Detecting Conflicts and Inconsistencies in Web Application Requirements

Abstract

Web applications evolve fast. One of the main reasons for this evolution is that new requirements emerge and change constantly. These new requirements are posed either by customers or they are the consequence of users' feedback about the application. One of the main problems when dealing with new requirements is their consistency in relationship with the current version of the application. In this paper we present an effective approach for detecting and solving inconsistencies and conflicts in web software requirements. We first characterize the kind of inconsistencies arising in web applications requirements and then show how to isolate them using a modeldriven approach. With a set of examples we illustrate our approach.

Software Requirements Conflict Identification: Review and Recommendations

“…While Urbieta et al [19], [20] proposed a model-driven approach to detect requirement conflicts in Web applications in early stage of software development. The approach starts automatically listing the candidate structural and navigational conflicts by structural analysis using the Navigational Development Techniques (NDT) model.…”

“…They have proposed different approaches for conflict analysis and detection. For automatic techniques, the conflict analysis approach can be classified into the following four groups: (1) semantic approach for technique that use ontology like [13], [18]; (2) syntax approach when a syntax analysis is done for requirement specification like [9]; (3) graphical analysis when a specific model used like [20], [19], [27]; and (4) tractability approach when tractability technique is used like [26], [8].…”

“…3. Analysis results based on type of requirements [13], [18]; structural model if graphical analysis is used like [20], [19], [8], [29]; or formalization methods which are different from schematic version in [3], structure requirements in [6], two canonical forms in [16], semi-formal ontology driven domain-special requirement language in [17], DSL in [20] and [19], OWL in [27], and EBNF in [21]. While some proposed techniques used more than one type of representation like [20], [19] which used ontology and DSL as formalization method, [27] used ontology, OWL as formalization method and goal graph as a structural model.…”

Identifying and Modelling Complex Workflow Requirements in Web Applications

“…Semantic equivalences between requirements are detected for warning requirement analysts. For more information see [13]. In the case of adaptation requirements, a previous weaving is performed among both kind of requirements obtaining instantiated PS.…”

“…The former type corresponds to a difference in the data belonging to a business concept meanwhile the latter defines a difference in the way interaction occurs. For more information see [13]. User story US3 proposes a slightly different workflow with respect to the one presented in Figure 6 corresponding to US1.…”