Fundamental Research on Detecting Contradictions in Requirements: Taxonomy and Semi-Automated Approach

Abstract

Requirements documents can contain several thousand individual requirements. They must be error-free to avoid unnecessary complications and costs in the later product development stages. An important part of this is to identify contradictions between two requirements. The first step is therefore to define what contradictions are and in what form they can occur in requirement documents. In this paper the scientific theories regarding contradictions are discussed, concerning to their usefulness for the topic. In doing so, the Aristotelian Logic proved to provide the best basis for an application in the Requirements Engineering context. Based on this theory, we have created specific subtypes of contradictions to match them to the requirements engineering field. The identification of these subtypes is done by a formalization of the requirement sentences and a subsequent analysis by means of simple questions. To validate the method, industrial requirement documents were searched for contradictions. For each detected type of contradiction, we present an example of the detection process. Thereby, we show that the method is easy to apply and may also be used by non-specialists. Thus, our method provides a taxonomy as a basis for further research on automated contradiction detection as well as on automated quality analysis of requirements documents.

Automated Condition Detection in Requirements Engineering

“…Moving in this direction, this paper creates the possibility to identify contradictions between two requirements automatically. The approach is, that if two requirements had the same condition and in the effect the same variables, but different actions, this would indicate a contradiction (Gärtner et al 2022). This research contributes a building block for this approach by identifying these corresponding verbal expressions.…”

“…These terms are also referred to as constituents (words or groups of words that function as a unit). Unlike Fischbach's terminology (2021a) which we adopted in our previous paper, (Gärtner et al 2022), we now propose the following terms: variable and action. For example, in the sentence If the threshold is reached, the controller must limit the speed decay, we must first differentiate between the condition (If the threshold is reached) and the effect (the controller must limit the speed decay), as shown in Figure 1.…”

“…Our work aims at providing the necessary building blocks for an automated quality assurance of specification documents. In a previous study (Gärtner et al 2022), we presented different types of contradictions that can occur in requirements specifications and developed a method to classify contradictions. In this context, the question of whether the requirement includes a condition or not plays a central role.…”

“…Create a basis to identify contradictions automatically. According to our previous study, the conditions and effects must be compared regarding their verbal expressions to find contradictions (Gärtner et al 2022). For this purpose, we compare two natural language processing (NLP) techniques: the first one is a model that consists mainly of grammatical rules that are directly embedded into the source code.…”

Towards an automatic contradiction detection in requirements engineering

“…Distinguishing itself from existing literature, our approach does not rely on classification for a specific, limited dataset; instead, it adheres to a widely accepted and rigorously tested systematic model. Therefore, we draw upon the logical philosophy of Aristoteles (1986), as elaborated in our previous paper (Gärtner et al 2022). Aristotle's logic, known as term logic, traditional logic, or formal logic, centers on the law of non-contradiction (LNC) (Horn 2018), forming the basis for our understanding.…”