CN107168762A - A kind of RUCM model consistency inspection methods based on body - Google Patents

Abstract

The invention relates to an ontology-based RUCM model consistency checking method, the method comprising the following steps: modeling domain knowledge, forming a domain ontology model including domain concepts, relationships between domain concepts and related attributes of domain concepts ; Combining the core concepts in the ontology database, establish and analyze the mapping relationship between the related concepts of the ontology database and the concepts of the RUCM model; finally, check the inconsistency defects in the RUCM model according to the defined consistency rules. The invention designs an extended RUCM model that supports ontology model checking, and provides an algorithm tool for an ontology-based RUCM model consistency checking method. The invention solves the defect of quickly and efficiently checking the inconsistency in the requirement by using domain knowledge; meanwhile, it can solve the repetitive labor and provide the accuracy and completeness of the requirement.

Description

An Ontology-Based RUCM Model Consistency Checking Method

technical field

The invention relates to the technical field of RUCM modeling, in particular to an ontology-based RUCM model consistency checking method.

Background technique

Most software requirements specifications are written in natural language. In the software requirements specification document, most domain concepts are defined and described by natural language, and the specific requirement items of software requirements are also written by natural language. Therefore, in these requirement descriptions, there are often ambiguous or ambiguous expressions. Such uncertain expressions often lead to disagreements and misunderstandings in different projects, which may lead to the failure of software projects. When writing software requirements specifications, in addition to using the traditional pure natural language description, a structured use case description template (RUCM modeling) based on rule-constrained natural language is also widely used. The method defines a rule template for describing software requirement use cases, and the description items contained in the specific rule description template are described in natural language. In the RUCM (Reastricted Use Case Modeling) modeling method, the requirement description is described using a restricted natural language and a use case specification template. Because software requirements analysts lack the domain knowledge of software requirements in the process of using RUCM modeling method, the final software requirements specifications often have inconsistent domain-related errors. For the same function, different software requirements analysts may have unequal and incomplete information based on their understanding of the domain knowledge of software requirements, and there may be great inconsistencies in the final written software requirements documents.

Contents of the invention

In view of the above analysis, the present invention aims to provide an ontology-based RUCM model consistency checking method to solve the problem of inconsistency of requirements existing in writing software requirements by existing software requirements analysts.

The method combines the domain knowledge expressed by the ontology model to analyze the consistency and completeness of the software requirements including the requirement text described in the natural language and the use case requirements described in the RUCM model. The ontology is a specific abstract expression of the knowledge contained in the domain. In the field of computer science research, the ontology model is a model that abstracts reality according to the concepts in the field, the relationship between concepts, and the attributes of concepts.

The purpose of the present invention is mainly achieved through the following technical solutions:

Including the following steps:

Step S1, establishing an ontology model library for RCUM model verification;

Step S2, establishing a conceptual mapping relationship from the ontology model library in step S1 to the RUCM model, forming a relationship mapping template;

Step S3: Define corresponding consistency check rules according to the mapping relationship established in step S2, and automatically detect inconsistencies described in the RUCM requirements through the rules, thereby forming a corresponding evaluation report.

The step S1 includes the following sub-steps:

Step S101, forming an abstract domain concept by analyzing domain-related background materials, and building a domain-related ontology concept library;

Step S102, analyzing the relationship between the core concepts in combination with the domain background of the target software system that needs to be realized by the software requirements, and obtaining the ontology relationship library;

Step S103, according to the relationship between domain core concepts, add domain concept specific attributes to obtain ontology concept attribute library;

Step S104, according to the above-mentioned ontology concept database, ontology relationship database and ontology concept attribute database, a template set of ontology model is formed, and then the template set is integrated to form an ontology model database for RUCM model verification, that is, a domain ontology database.

The step S2 includes the following sub-steps:

Step S201, determining and labeling the core concept attributes of the ontology;

Step S202, identifying and labeling concepts of software requirements;

Step S203 , on the basis of marking the core concept attributes of the ontology and the concepts of the software requirements, the corresponding rules between the two are obtained.

The step S3 includes the following sub-steps:

Step S301, defining corresponding consistency check rules according to the mapping relationship established in step S2;

Step S302, designing a corresponding algorithm for analyzing requirements, and automatically forming a corresponding evaluation report.

The present invention extends the traditional RUCM model. Aiming at three types of elements in the ontology model: concepts, relations between concepts and concept attributes, RUCM models that extend and support ontology model checking are designed respectively to form OntoRUCM (Ontology-based RUCM) models.

The modeling tool framework of the OntoRUCM model reuses three technologies of the Eclipse integrated development environment at the basic layer. First, the framework reuses the Eclipse workbench, and its functions include providing basic graphical interface elements and file resource management functions, namely Workbench UI and Workspace, two Eclipse core modules; second, the framework reuses the Eclipse plug-in development environment PDE, In order to realize the dynamic extension function at the meta-model level and the editor level; thirdly, the framework reuses some functions of the Eclipse modeling framework EMF to realize the compatibility of model files with other tools. The OntoRUCM model provides the corresponding extension class library in the extension layer, which is used to increase the extension points of the consistency check. The top editor layer uses an independent architecture of editors and plug-ins, and finally forms the overall design architecture of the OntoRUCM model.

The beneficial effects of the present invention are as follows:

The present invention defines the mapping relationship between the user-defined requirement model and the ontology extended by RUCM, clarifies the relevant important characteristics of the inspection rules, uses domain knowledge to inspect the user's requirements; standardizes the consistency and completeness inspection method of RUCM software requirements, It provides an automatic implementation method for the consistency detection method of software requirements.

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be apparent from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention may be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

Description of drawings

The drawings are for the purpose of illustrating specific embodiments only and are not to be considered as limitations of the invention, and like reference numerals refer to like parts throughout the drawings.

Figure 1 is a diagram of the ontology-based RUCM model consistency check process

Figure 2 is a diagram of the establishment process of the domain ontology library

Figure 3 is the architecture diagram of OntoRUCM requirements modeling tool

Figure 4 is a scheme diagram of the ontology-based RUCM model consistency rule tool implementation

detailed description

Preferred embodiments of the present invention will be specifically described below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of the application and are used together with the embodiments of the present invention to explain the principle of the present invention.

The invention discloses a method for checking the consistency of a requirement model in order to automatically solve the related problems of the inconsistency of the current software requirements and the violation of the software requirements and domain knowledge. The method combines the domain knowledge expressed by the ontology model to analyze the consistency and completeness of the software requirements including the requirement text described in the natural language and the use case requirements described in the RUCM model. The ontology is a specific abstract expression of the knowledge contained in the domain.

The consistency check method provided in this embodiment, as shown in Figure 1, includes the following steps:

Step S1, establishing an ontology model library for RCUM model verification.

The specific establishment steps of the ontology model library, as shown in Figure 2, include the following sub-steps:

Step S101 , by analyzing domain-related background materials, an abstract domain core concept is formed, and a domain-related ontology concept library is constructed.

Usually software engineering projects are accompanied by the use of a large amount of domain-specific knowledge and concepts. Standard field materials include textbooks introducing knowledge in this field, collections of relevant standards in the field, specifications, etc. For example: related to embedded operating systems, it is necessary to analyze textbooks, Wikipedia, and related standards and specifications in the field of operating systems to obtain core concepts in the field of operating systems, such as processes, threads, and partitions. The abstracted core concepts build a domain-related ontology concept library.

The ontology concept base can be formed by using NLP processing to extract concepts and manually add concepts.

Step S102 , analyzing the relationship between core concepts in combination with the domain background of the target software system to be realized by the software requirements, and obtaining an ontology relational library.

The ontology relational library describes the mutual dependencies among the core concepts in the domain. Such as embedded operating system, typical relationships in this field include: use and call relationship between functions, resource association relationship between process and operating system, kernel will contain partitions, this "inclusion" is the relationship between kernel and partitions, etc. .

Step S103, according to the relationship between domain core concepts, add domain concept specific attributes to obtain an ontology concept attribute library.

The concept attribute is the additional information of the concept, which plays the role of explaining and supplementing the concept, and is used to provide additional information and standards for checking the consistency. For example, in the field of operating systems, adding an attribute such as "required" to a process indicates that a process is an important core concept in all operating systems. The graphical interface is not required for all operating systems, so the attribute "not required" can be added to the graphical interface, indicating that the graphical interface is only a core concept attached to the operating system.

Step S104, according to the above-mentioned ontology concept database, ontology relationship database and ontology concept attribute database, a template set of ontology model is formed, and then the template set is integrated to form an ontology model database for RUCM model verification, that is, a domain ontology database.

The ontology model used in the present invention to express domain knowledge includes three parts: domain concept, domain concept relationship and domain concept attribute.

According to the characteristics of the present invention, the final ontology library needs to have a method of establishment and representation. The ontology concept elements designed and established by the present invention can be represented by the templates in the following table. Thus forming the view of the ontology model.

Table 1 Enumeration of ontology elements

concept node Describe related concepts in the domain relationship between concepts Describe the relationship between concepts Concept Node Properties Describe other attributes of concept nodes for software requirements analysis and derivation

Step S2, establishing a conceptual mapping relationship from the ontology model base in step S1 to the RUCM model, to form a relationship mapping template.

Specifically include the following sub-steps:

Step S201, determining and labeling the core concept attributes of the ontology.

For all the concepts in the domain ontology, not all concepts are equivalent. In specific software requirements, different concepts need to be marked with different attributes to distinguish the concepts defined in the ontology from the software requirements concepts. effect.

Step S202, identifying and labeling concepts of software requirements.

The RUCM use case of a software requirement item is relatively complex, including not only the data flow of the requirement, but also the control flow between data processing. Therefore, the semantics of RUCM is extended, and the concepts used or appearing in RUCM use cases are marked by extension.

Step S203 , on the basis of marking the core concept attributes of the ontology and the concepts of the software requirements, the corresponding rules between the two are obtained.

The corresponding rules have the following two modes:

1. The concepts in the ontology model and the concepts in the software requirements are completely consistent in wording, so it can be determined that they are the same concept;

2. The relationship between the concepts in the ontology model and the verbs in the context of the concept in the software requirements are consistent. It can be determined that there is a high probability that it is a statistical concept. The present invention uses the current natural language word segmentation tools to identify.

The present invention extends the traditional RUCM model. Considering that the ontology model contains three types of elements: concepts, relations between concepts and concept attributes. For these three types of elements, an extended RUCM model that supports ontology model checking is designed, and the extended RUCM model is called OntoRUCM (Ontology-based RUCM) from now on. The OntoRUCM model includes support for the identification of concepts in the ontology, and the attachment template for the concept in the ontology can be extended as an attachment template. The template includes the name of the concept specifically used in RUCM, the description of the concept, the dependence of the concept, and the relationship between the concept and the ontology. node reference. See Table 2 for the extension of the concepts in the ontology. After the OntoRUCM template is added, the domain concept used in the RUCM template can be associated with the ontology model, and such software requirements can be used as a basis for ontology recognition analysis. The OntoRUCM template needs to include support for identifying the relationship between concepts in the ontology, and needs to mark the relationship identified by the concept. The marked template can be used as the identification basis for the derivation of the relationship in the ontology, combined with the ontology-based RUCM extended view OntoRUCM template in Table 2 to represent the mapping relationship in the ontology.

Ontology-based extensions to RUCM model analysis increase the usability of the OntoRUCM model extensions.

Table 2 Extended OntoRUCM template

Corresponding to the RUCM modeling method, relevant personnel have done specific research and tool implementation. The OntoRUCM model proposed in the present invention is an extension of the traditional RUCM modeling method, which not only includes the expansion of the attachment to the template, but also requires the user to add the label of the ontology concept while writing the RUCM requirement. This kind of actual tool writing is implemented according to the architecture in Figure 3. As shown in Figure 2, the modeling tool framework of the OntoRUCM model reuses three technologies of the Eclipse integrated development environment at the basic layer. First, the framework reuses the Eclipse workbench, and its functions include providing basic graphical interface elements and file resource management functions, namely Workbench UI and Workspace, two Eclipse core modules; second, the framework reuses the Eclipse plug-in development environment PDE, In order to realize the dynamic extension function at the meta-model level and the editor level; thirdly, the framework reuses some functions of the Eclipse modeling framework EMF to realize the compatibility of model files with other tools. The OntoRUCM model provides the corresponding extension class library in the extension layer, which is used to increase the extension points of the consistency check. The top editor layer uses an independent architecture of editors and plug-ins, and finally forms the overall design architecture of the OntoRUCM model.

Step S3: Define corresponding consistency check rules according to the mapping relationship established in step S2, and automatically detect inconsistencies described in the RUCM requirements through the rules, thereby forming a corresponding evaluation report.

Specifically include the following sub-steps:

Step S301, define corresponding consistency check rules according to the mapping relationship established in step S2.

The RUCM model of the software requirement in the present invention includes the reference from the concept in the software requirement to the ontology model, and then uses the corresponding rules from the ontology concept to the software requirement. Such analysis rules mainly include the definition of the following related characteristics:

consistency. The consistency means that in the same field, a concept, or two or more terms of the same concept cannot be confusedly used, such as the standard full name of a concept, its abbreviation, and various aliases or translations cannot be confusedly used . The relationship referenced by different concepts must be consistent when used.

integrity. The complete description refers to whether the core concepts in the domain are fully described in the software requirements. At the same time, relationships in software requirements that use concepts but do not add concepts are also considered incomplete.

Indexes for analyzing software requirements based on the ontology model include consistency and completeness of requirements, and the two inspection indexes are based on consistency inspection rules and integrity inspection rules.

1. Software Requirements Consistency Rules

The consistency of software requirements describes that the same concepts in the field cannot be confusedly used in software requirements, and the relationship referenced by different concepts must be consistent when used.

●The same concept of the ontology model, the RUCM model cannot have inconsistent vocabulary when using

●The conceptual relationship of the ontology model, the RUCM model cannot be misused when using it

The references to the same concept in the ontology model in the RUCM model must be completely reduced to one class

2. Software Requirements Integrity Rules

The domain knowledge described in the ontology can mark out the complete set of core concepts in a specific domain. Domain concepts in ontology models have different attributes.

The domain concepts referenced in the RUCM model should be defined in the domain ontology

●Relational flow of concepts in the ontology model, concepts in the RUCM model must be fully referenced when used.

Because the domain ontology is the abstraction of all the core concepts of a specific domain, and the requirements information base is the abstraction of the expressed content of the system requirements. Corresponding analysis algorithms must be established among the two abstractions to detect defects and deficiencies in system requirements.

The principles of the rules for analyzing software requirements mainly include the rules for checking the consistency of the requirements and the integrity of the requirements.

The rule of the consistency of the requirements is mainly to solve the problem that the same concept appearing in the requirement information base in the domain ontology must ensure that the concepts are not misused and prevent the concepts in the requirement information base from being confused.

The rules of the completeness of the requirements are mainly to solve the core concepts in the domain ontology must have relevant descriptions in the requirements document, otherwise, it can be considered that the requirements document is only for a specific part of the domain and does not have completeness.

The rules of the requirements can also be refined and improved in subsequent work.

Step S302, designing a corresponding algorithm for analyzing requirements, and automatically forming a corresponding evaluation report.

Based on the relevant rules for analyzing requirements, the present invention designs corresponding algorithms for analyzing requirements, performs analysis in the established domain ontology and requirements information database, and automatically detects inconsistencies described in RUCM requirements, thereby forming a corresponding evaluation report. The requirement information base is a collection of requirement information generated by extracting two types of software requirements, free text requirements and RUCM requirement models.

The algorithm for analyzing requirements uses the following three steps:

1. Traverse the requirements information base to find the requirements that do not meet the rules.

2. Traverse the domain ontology to find incomplete requirements in the requirements information base.

3. Make statistics on the found requirements and generate a corresponding analysis report.

The algorithm for analyzing requirements can be further optimized according to the specific ontology model and the details of the requirements information base.

The present invention provides a tool implementation design embodiment of an algorithm aiming at the related requirements, as shown in FIG. 4 ,

The input of the algorithm tool is software requirements and domain ontology;

The software requirements include free text requirements and RUCM models;

The free text requirements are a summary of the requirements in each requirement item, and the rule base is a predefined verification rule. The rules contained in the rule base and user-defined rules are parsed in the rule parser together, and then the text rules are verified. Analytical filtered free-text requirements output by the definer;

The general requirement format of the semi-formal requirement output outputted by the text rule definer after the free text requirement is defined by the text rule definer, and the semi-formal requirement generated by the sentence rule as a part of the input of the rule executor;

The RUCM model is expanded to the model generator to form an OntoRUCM model that can be recognized by the rule executor through the expansion of the UCMeta converter, as another part of the rule executor's input;

In the construction process of the domain text, the domain text is passed through the concept analysis ruler to obtain the general format file of the domain concept text, and after the relationship analysis ruler is used to obtain the general format file of the domain relational text, the two together constitute the domain ontology knowledge and serve as a rule executor another input of

The rule executor automatically detects the inconsistency described in the RUCM requirements through the rules, the verification algorithm included in the rule executor is a verification of the consistency verification of the requirements input by the rule executor, and the analysis algorithm is the verification between the requirement rules Find and evaluate the cited analysis, and then output the analysis results to form a corresponding evaluation report.

To sum up, the embodiment of the present invention provides a method for checking the consistency of the requirement model. The method defines the mapping relationship between the user-defined requirement model and the ontology extended by RUCM, and clarifies the relevant important characteristics of the verification rules. Domain knowledge is used to test the user's needs; standardizing the consistency and completeness checking method of RUCM software requirements provides an automatic implementation method for the consistency testing method of requirements.

The above is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of changes or modifications within the technical scope disclosed in the present invention. Replacement should be covered within the protection scope of the present invention.

Claims (7)

1. a kind of RUCM model consistency inspection methods based on body, it is characterised in that comprise the following steps：

Step S1, set up ontology model storehouse for RCUM model testings；

Ontology model storehouse in step S2, establishment step S1 forms relationship map template to the concept mapping relations of RUCM models；

Step S3, according to the step S2 mapping relations set up corresponding consistency check rule is defined, by regular automatic Detect inconsistent described in RUCM demands, so as to form corresponding assessment report.

2. the RUCM model consistency inspection methods according to claim 1 based on body, it is characterised in that

The step S1 includes following sub-step：

Step S101, the background material by analysis field correlation, form abstract field key concept, build field correlation Ontological concept storehouse；

Between step S102, the domain background analysis key concept for the destination software systems realized with reference to required for software requirement Relation, obtains ontological relationship storehouse；

Step S103, according to the relation between the key concept of field, addition field key concept particular community obtains Ontological concept Attribute library；

Step S104, according to above-mentioned Ontological concept storehouse, ontological relationship storehouse and Ontological concept attribute library, form the mould of ontology model Plate set, is then integrally formed the ontology model storehouse i.e. field ontology library for RUCM model testings by template set.

3. any RUCM model consistency inspection methods based on body according to claim 1 or 2, its feature exists In,

The step S2 includes following sub-step：

Step S201, the key concept attribute for determining and marking body；

Step S202, identification and marking software demand concept；

Step S203, on the basis of the key concept attribute and the concept of software requirement for having marked body, obtain between the two The rule of correspondence.

4. the RUCM model consistency inspection methods based on body according to claim 3, it is characterised in that

The step S3 includes following sub-step：

Step S301, according to the step S2 mapping relations set up define corresponding consistency check rule；

The algorithm of the corresponding analysis demand of step S302, design, automatically forms corresponding assessment report.

5. the RUCM model consistency inspection methods based on body according to claim 1, it is characterised in that

The step S3 includes following sub-step：

Step S301, according to the step S2 mapping relations set up define corresponding consistency check rule；

The algorithm of the corresponding analysis demand of step S302, design, automatically forms corresponding assessment report.

6. any RUCM model consistency inspection methods based on body according to claim 1,2,4,5, it is special Levy and be,

RUCM models in the step S2 are OntoRUCM models, concept of the OntoRUCM models in body, general The corresponding template of RUCM models is extended by relation and concept attribute between thought, is added to RUCM model consistency inspections Mapping relations.

7. the RUCM model consistency inspection methods according to claim 6 based on body, it is characterised in that described The modeling tool framework of OntoRUCM models has been multiplexed Eclipse workbench, Eclipse developing plug environment in basal layer PDE, Eclipse modeling framework EMF partial function, extension class libraries is provided in extension layer, and editor has been used in editing machine layer Device and the separate framework mode of plug-in unit；The workbench of the Eclipse provides fundamental figure interface element and file resource Management function, the developing plug environment PDE of the Eclipse realizes the dynamic expansion work(of meta-model aspect and editing machine aspect Can, compatibility of the Eclipse modeling frameworks EMF to implementation model file and other instruments, the extension class libraries is used for Increase the extension point of consistency check.