Aug 2026· 2026 IEEE 34th International Requirements Engineering Conference Workshops (REW)· pp. 134-140· 0 citations· 31 references
Abstract
As software systems become more complex, end users' needs for explanation of certain system aspects also increase. A useful method for eliciting the needs of large user groups are common CrowRE methods, such as feedback mining in app reviews. However, these methods are not ideal for addressing the need for explanation, as minor issues rarely result in explicit feedback. To address this, we seek to explore new modalities for identifying the need for explanations among large user groups. To this end, we conducted an exploratory user study in which we observed 30 participants as they encountered situations in which they needed explanations within a software system. We identified behavioral patterns within the system, such as click spamming, that occur frequently in connection with the need for explanations. Physical reactions such as frowning or verbal expression were also observed in situations where an explanation was needed. Our vision is that in the future, after some validation steps and further refinement, these behavior patterns can be used in deployed systems as needs indicators. This would enable the automatic collection of explanation needs during runtime using telemetry data or other monitoring devices.
If software is complex and difficult to use, endusers increasingly rely on external, often inaccurate information sources. To avoid this, system designers must elicit and implement appropriate explanation requirements, enabling their software to explain itself without external help. The accurate elicitation of explanat...
Jakob Droste, C. Wellhausen, Hannah Deters et al.· 2026 IEEE 34th International...· 0 citations
There seems to be a consensus that adding explanations to AI systems, such as recommender systems, can have positive effects, such as increasing a user’s trust, the transparency of the system, or the efficiency of a decision-making process. It is to this date unclear, though, how an explanation needs to be designed to...
Generative AI has moved quickly from laboratory demos to everyday mobile apps in China, and user satisfaction has become the clearest test of whether these products actually deliver value. Yet most satisfaction studies still lean on questionnaires — useful, but too structured to capture what people really complain abou...
Yi-Zhe Wang, Chao Han, Guan Zhang et al.· Frontiers in Computing and I...· 0 citations
Evaluated SWIRL, which supports interpreting tool-generated warnings through interactive, customized summarization, suggests that SWIRL's active learning-based summarization can enhance the sensemaking process of tool-generated warnings.
Usability is a key non-functional requirement in interactive software systems, and heuristic evaluation can support the verification and validation of user interfaces by identifying violations of usability principles. However, this process still depends on human expertise, is subject to evaluator variability, and may b...
Ana Beatriz Farias, D. Albuquerque, Emanuel Dantas et al.· Brazilian Symposium on Syste...· 0 citations
This paper investigates whether LLMs can be steered to generate explanations that are tailored to a specific group of people and proposes an approach that first identifies group-specific attributes in terms of explanatory style and knowledge of a specific target group and tailors the explanations significantly better t...
L. Fichtel, Janek Prange, Henning Wachsmuth· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
