Earticle

AI 기술의 비약적인 발전은 사회 전반에 걸쳐 다양한 기능을 수행할 수 있는 모델들이 등장했다. 이에 따라 AI 모델들은 점차 대규모화되고 복잡한 구조를 갖추게 되었으며, 그 결과 이들의 의사결정 과정은 인간이 직관적으로 이해하기 어려운 수준에 이르렀다. 이러한 복잡성을 해소하고자 설명 가능 한 인공지능(XAI, eXplainable Artificial Intelligence)에 대한 연구가 활발히 진행되고 있으며, 다양 한 설명 모델들이 제안되고 있다. 설명 가능한 인공지능 모델은 각기 다른 접근법을 통해 모델의 판단 근거를 설명하려고 시도하고 있지만, 제공되는 설명의 대상, 평가 방식, 그리고 설명이 충족해야 할 요 구사항 등에 대해서는 여전히 명확한 기준이 부재한 실정이다. 이로 인해 AI 시스템을 실제 사회 기 반 시스템에 적용하는 데에는 여러 제약이 존재한다. 또한, 기존 XAI 모델들은 직관적인 설명을 제공 하는 데 집중해왔으나, 그 설명에 대한 정량적‧정성적 평가 또는 명확한 요구사항 정립은 부족했다. 특 히, 공공 및 주요 사회 시스템에서 신뢰할 수 있는 XAI 개발이 요구되고 있음에도 불구하고, 어떤 기 능적 또는 비기능적 요구사항을 충족해야 하는지에 대한 논의는 미비하다. 본 연구는 텍스트 및 이미 지 도메인 전반을 아우르며, XAI 모델이 갖춰야 할 핵심 요구사항을 정의하고 이를 분류하는 것을 목 표로 한다. 이를 위해 다양한 XAI 모델들을 분석하고, 이들이 사용하는 핵심 요소와 한계를 식별하며, 소프트웨어공학적 관점에서 이를 체계적으로 해석한다. 특히 Goal-Question-Metric(GQM) 접근법 을 적용하여 각 모델의 신뢰성, 설득력, 효용성에 대해 평가하고 요구사항 도출을 위한 근거를 마련한다.

본 연구에서는 Normalizing Flow 기반 비지도 이상탐지 모델인 Multi-Scale Flow의 특징추출기에 다양한 모델을 적용하여 포트홀 이상탐지 성능에 미치는 영향을 정량·정성적으로 실험을 통해 분석한 다. 성능 평가는 이상 존재 여부를 판단하는 Detection AUROC와 이상 위치를 판단하는 Localization AUROC를 결합한 평균 성능을 기준으로 수행되었다. 실험 결과, 평균 AUROC 기준 85.84를 달성한 Swin Transformer가 포트홀 이상탐지에 있어 가장 높은 성능을 보임을 확인하였다. 또한 포트홀 이상 탐지에 가장 강인한 성능을 가지는 Swin Transformer 모델의 추가적인 성능 고도화를 위해 일반적으 로 사용하는 데이터 증강 기법과 Mix-Up, Cut-Mix 적용 여부에 따른 성능 분석을 진행하였다. 실험 결과, 데이터 증강 기법의 적용을 통해 포트홀 이상탐지 성능에 있어 소폭 상승하는 경향을 확인하였다.

소프트웨어 기반의 시스템 설계과정에서 설계자가 고려해야 하는 요소들이 다양해지면서 시스템 설계가 점점 어려워지고 있다. 다양한 설계 요소들이 존재하지만 사용자의 특성 및 직무, 사용가능한 정보기술의 특성 등이 핵심적인 요소로 간주된다. 또한 정보기술이 발달하면서 인간과 시스템의 상호작용이 점점 인지적인 특징을 지니게 되었다. 따라서 사용성 높고, 효율적이면서 안전한 소프트웨어 기반의 시스템을 개발하기 위해서는 시스템 설계자가 사용자의 인지적인 요구사항 및 그들의 직무를 시스템 설계과정에서 체계적으로 다룰 수 있어야 한다. 그러나 소프트웨어 공학, 시스템 공학 및 인간-컴퓨터 상호작용 등에서의 전통적인 시스템 설계 방법은 이러한 설계자의 설계활동을 효과적으로 지원하는데 한계가 있었다. 그 대안으로 인지시스템공학(cognitive systems engineering; CSE)은 인간중심의 설계철학을 바탕으로 소프트웨어 기반의 복잡한 시스템 설계과정에서 설계자의 활동을 체계적으로 도와줄 수 있는 유용한 개념과 방법을 제공해주고 있다. CSE는 원래 사람이 실시간으로 감시 및 제어해야 하는 복잡한 사회기술적 시스템(예: 원자력발전소 및 공항관제소)의 분석, 설계 및 평가를 위해 태동한 학문이다. 그러나 CSE에서 제공하는 이론적 및 방법론적 프레임워크는 소프트웨어 기반의 시스템을 설계하는 데에도 유용하게 활용할 수 있는 충분한 가능성을 갖고 있다. 이 논문은 CSE의 근간을 이루는 핵심 개념 및 원칙을 고찰하고 소프트웨어 기반 시스템 설계에의 활용가능성 및 그 방안을 논의한다.

The digitization of traditional medical knowledge in Tanzania will greatly enhance its preservation and dissemination. This is especially important given the challenges facing the current methods of preserving and managing such knowledge. This study presents the requirements engineering approaches and requirements for a web-mobile application that would successfully digitize indigenous knowledge of phytomedicine and relevant practitioners licensing and registration processes. To establish the requirements of such a digital system application, the study sought the opinion of 224 stakeholders whose suggestions were used to analyze and model the requirements for designing such a web-mobile tool. The study was carried out in Arusha, Kagera and Dar es Salaam regions of Tanzania which involved ethnobotanical researchers, herb practitioners, curators from herbaria and registrar officers from Traditional and Alternatives Health Practice Council. Structured interview, survey, observation and document review were employed to find out the basic functional and non-functional requirements for possible designing and implementation a web-mobile application that would digitize indigenous knowledge of medicinal plants. The requirements were modelled using the use case and context diagrams. Finally, the study came up with a list of items for both functional and non-functional requirements that can be used as guidelines to develop a web-mobile application that will capture and document traditional medical knowledge of medicinal plants in Tanzania and, enabling relevant authorities to regulate and manage stakeholders.

Most of the data warehouse (DW) requirements engineering approaches have not distinguished the early requirements engineering phase from the late requirements engineering phase. There are very few approaches seen in the literature that explicitly model the early & late requirements for a DW. In this paper, we propose an AGDI (Agent-Goal-Decision-Information) model to support the early and late requirements for the development of DWs. Here, the notion of agent refers to the stakeholders of the organization and the dependency among agents refers to the dependencies among stakeholders for fulfilling their organizational goals. The proposed AGDI model also supports three interrelated modeling activities namely, organization modeling, decision modeling and information modeling. Here, early requirements are modeled by performing organization modeling and decision modeling activities, whereas late requirements are modeled by performing information modeling activities. The proposed approach has been illustrated to capture the early and late requirements for the development of a university data warehouse exemplifying our model's ability of supporting its decisional goals by providing decisional information.

The recent introduction of information and communications technology (ICT) in the maritime sector has increased the number of contacts with external networks, increasing cybersecurity threats. Accordingly, the importance of cyber resilience to ships is being emphasized. The International Association of Classification Societies (IACS) emphasized holistic cyber resilience in its Unified Requirements (UR) E26, which lays out five functional elements and requirements for ships: identify, protect, detect, respond, and recover. However, UR E26 provides requirements but no practical techniques. This paper proposes a framework that integrates IACS UR E26 and the MITRE Cyber Resilience Engineering Framework (CREF) to provide a systematic way to enhance and manage ship cyber resilience. Since the proposed framework provides a practical description of the requirements of UR E26, it can be used for training cyber resilience response to ships and can improve cyber resilience through continuous improvements.

시스템엔지니어링은 오늘날의 기업에 무척 중요하며 요구공학은 전반적인 과정상 중요한 단계이다. 요구공학은 고객 으로부터 요구사항을 도출하고 문서화하는 것으로 시스템 개발의 첫 번째 단계이며, 시스템을 설계 제작하고 시험하고, 운영 유지하는 모든 과정이 요구사항에 관련되어 있어 프로젝트 성공의 매우 중요한 사항으로 평가된다. 프로젝트 성공 을 위해 빈약한 요구사항을 최소화하도록 객체 지향 요구공학이 제안되어 졌다. 객체 지향 요구공학은 요구사항 각각에 대하여 기능성은 물론 제품과 프로세스에 대한 정보를 통합하여 관리하는 접근방식이다. 본 논문은 객체 지향적 요구사 항 관리의 모범인 OCH(Operations Concept Harbinger)의 개념 적용을 통하여 사용자 요구 수집단계에서 다양한 이해관 계자의 의견을 동시에 통합적으로 수렴하여 사용자 요구를 요구사항으로 전환하고 요구사항별로 추적함으로써 무기체계 획득을 위한 정확한 요구사항 개발과 관리를 가능케 하고 사용자 요구누락 및 왜곡을 방지하며 통합적 프로젝트 관리가 가능토록 제안하는 것이다.

System engineering is critical in today's industry and requirements engineering is an important stage of overall process. Requirements Engineering is the initial step of system development activity in which the requirements from the customer are elicited and documented. This activity is very much vital for the success of the project because all of activities depends upon requirements engineering such as designing, implementation, testing, operation and maintenance. The development process begins by clarifying the need and then articulating the need as a high level solution. In order to minimize the poor requirements and to sure successful projects, Object-Oriented requirements engineering was proposed. Object-Oriented requirements engineering is an approach to encapsulating information about the process and product, as well as functionality in to a requirements object. This paper proposes using the concept of an Operations Concept Harbinger(OCH) that is the prototype of Object-Oriented requirements engineering to develop the requirements with consolidating simultaneously the opinions of various stakeholder in the customer level and trace accurately the transition from User needs to requirements. Then the customer can secure the accurate requirements to meet their needs and traceability from user needs to requirements.

Software requirements should be accurately captured in the form of user stories so that development teams are able to meet customer expectations for the software. This study investigates the various challenges in capturing requirements of the agile scrum methodology and proposes some improvements. A user story writing workshop was conducted based on a standard set of business requirements. The resulting user stories were analyzed for correctness and problem areas were identified. Then a separate industry survey was conducted to confirm that the problem areas found in this workshop are indeed the common problem areas for industry practitioners. A set of improvements were then identified to address the problem areas. A second workshop was then conducted to check whether the improvements actually result in better user story writing. The results of the second workshop suggest that the improvements were quite effective in resolving most of the problem areas. They can thus be adopted by industry practitioners as best practices.

Errors in requirements elicitation cause much more expensive errors in later phases of software engineering. A major source of errors is the ambiguity of the natural languages initially used to write the user requirements. Most of the research studies in the literature have focused on ambiguity detection and correction after the user requirements are written. A better approach would be avoiding the introduction of ambiguities while the user requirements are being written to reduce time and cost. Unfortunately, this problem has been overlooked in the literature except for few research studies concerned with the English language. This paper is concerned with the Arabic language due to its complex structure and because the ambiguity of Arabic user requirements is a serious problem hindering the production of quality software in the middle east. The paper studies main sources of ambiguities in the Arabic language in addition to best practices in writing Arabic user requirements to avoid the introduction of ambiguities as suggested by interviewed Arabic software engineers or as suggested in the literature for English user requirements. Accordingly, the paper suggests a set of best practices suitable for Arabic user requirements based on both practical experience and theoretical background. Examples are provided for justification. Human subject testing and empirical evaluation have shown the effectiveness of the proposed practices.

Traditional requirements analysis is focused on capture functions of the system that are being built from scratch. As the size and complexity of process-aware information system continues to grow, the use of COTS (Commercial off-the-shelf) software packages is being viewed as a solution. Use of COTS packages accelerates the implementation process to some extent, however, it puts enterprises into a dilemma: whether to adapt as-is business processes or to customize software packages. In order to help the enterprises get out of the dilemma, a methodology is proposed to automatically elicit software requirement using best practice as domain knowledge. The methodology involves three phases: business modeling, gap detecting and gap bridging. Firstly, VPML is employed to describe as-is business process, Secondly, semantics computing technology is employed to analyze the gap between enterprise needs and COTS software capacity. At last, Goal Reasoning Technology is employed to encourage enterprise users and software vendors take participate in decision making process together.

Requirements need to be validated at the early phase of the software development to avoid errors such as inconsistency, incompleteness and incorrectness. Drawn from this argument, a requirements validation process needs to consider Consistency, Completeness and Correctness (“3 Cs”) for the production of a quality software specifications. This paper provides a review of requirements validation and consistency management based on the existing literature in order to identify the gaps in the existing knowledge on the process of software requirements specifications. This paper begins with a review of the definitions of the 3Cs,upon which the understanding of the 3Cs is derived. Next comprehensive review of related works on the identified consistency management techniques: traceability and analysis approaches are then presented. This is supported with a heat map representations of the related research on the types of contributions, techniques, specifications and semantics used in consistency management. Since semi-formal specifications were found as the most common representation of the requirements, the types of models used as semi-formal specifications to represent the requirements were also discussed. Overall, this paper identifies the various gaps existing within the process of validating and managing the consistency of requirements to avoid re-inventing the wheels in the diverse and comprehensive knowledge of requirements engineering.

This paper describes an agent-oriented requirements engineering approach. Agent-oriented techniques represent an exciting new means of analysing, designing and building complex software systems. They have the potential to significantly improve current practice in software engineering and to extend the range of applications that can feasibly be tackled. i * can be used to model social dependencies between agents and how process design choices affect the agents’ goals. Agent-oriented approaches are becoming popular in software engineering, both as architectural frameworks, and as modeling frameworks for requirements engineering and design. i* is an informal diagram-based language for early -phase requirements engineering that supports the modeling of social dependencies between agents and how process design choices affect the agents’ goals both functional and non-functional. ConGolog is an expressive logic-based formalism for specifying processes that involves multiple agents. The two formalisms complement each other well, and in this work, we used the combination of the two frameworks in requirements engineering. The i* SR-diagram language is extended with process specification annotations, which allow the SR model of a system to be refined and then mapped into a ConGolog model. The mapping must satisfy a set of mapping rules, which ensure that it specifies which elements in the two models are related and that the models are consistent.

Requirement elicitation is a critical activity in the requirement development process and it xplores the requirements of stakeholders. The common challenges that analysts face during licitation process are to ensure effective communication between analyst and the users. Mostly rrors in the systems are due to poor communication between user and analyst. This paper roposes an improved approach for requirements elicitation using paper prototype. The paper rogresses through an assessment of the new approach using student projects developed for arious organizations. A case study project is explained in the paper.

Requirements elicitation is the important activity in requirements engineering process and it involves in discovering needs, capturing and gathering software requirements from the stakeholders. However, there are few problems faced by requirements engineer in order to carry out requirements elicitation process. The requirements elicitation is lacking in terms of various issues; stakeholders involvement, support tool and complete documentation so as to achieve on time delivery. This process becomes easier if the requirements engineer and the other stakeholders shared a preferred method even though they are in distributed places, geographically. In our case, the requirements engineer (courseware developer) and other stakeholders (schoolteachers) agreed to use Focus Group Discussion (FGD). Therefore, a prototype of Focus Group Discussion for requirements elicitation tool, FGD-RElicit is proposed to assist all stakeholders involved and aims to tackle some of the difficulties in conducting requirements elicitation process. FGD-RElicit is aimed to improve the requirements elicitation process, enhances the software requirements specification (SRS), and feasibly produce much better software. This paper presents the multi viewpoint approach adopted, FGD-RElicit details feature, evaluation and the results of its application in a case study.

Visibility in search engines is a competitive advantage of commercial websites. In this context, Search Engine Optimization (SEO) techniques are devised to enhance the ranking of websites in search engine results. The existing approaches toward SEO are primarily modifying the content or the structure of websites during the post-development activities. These modifications can introduce new defects to the websites. To tackle this problem, we have proposed caring for visibility in the requirements of commercial websites. We further contribute a goal-based framework for modeling and description of visibility in requirement engineering phase. The framework lays a foundation for automated analysis of visibility in commercial websites.

This paper presents RE4Gaia, which is a requirements modeling approach for the development of multi-agent systems that extends the Gaia methodology. The approach focuses on dealing with the organizational structure as a means to adequately capture and understand required roles and associated functions in the context of an organization prior to the analysis and design of MAS using Gaia. In addition, a traceability framework is introduced to facilitate moving from the requirements models to the analysis and design models proposed in Gaia.

This paper gives an overview of various standard lifecycle development processes. It then provides a roadmap for developing security-critical projects using Rational Unified Process as a framework for development. The Security Quality Requirements Engineering (SQUARE) methodology provides a way to address security issues early in the development lifecycle. SQUARE can be more effective when it fits into an organization’s existing development process. Hence this paper describes a way to fit the SQUARE methodology into the Rational Unified Process.

The principal role of Data warehouses in making strategy decisions makes its quality crucial for organizations. Quality of data models for data warehouse has significant effect on the quality of data warehouse. Therefore, we need methods, models and tools to have quantitative and objective measurement of quality of data models for data warehouses. Lack of a set of defined measures made us to define a set of metrics that can be used to measure the complexity, understandability, believability, usability, accessibility, data currency and modifiability which are in turn defined as quality characteristics for data warehouse systems. The proposed metrics may act as objective indicators of the quality of data warehouse models and also help the developers to select quality data model among various semantically equivalent models. The other advantage of these metrics is that they are available during early phase of software development life cycle.

For the past decade, agile methodologies have been part of a shift from a predictive to adaptive approach towards software development in many organisations. This paper views agile requirements engineering as a kind of social engineering, and explores the potential of the Social Web in facilitating it. In doing so, the interplay between stakeholders, the salient activities of the agile requirements engineering process, and Social Web Applications is illustrated by means of representative examples. The implications towards other agile project artefacts and agile requirements engineering education are highlighted.

In the requirement analysis phase, good quality requirements are needed to develop the foundation of good quality software. This paper proposes quality attributes from the SQA activities in requirement phase from the end users’ perspective in an e-commerce application. By analyzing SQA activities in requirement engineering process, we found out five quality attributes that most affect customer’s satisfaction: functionality, security, usability, reliability and efficiency. This paper describes the quality attributes that are gained from the requirements elicitation, requirements documentation, requirements validation, negotiation and the requirements management planning activities. The results show that functionality, security, usability, reliability and efficiency affect e-commerce customers’ satisfaction. Most of the online shopping websites comply with customer requirement and requirement expectations.