator results of the project model data to the knowledge base architecture model data;

[0150] (5) Capacity satisfaction analysis: Based on the calculation results of capability indicators and the calculation weights of various indicators set by the user, the capacity satisfaction is calculated. The user can manually adjust the capacity satisfaction analysis results.

[0151] The calculation method of capability indicators is as follows:

[0152] (1) Capability matching degree: the number of capabilities similar to the project to be reviewed and the knowledge base project and the number of their sub-nodes / the number of capabilities of the project to be reviewed;

[0153] (2) Capability coverage: the number of capabilities similar to those of the knowledge base items and the items to be reviewed / the number of capabilities of the knowledge base items defined by experts;

[0154] (3) Matching degree of decomposition relationship of pending capability: the number of decomposition relationships similar to the pending item and the knowledge base item / the number of decomposition relationships of the pending item;

[0155] (4) The proportion of capabilities with system support in the projects to be reviewed: the number of capabilities with system support / the number of capabilities in the projects to be reviewed;

[0156] (5) Effect attribute matching: the number of ability-effect relationships that are similar between the pending project and the knowledge base project / the number of ability-effect relationships in the pending project;

[0157] (6) Effect attribute coverage: the number of ability-effect relationships that are similar between the knowledge base project and the project to be reviewed / the number of ability-effect relationships in the knowledge base project.

[0158] 2. The process is in line with the situation

[0159] Through the quantitative review of nodes and process interaction relationships between nodes, the process compliance is analyzed. The corresponding calculation indicators are shown in Table 13.

[0160] Table 13 Process compliance metrics table

[0161]

[0162] The process for reviewing process compliance is as follows:

[0163] First, experts define the scope of activities and business nodes / external nodes in the knowledge base project, and the review system automatically calculates indicators and conducts review analysis to obtain initial review results; secondly, experts confirm the results of the review analysis and calculate indicator values ​​such as activity matching, interaction relationship matching, business node / external node matching, and process mapping relationship matching; finally, the process compliance is obtained.

[0164] (1) Experts define activities, business nodes, and external nodes in the knowledge base project: In order to calculate process compliance more accurately, experts are first required to define the activities, knowledge base business nodes, and external nodes for comparison to improve the scientificity and accuracy of the calculation results.

[0165] (2) Automatic review and analysis: Automatically compare model nodes and display the review results in a digital and graphical manner. Different types of visual display methods can assist experts in reviewing the models of pending projects;

[0166] (3) Experts confirm the review and analysis results: During the review and analysis process, different experts may have different understandings of the review results and there may be limitations in automatic review. After the model is automatically reviewed and analyzed, the review experts can manually adjust the review results such as node correspondence, line correspondence, and violation relationship;

[0167] (4) Calculation of process indicators: Based on the conformity evaluation calculation model established in accordance with the technical system review specifications, quantitative calculations are performed based on the comparative analysis results of the project model to be reviewed and the knowledge base architecture model to obtain the conformity evaluation calculation results of the review indicators. The evaluation calculations are used to display the coverage rate, matching degree, relationship conformity, relationship violation, data support in the model to be reviewed, and other review indicator results of the project model data to the knowledge base architecture model data;

[0168] (5) Process compliance analysis: Based on the calculation results of process indicators and the calculation weights of various indicators set by the user, the process compliance is calculated. Users can manually adjust the process compliance analysis results.

[0169] The process indicators are calculated as follows:

[0170] (1) Activity matching: the number of activities similar to those in the pending project and the knowledge base project / the number of activities in the pending project;

[0171] (2) Activity coverage: the number of activities similar to those in the knowledge base project and the project to be reviewed / the number of activities in the designated knowledge base project;

[0172] (3) Matching degree of interaction relationships between pending activities: the number of interaction relationships between activities similar to the pending projects / the number of interaction relationships between activities of the pending projects;

[0173] (4) Proportion of activities with system function support: number of activities with system function support / number of activities in pending projects;

[0174] (5) Node matching degree: the number of (external) business nodes similar to the project to be reviewed and the project in the knowledge base / the number of (external) business nodes of the project to be reviewed;

[0175] (6) Node coverage: the number of (external) business nodes similar to the knowledge base project and the project to be reviewed / the number of (external) business nodes of the designated knowledge base project;

[0176] (7) Business node process mapping relationship matching degree: the number of resource flows similar to the pending project and the knowledge base project / the number of resource flows of the pending project;

[0177] (8) Proportion of content containing communication means attributes: the number of information flows (resource flows) containing communication means in the model to be reviewed / the number of information flows (resource flows) in the model to be reviewed;

[0178] (9) Proportion of content containing information exchange method attributes: the number of information flows (resource flows) containing information exchange methods in the model to be reviewed / the number of information flows (resource flows) in the model to be reviewed;

[0179] (10) Proportion of content containing protocol attributes: the number of information flows (resource flows) containing protocols in the model to be reviewed / the number of information flows (resource flows) in the model to be reviewed.

[0180] 3. System support

[0181] System support is analyzed through quantitative review of system / external system, system function, composition / decomposition relationship, interaction relationship, etc. The corresponding calculation indicators are shown in Table 14.

[0182] Table 14 System attribute measurement index table

[0183]

[0184] The system support review process is as follows:

[0185] First, experts define the scope of systems and system functions in the knowledge base project, and the review system automatically calculates indicators and conducts review analysis to obtain initial review results; secondly, experts confirm the review and analysis results, and calculate the matching degree, coverage rate, decomposition relationship matching degree, and related ratio indicator values; finally, the system support situation is obtained.

[0186] (1) Experts define the scope of knowledge base systems and system functions: In order to more accurately calculate system indicators, experts are first required to define the systems that need to be built and the system functions that need to be provided in the knowledge base model, focus on the systems and functions within the scope, exclude other systems with lower correlation, and improve the scientificity and accuracy of the calculation results;

[0187] (2) Automatic review and analysis: Automatically compare model nodes and display the review results in a digital and graphical manner. Different types of visual display methods can assist experts in reviewing the models of pending projects;

[0188] (3) Experts confirm the review and analysis results: During the review and analysis process, different experts may have different understandings of the review results and there may be limitations in automatic review. After the model is automatically reviewed and analyzed, the review experts can manually adjust the review results such as node correspondence, line correspondence, and violation relationship;

[0189] (4) Calculation of system indicators: Based on the conformity evaluation calculation model established in accordance with the technical system review specifications, quantitative calculations are performed based on the comparative analysis results of the project model to be reviewed and the knowledge base architecture model to obtain the conformity evaluation calculation results of the review indicators. The evaluation calculations are used to display the coverage rate, matching degree, relationship conformity, relationship violation, data support in the model to be reviewed, and other review indicator results of the project model data to the knowledge base architecture model data;

[0190] (5) System support situation analysis: Based on the quantitative review results of the system / external system, system functions, composition / decomposition relationships, interaction relationships, etc., as well as the calculation weights of various indicators set by the user, the system support situation is calculated. The user can manually adjust the system support situation analysis results.

[0191] The calculation method of system indicators is as follows:

[0192] (1) System matching degree: the number of systems and all their sub-nodes that are similar to the project to be reviewed and the knowledge base project / the number of systems of the project to be reviewed;

[0193] (2) Proportion of systems with decomposed system functions: the number of systems with system functions in the projects to be reviewed / the number of systems in the projects to be reviewed;

[0194] (3) System function matching degree: the number of system functions and all sub-nodes similar to the items under review and the items in the knowledge base / the number of system functions of the items under review;

[0195] (4) The ratio of system functions to corresponding systems: the number of system functions supported by systems in the pending projects / the number of system functions in the pending projects;

[0196] (5) The proportion of system functions to be reviewed that have system function processes: the number of system functions with system function processes / the number of system functions in the projects to be reviewed;

[0197] (6) The ratio of system functions with system function processes to corresponding systems: the number of system functions with system function processes and system support / the number of system functions with system function processes;

[0198] (7) System composition relationship matching degree: the number of system composition relationships that are similar between the pending project and the knowledge base project / the number of system composition relationships of the pending project;

[0199] (8) System function decomposition relationship matching degree: the number of system function decomposition relationships similar to the items to be reviewed and the items in the knowledge base / the number of system function decomposition relationships of the items to be reviewed;

[0200] (9) Matching degree between system function interaction and system interaction relationship: the number of similar system function interaction relationships between the pending projects and the knowledge base projects / the number of system function interaction relationships between the pending projects.

[0201] 4. Information is consistent with the situation

[0202] Information compliance is analyzed through quantitative review of information, information categories and their relationships. The corresponding calculation indicators are shown in Table 15.

[0203] Table 15 Information / information attribute measurement index table

[0204]

[0205] The information compliance review process is as follows:

[0206] First, experts define the information categories and scope of information in the knowledge base project, and the review system automatically calculates indicators and conducts review analysis to obtain initial review results; secondly, experts confirm the review and analysis results and calculate indicator values ​​such as information / information category matching, information / information category coverage, and information and information attribute mapping relationship matching; finally, the information compliance status is obtained.

[0207] (1) Experts define the scope of information categories and information in the knowledge base project: In order to more accurately calculate information / information category indicators, experts are first required to define the scope of raw data, product data, demand data, user data, and catalog data in the knowledge base project;

[0208] (2) Automatic review and analysis: Automatically compare model nodes and display the review results in a digital or graphical manner. Different types of visual display methods can assist experts in reviewing the models of pending projects;

[0209] (3) Experts confirm the review and analysis results: During the review and analysis process, different experts may have different understandings of the review results and there may be limitations in automatic review. After the model is automatically reviewed and analyzed, the review experts can manually adjust the review results such as node correspondence, line correspondence, and violation relationship;

[0210] (4) Calculation of information / information indicators: Based on the conformity evaluation calculation model established in accordance with the technical system review specifications, quantitative calculations are performed based on the comparative analysis results of the project model to be reviewed and the knowledge base architecture model to obtain the conformity evaluation calculation results of the review indicators. The evaluation calculations are used to display the coverage rate, matching degree, relationship conformity, relationship violation, data support in the model to be reviewed, and other review indicator results of the project model data to the knowledge base architecture model data;

[0211] (5) Information compliance analysis: Based on the quantitative review results of information, information categories and their relationships, as well as the calculation weights of various indicators set by the user, the information compliance is calculated. Users can manually adjust the information compliance analysis results.

[0212] Information / information indicators are calculated as follows:

[0213] (1) Information coverage: the number of similar information between knowledge base items and items to be reviewed / the number of information on knowledge base items identified by experts;

[0214] (2) Information category coverage: the number of information categories similar to the items in the knowledge base and the items to be reviewed / the number of information categories of the items in the knowledge base identified by experts;

[0215] (3) Information matching degree: the number of similar information between the pending item and the knowledge base item / the number of information of the pending item;

[0216] (4) Information category matching degree: the number of information categories similar to the items to be reviewed and the items in the knowledge base / the number of information categories of the items to be reviewed;

[0217] (5) Matching degree between information category and information affiliation: the number of similar information affiliations between the pending items and the knowledge base items / the number of similar information affiliations between the pending items;

[0218] (6) Matching degree of information and information attribute mapping relationship: the number of information and information attribute mapping relationships that are similar between the pending review items and the knowledge base items / the number of information and information attribute mapping relationships between the pending review items;

[0219] 5. Standard coverage

[0220] Standard coverage is analyzed through quantitative review of standards, standard classification, standard system and mapping relationship. The corresponding calculation indicators are shown as follows.

[0221] Table 16 Standard class attribute measurement index table

[0222]

[0223] The standard coverage review process is as follows

[0224] First, based on the standard system, standard classification and standards in the knowledge base projects that need to be followed as defined by experts, as well as the scope of resource layer, data layer and application layer standards in the knowledge base projects, the review system automatically calculates indicators and conducts review analysis to obtain initial review results; secondly, experts confirm the results of the review analysis and calculate the indicator values ​​of standard matching degree, coverage rate, standard classification matching degree, coverage rate, etc.; finally, the standard coverage situation is obtained.

[0225] (1) Experts define the scope of the knowledge base project standard system, standard classification and standards: In order to more accurately calculate standard indicators, experts are first required to define the standard system, standard classification and standards in the knowledge base project to be followed, as well as the scope of the resource layer, data layer and application layer standards in the knowledge base project, focus on the standard system to be followed, exclude other standards with low relevance, and improve the scientificity and accuracy of the calculation results;

[0226] (2) Automatic review and analysis: Automatically compare the nodes of the memory model and display the review results in a digital and graphical manner. Different types of visual display methods can assist experts in reviewing the models of the projects to be reviewed;

[0227] (3) Experts confirm the review and analysis results: During the review and analysis process, different experts may have different understandings of the review results and there may be limitations in automatic review. After the model is automatically reviewed and analyzed, the review experts can manually adjust the review results such as node correspondence, line correspondence, and violation relationship;

[0228] (4) Calculation of standard indicators: Based on the conformity evaluation calculation model established in accordance with the technical system review specifications, quantitative calculation is performed based on the comparative analysis results of the project model to be reviewed and the knowledge base architecture model to obtain the conformity evaluation calculation results of the review indicators. The evaluation calculation shows the review indicator results such as the coverage rate, matching degree, relationship conformity, relationship violation, and data support in the model to be reviewed of the project model data for the knowledge base architecture model data;

[0229] (5) Standards coverage analysis: Based on the quantitative review results of standards, standard classifications, standard systems, and mapping relationships, as well as the calculation weights of various indicators set by the user, the standards coverage is calculated. Users can manually adjust the standards coverage analysis results.

[0230] The calculation method of standard indicators is as follows:

[0231] (1) Standard matching degree: the number of standards in the projects to be reviewed that are similar to the experts' classification standards / the number of standards in the projects to be reviewed;

[0232] (2) Standard coverage: the number of standards that are similar to those in the projects to be reviewed / the number of standards that are similar to those in the projects to be reviewed;

[0233] (3) Ratio of newly added standards: the number of standards in the pending projects with the status of newly added / the number of standards in the pending projects;

[0234] (4) Proportion of revised standards: the number of standards in the status of revised in the projects to be reviewed / the number of standards in the projects to be reviewed;

[0235] (5) Standard classification matching degree: the number of standard classifications and their sub-standard classifications in the projects to be reviewed that are similar to the standard classifications divided by experts / the number of standard classifications in the projects to be reviewed;

[0236] (6) Standard classification coverage: the number of standard classifications divided by experts that are similar to those in the pending projects / the number of standard classifications divided by experts;

[0237] (7) Standard system matching degree: the number of standard systems in the projects to be reviewed that are similar to the standard systems classified by experts / the number of standard systems in the projects to be reviewed;

[0238] (8) Standard system coverage rate: the number of standard systems classified by experts and similar to those in the pending projects / the number of standard systems classified by experts;

[0239] (9) Matching degree of mapping relationship between standard system and standard classification: the number of mapping relationships between standard system and standard classification similar to the projects to be reviewed / the number of mapping relationships between standard system and standard classification of the projects to be reviewed;

[0240] (10) Coverage of mapping relationships between standard systems and standard classifications: the number of mapping relationships between standard systems and standard classifications similar to knowledge base projects / the number of mapping relationships between standard systems and standard classifications classified by experts;

[0241] (11) Matching degree between standard classification and standard mapping relationship: the number of standard classification and standard mapping relationships similar to the projects to be reviewed / the number of standard classification and standard mapping relationships of the projects to be reviewed;

[0242] (12) Coverage of standard classification and standard mapping relationships: the number of standard classifications and standard mapping relationships similar to knowledge base projects / the number of standard classifications and standard mapping relationships classified by experts.

[0243] In some embodiments, according to the quantitative review system process and the comparison result, the technical system review conclusion of the target project to be reviewed is obtained, such as Figure 6 As shown, including:

[0244] S611: Obtain the capability satisfaction, process compliance, system support, information compliance and standard coverage of the target project to be reviewed to form a quintuple, and input the quintuple into the XGBoost model to generate the technical system review conclusion of the target project to be reviewed.

[0245] S612: Based on the technical system review conclusion, the parameters of the XGBoost model are trained again to obtain the updated XGBoost model.

[0246] Specifically, based on the review results of capability satisfaction, process compliance, system support, information compliance and standard coverage, the technical system review conclusions are summarized and generated based on the XGBoost training model. They are divided into three categories: full compliance, basic compliance and non-compliance. Users can manually adjust the review conclusions to obtain the final results of the technical system review.

[0247] For data collection. First, a total of 2056 projects that have completed technical system review between 2015 and 2022 are collected, of which 601 projects fully meet the review requirements, 993 projects basically meet the review requirements, and 462 projects do not meet the review requirements. 80% of the projects in these three categories are used as training sets, and the remaining projects are used as test sets. Finally, the training set consists of 1645 projects, including 481 fully compliant projects, 794 basically compliant projects, and 370 non-compliant projects. The test set consists of 411 projects, including 120 fully compliant projects, 199 basically compliant projects, and 92 non-compliant projects. Secondly, the above projects are modeled, and the models to be reviewed are constructed respectively, uploaded to the review system, and the capability satisfaction, process compliance, system support, information compliance and standard coverage of each project are obtained to form a quintuple.<Cap,Seq,Sys,Inf,Std> The five-tuple data of 1645 items in the training set and their corresponding labels (3 marks completely compliant, 2 marks basically compliant, and 1 mark does not comply) were input into the XGBoost model to form the initial training model.

[0248] For XGBoost model training. Since XGBoost is a supervised learning algorithm, it is a variant of the GBDT algorithm. In the process of model training, the residual is focused, and the second-order Taylor expansion is used in the objective function and the regularization term is added. In the process of generating the decision tree, the idea of ​​precise greed is adopted. When looking for the best split point, the pre-sorting algorithm is used, and then all split points on all features are traversed, and the objective function gain of all samples after splitting according to these candidate split points is calculated. The features and candidate split points corresponding to the maximum gain are found and split. XGBoost is trained by addition. Each time a tree is trained by focusing on the residual, the final prediction result is the sum of all trees.

[0249] For a dataset containing n samples and m features, the prediction output of the XGBoost model is expressed as: Where f k Represents a regression tree, K is the number of regression trees, indicating that given an input x i , the output value is the sum of the predicted values ​​of K regression trees.

[0250] The data of 1645 items in the training set were input into the XGBoost model to form a training model. The model was then used to classify 411 items in the test set with an accuracy of 89.93% and a precision of 90.85%.

[0251] For XGBoost model optimization. Users upload new models to be reviewed to the review system and obtain the corresponding capability satisfaction, process compliance, system support, information compliance and standard coverage. Classify the models to be reviewed based on the trained model and obtain the initial technical system review results (3: fully compliant, 2: basically compliant, 1: non-compliant). Users can manually adjust the technical system review results to obtain the final review results. Then, the system automatically incorporates the project data into the training set of the XGBoost model. When the number of included items exceeds 50, the model is retrained and the retrained model is used to classify new projects to be reviewed.

[0252] The embodiment of the present application first conducts automated comparative analysis based on architecture model data, including extracting keywords in model node descriptions based on named entity recognition technology; obtaining the similarity of node names and node descriptions based on topic similarity calculation technology; obtaining the total similarity between nodes based on user-set weights; and displaying similar or violated nodes and relationships in a visual manner. Second, a method for conducting technical system review and analysis based on multiple perspectives, including capability satisfaction analysis, process compliance analysis, system support analysis, information compliance analysis, and standard coverage analysis. Third, a method for summarizing and generating technical system review results based on multi-perspective review and analysis results and using the XGBoost model. First, based on historical project data, an XGBoost model is trained; second, an initial review conclusion is given based on the multi-perspective review and analysis results of the model of the project to be reviewed newly uploaded by the user and the XGBoost model; then, the user can manually adjust the review conclusion and automatically incorporate the new review data into the training set of the XGBoost model; finally, the XGBoost model is iteratively optimized for the next review result summary.

[0253] Embodiment three:

[0254] Figure 7 Schematic diagram of a device for constructing an architecture model for automatic review of a technical system provided in an embodiment of the present application. Figure 7 As shown, the device comprises:

[0255] The review perspective predefinition module 701 is configured to obtain predefined review perspectives, which include capability perspective, business perspective, system perspective, information activity perspective and standard perspective;

[0256] The model data extraction module 702 is configured to extract corresponding model data according to the review perspective;

[0257] The review model construction module 703 is configured to construct a target perspective review model according to the model data and the review perspective item by item;

[0258] The review result acquisition module 704 is configured to obtain the multi-perspective review results of the target document based on the multiple target perspective review models.

[0259] It should be understood that the architecture model construction device for automatic review of technical systems in the embodiment of this specification can also execute Figure 1 to Figure 2 A method for executing an architecture model construction device for automatic review of a technical system in the present invention, and realizing an architecture model construction device for automatic review of a technical system in the present invention Figure 1 to Figure 2 The functions of the example shown will not be described in detail here.

[0260] Figure 8 Schematic diagram of a system architecture model review device for automatic review of technical systems provided by the embodiment of the present application. Figure 8 As shown, the device comprises:

[0261] The review model acquisition module 801 is configured to adopt the architecture model construction method for automatic review of technical systems described in the first aspect to obtain a multi-perspective review model of the target project to be reviewed;

[0262] A comparison result acquisition module 802 is configured to use a model node comparison algorithm to obtain a comparison result between the multi-view review model and a pre-built reference review model;

[0263] The quantitative review system process construction module 803 is configured to construct a corresponding quantitative review system process according to the predefined review content;

[0264] The review conclusion acquisition module 804 is configured to obtain the technical system review conclusion of the target project to be reviewed based on the quantitative review system process and the comparison result.

[0265] It should be understood that the architecture model review device for automatic review of technical systems in the embodiment of this specification can also execute Figures 3 to 6 A method for executing an architecture model review device for automatic review of a technical system in the present invention, and realizing an architecture model review device for automatic review of a technical system in the present invention Figures 3 to 6 The functions of the example shown will not be described in detail here.

[0266] At the same time, the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0267] Embodiment 4:

[0268] Fig. 9Schematic diagram of the electronic server 9 provided in the embodiment of the present application. Fig. 9 As shown, the electronic server 9 of this embodiment includes: a processor 901, a memory 902, and a computer program 903 stored in the memory 902 and executable on the processor 901. When the processor 901 executes the computer program 903, the steps in the above-mentioned various method embodiments are implemented. Alternatively, when the processor 901 executes the computer program 903, the functions of the modules / units in the above-mentioned various device embodiments are implemented.

[0269] The electronic server 9 may be a desktop computer, a notebook, a PDA, a cloud server, or other electronic servers. The electronic server 9 may include, but is not limited to, a processor 901 and a memory 902. Those skilled in the art will appreciate that Fig. 9 The electronic server 9 is merely an example and does not constitute a limitation of the electronic server 9 , which may include more or less components than those shown in the figure, or different components.

[0270] The memory 902 may be an internal storage unit of the electronic server 9, for example, a hard disk or memory of the electronic server 9. The memory 902 may also be an external storage server of the electronic server 9, for example, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic server 9. The memory 902 may also include both an internal storage unit of the electronic server 9 and an external storage server. The memory 902 is used to store computer programs and other programs and data required by the electronic server.

[0271] The processor 901 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 901 reads the corresponding computer program from the non-volatile memory into the memory and then runs it, forming a shared resource access control device at the logical level. The processor executes the program stored in the memory, and is specifically used to perform the following operations:

[0272] Obtaining predefined review perspectives, including capability perspective, business perspective, system perspective, information activity perspective, and standard perspective;

[0273] Extracting corresponding model data according to the review perspective;

[0274] Based on the model data, follow the review perspective item by item to build a target perspective review model;

[0275] Based on the multiple target perspective review models, a multi-perspective review result of the target document is obtained.

[0276] or,

[0277] Adopting the architecture model construction method for automatic technical system review described in the first aspect, a multi-perspective review model of the target project to be reviewed is obtained;

[0278] Using a model node comparison algorithm, obtaining a comparison result between the multi-view review model and a pre-built reference review model;

[0279] Build corresponding quantitative review system processes based on predefined review content;

[0280] Based on the quantitative review system process and the comparison results, the technical system review conclusion of the target project to be reviewed is obtained.

[0281] The above is as in this manual Figure 1 to Figure 2 The method or specification for constructing an architecture model for automatic review of a technical system disclosed in the embodiment shown Figures 3 to 6 The architecture model review method for automatic review of the technical system disclosed in the illustrated embodiment can be applied to the processor 901, or implemented by the processor 901. The processor 901 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor or an instruction in the form of software. The above processor can implement or execute the various methods, steps and logic block diagrams disclosed in the embodiments of this specification. The general processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiments of this specification can be directly embodied as a hardware decoding processor to perform, or a combination of hardware and software modules in the decoding processor to perform. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0282] Of course, in addition to software implementation, the electronic server of the embodiment of this specification does not exclude other implementation methods, such as logic devices or a combination of software and hardware, etc., that is to say, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0283] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units.

[0284] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. The computer program may include computer program code, and the computer program code may be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electric carrier signals and telecommunication signals.

[0285] Embodiment five:

[0286] The embodiment of the present specification also provides a computer-readable storage medium, which stores one or more programs, wherein the one or more programs include instructions, which, when executed by a portable electronic server including multiple application programs, can enable the portable electronic server to perform the above-mentioned Figure 1 to Figure 2The method or specification for constructing an architecture model for automatic review of a technical system disclosed in the embodiment shown Figures 3 to 6 The embodiment disclosed herein discloses an architecture model review method for automatic review of a technical system, and is specifically used to perform the following method:

[0287] Obtaining predefined review perspectives, including capability perspective, business perspective, system perspective, information activity perspective, and standard perspective;

[0288] Extracting corresponding model data according to the review perspective;

[0289] Based on the model data, follow the review perspective item by item to build a target perspective review model;

[0290] Based on the multiple target perspective review models, a multi-perspective review result of the target document is obtained.

[0291] or,

[0292] Adopting the architecture model construction method for automatic technical system review described in the first aspect, a multi-perspective review model of the target project to be reviewed is obtained;

[0293] Using a model node comparison algorithm, obtaining a comparison result between the multi-view review model and a pre-built reference review model;

[0294] Build corresponding quantitative review system processes based on predefined review content;

[0295] Based on the quantitative review system process and the comparison results, the technical system review conclusion of the target project to be reviewed is obtained.

[0296] In short, the above is only a preferred embodiment of this specification, and is not intended to limit the protection scope of this specification. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this specification should be included in the protection scope of this specification.

[0297] It should be noted that the above detailed descriptions are exemplary and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the art to which the present application belongs.

[0298] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should also be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0299] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the terms used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein.

[0300] In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or inherent to these processes, methods, products, or servers.

[0301] For ease of description, spatially relative terms, such as "above", "above", "on the upper surface of", "above", etc., may be used herein to describe the spatial positional relationship between a device or feature and other devices or features as shown in the figure. It should be understood that spatially relative terms are intended to include different orientations of the device in use or operation in addition to the orientation described in the figure. For example, if the device in the accompanying drawings is inverted, the device described as "above other devices or structures" or "above other devices or structures" will be positioned as "below other devices or structures" or "below other devices or structures". Thus, the exemplary term "above" may include both "above" and "below". The device may also be positioned in other different ways, such as rotated 90 degrees or in other orientations, and the spatially relative descriptions used herein are interpreted accordingly.

[0302] In the above detailed description, reference is made to the accompanying drawings, which form a part of this document. In the accompanying drawings, similar symbols typically identify similar components unless the context indicates otherwise. The illustrated embodiments described in the detailed description, drawings, and claims are not meant to be limiting. Other embodiments may be used, and other changes may be made, without departing from the spirit or scope of the subject matter presented herein.

[0303] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A system architecture model review method for automatic technical system review, characterized in that: include: Adopt the architecture model construction method for automatic review of technical system to obtain the multi-perspective review model of the target project to be reviewed; Using a model node comparison algorithm, obtaining a comparison result between the multi-view review model and a pre-built reference review model; Build corresponding quantitative review system processes based on predefined review content; According to the quantitative review system process and the comparison results, the technical system review conclusion of the target project to be reviewed is obtained; including: Obtain the capability satisfaction, process compliance, system support, information compliance and standard coverage of the target project to be reviewed to form a five-tuple, and input the five-tuple into the XGBoost model to generate the technical system review conclusion of the target project to be reviewed; Based on the technical system review conclusion, training the parameters of the XGBoost model to obtain an updated XGBoost model; The method for constructing an architecture model for automatic review of technical systems includes the following steps: Obtaining predefined review perspectives, including capability perspective, business perspective, system perspective, information activity perspective, and standard perspective; Extracting corresponding model data according to the review perspective; According to the model data, the target perspective review model is constructed item by item in accordance with the review perspective; wherein, corresponding to the capability perspective, the target perspective review model is constructed to include a capability decomposition model and an effect attribute model; corresponding to the business perspective, the target perspective review model is constructed to include a business activity process model and a business resource flow model; corresponding to the system perspective, the target perspective review model is constructed to include a system composition model, a system function description model and a system interaction model; corresponding to the information activity perspective, the target perspective review model is constructed to include an information activity process model and an information classification model; corresponding to the standard perspective, the target perspective review model is constructed to include a standard system model; Based on the multiple target perspective review models, a multi-perspective review result of the target document is obtained.

2. The architecture model review method for automatic review of technical system according to claim 1 is characterized in that: Corresponding to the system perspective, it also includes building a mapping relationship between system and capability, a mapping relationship between system function and activity, a mapping relationship between system and information activity, and a mapping relationship between system and business activity.

3. The system architecture model review method for automatic review of technical system according to any one of claims 1 to 2, characterized in that: The target perspective review model has description, paradigm and review granularity requirements that meet consistency requirements.

4. The system architecture model review method for automatic technical system review according to claim 1 is characterized in that: The model node comparison algorithm obtains the comparison result by calculating the model node similarity between the multi-view review model and the reference review model, including: A named entity recognition algorithm combining rules and conditional random fields is used to extract attribute information of multiple model nodes; Using a subject content similarity algorithm, respectively obtain the similarity of multiple model node attribute information between the multi-view review model and the reference review model; Based on the similarity of attribute information of multiple model nodes, the model node similarity between the multi-view review model and the reference review model is obtained.

5. The system architecture model review method for automatic review of technical system according to claim 4 is characterized in that: The model node attribute information includes a node name and a node description; and / or, based on the similarity of multiple model node attribute information, the node similarity between the multi-perspective review model and the reference review model is obtained, including: calculating the final model node similarity based on the weights of the node name and the node description.

6. The system architecture model review method for automatic technical system review according to claim 1 is characterized in that: Before building the corresponding quantitative review system process, it also includes: building a corresponding relationship between the review perspective and the review content, and the review content includes the ability to meet the situation, process compliance, system support, information compliance and standard coverage; among which, The capability perspective corresponds to the capability satisfaction situation; The business perspective corresponds to the process compliance; The system perspective corresponds to the capability satisfaction, the process compliance and the system support respectively; The information activity perspective corresponds to the information compliance situation; the standard perspective corresponds to the standard coverage situation.

7. The system architecture model review method for automatic technical system review according to claim 6 is characterized in that: According to the predefined review content, build the corresponding quantitative review system process, including: For each of the review contents, construct the corresponding review content attributes and the calculation indicator system corresponding to each review content attribute; Based on the attributes of the review content and the calculation indicators, a quantitative review process corresponding to the review content is constructed.

Images