Preview

PDF, English - main document Download (22MB) | Terms of use

Abstract

[Context] User feedback plays an important role in software development, improving system acceptance, reducing project failures and enhancing customer loyalty. To achieve these benefits, software organizations actively collect, analyze and validate feedback to derive changes to existing requirements or new requirements. However, feedback collection and requirements derivation are hindered by several problems that are reported in literature. Feedback from a small selection of users can result in biased or incomplete requirements; vague or ambiguous feedback makes it difficult to map it to requirements; feedback lacking details to propose a change to the application cannot be used to derive requirements; gathering feedback at specific times is not possible by relying solely on user initiative and feedback that is used for requirements derivation must be validated to ensure support among users. [Objective] The goal of this thesis is to introduce an approach for requirements engineers that enables them to collect feedback and derive requirements without facing the mentioned problems. The approach consists of a process to collect feedback and derive requirements, as well as the platform “smartFEEDBACK” that supports the process. [Methods] To achieve the goal, this thesis follows the design science methodology consisting of problem investigation, treatment design and treatment validation. The problem investigation consists of a systematic mapping study to understand the current state and practice of collecting feedback over platforms. Platforms are online tools that facilitate the collection of feedback from multiple stakeholders and enable exchange about that feedback among them. The results of the problem investigation are the basis for the treatment design (our approach). The treatment validation validates whether the approach is feasible and effective, as well as whether the users are satisfied with it and whether the approach can be improved. For the treatment validation the approach is applied in the large-scale research project SMART-AGE that examines the use of four interconnected apps developed for older adults to improve their quality of life. [Contributions] We contribute our approach that enables researchers and practitioners to collect feedback and derive requirements without encountering the mentioned problems. Additionally, we contribute our mapping study that can serve as a foundation for future systematic mapping studies or as an orientation for designing individual feedback platforms. We also offer a dataset of change requests collected in SMART-AGE, providing insights into real-world feedback from older adults. This dataset is useful for researchers and practitioners aiming to understand the specific needs and preferences of this user group. Lastly, we contribute a validation of the approach's feasibility, effectiveness, user satisfaction and improvement, offering a benchmark for researchers to compare their approaches.