Context: The quality of requirements specifications is a critical success factor in software development. Assuring high-quality requirements, specifically in an automated way, poses a significant challenge due to their unstructured and multi-modal character. With the rise of deep learning and large language models (LLMs), new opportunities have developed to assess the quality of requirements automatically, particularly user stories in the context of agile software engineering, where short development cycles require efficient tool support. Objective: This study aims to systematically review and investigate the current landscape of approaches based on artificial intelligence techniques such as natural language processing and deep learning for assessing the quality of software requirements. The investigation focuses on the artificial intelligence techniques adopted, quality aspects considered, datasets used to tune and evaluate the proposed approaches, and their performance. Method: We conducted a systematic literature review of 26 peer-reviewed papers published between 2019 and 2025. We selected the papers after a title and abstract review of 353 papers identified through a literature databases query and forward–backward snowballing. Results: The results reveal significant overlap among considered quality aspects, which can be mapped onto the higher-order requirements quality model INVEST. Most studies focus on assessing requirement quality rather than improving requirements and rely heavily on synthetic and public datasets. LLMs have rapidly gained popularity since 2023, though model evaluation strategies remain inconsistent. Metrics such as accuracy, precision, recall, and F1-Score are common, yet a few studies use semantic or expert-based evaluations. Conclusion: The field is evolving toward LLM-driven, semantically rich models, yet lacks methodological standardization, reproducible datasets for evaluating the models, and integration of the approaches with real-world requirements engineering processes. Future work should address these limitations by developing benchmark datasets, standardizing evaluation metrics, and exploring hybrid systems that combine AI-based and traditional requirements quality assurance approaches.

Abstract Nowadays, systems containing components based on machine learning (ML) methods are becoming more widespread. In order to ensure the intended behavior of a software system, there are standards that define necessary qualities of the system and its components (such as ISO/IEC 25010). Due to the different nature of ML, we have to re-interpret existing qualities for ML systems or add new ones (such as trustworthiness). We have to be very precise about which quality property is relevant for which entity of interest (such as completeness of training data or correctness of trained model), and how to objectively evaluate adherence to quality requirements. In this article, we present how to systematically construct quality models for ML systems based on an industrial use case. This quality model enables practitioners to specify and assess qualities for ML systems objectively. In addition to the overall construction process described, the main outcomes include a meta-model for specifying quality models for ML systems, reference elements regarding relevant views, entities, quality properties, and measures for ML systems based on existing research, an example instantiation of a quality model for a concrete industrial use case, and lessons learned from applying the construction process. We found that it is crucial to follow a systematic process in order to come up with measurable quality properties that can be evaluated in practice. In the future, we want to learn how the term quality differs between different types of ML systems and come up with reference quality models for evaluating qualities of ML systems.

AI-based systems are software systems with functionalities enabled by at least one AI component (e.g., for image- and speech-recognition, and autonomous driving). AI-based systems are becoming pervasive in society due to advances in AI. However, there is limited synthesized knowledge on Software Engineering (SE) approaches for building, operating, and maintaining AI-based systems. To collect and analyze state-of-the-art knowledge about SE for AI-based systems, we conducted a systematic mapping study. We considered 248 studies published between January 2010 and March 2020. SE for AI-based systems is an emerging research area, where more than 2/3 of the studies have been published since 2018. The most studied properties of AI-based systems are dependability and safety. We identified multiple SE approaches for AI-based systems, which we classified according to the SWEBOK areas. Studies related to software testing and software quality are very prevalent, while areas like software maintenance seem neglected. Data-related issues are the most recurrent challenges. Our results are valuable for: researchers, to quickly understand the state of the art and learn which topics need more research; practitioners, to learn about the approaches and challenges that SE entails for AI-based systems; and, educators, to bridge the gap among SE and AI in their curricula.

As the development progresses, software projects tend to accumulate Technical Debt and become harder to maintain. Multiple tools exist with the mission to help practitioners to better manage Technical Debt. Despite this progress, there is a lack of tools providing actionable and self-learned suggestions to practitioners aimed at mitigating the impact of Technical Debt in real projects. We aim to create a data-driven, lightweight, and self-learning tool positioning highly impactful refactoring proposals on a Jira backlog. Bearing this goal in mind, the first two authors have founded a startup, called Skuld.ai, with the vision of becoming the go-to software renovation company. In this tool paper, we present the software architecture and demonstrate the main functionalities of our tool. It has been showcased to practitioners, receiving positive feedback. Currently, its release to the market is underway thanks to an industry-research institute collaboration with Fraunhofer IESE to incorporate self-learning technical debt capabilities.

Existing definitions and metrics of technical debt (TD) tend to focus on static properties of software artifacts, in particular on code measurement. Our experience from software renovation projects is that dynamic aspects - runtime indicators of TD - often play a major role. In this position paper, we present insights and solution ideas gained from numerous software renovation projects at QAware and from a series of interviews held as part of the ProDebt research project. We interviewed ten practitioners from two German software companies in order to understand current requirements and potential solutions to current problems regarding TD. Based on the interview results, we motivate the need for measuring dynamic indicators of TD from the practitioners' perspective, including current practical challenges. We found that the main challenges include a lack of production-ready measurement tools for runtime indicators, the definition of proper metrics and their thresholds, as well as the interpretation of these metrics in order to understand the actual debts and derive countermeasures. Measuring and interpreting dynamic indicators of TD is especially difficult to implement for companies because the related metrics are highly dependent on runtime context and thus difficult to generalize. We also sketch initial solution ideas by presenting examples of dynamic indicators for TD and outline directions for future work.

Published software quality models either provide abstract quality attributes or concrete quality assessments. There are no models that seamlessly integrate both aspects. In the project Quamoco, we built a comprehensive approach with the aim to close this gap. For this, we developed in several iterations a meta quality model specifying general concepts, a quality base model covering the most important quality factors and a quality assessment approach. The meta model introduces the new concept of a product factor, which bridges the gap between concrete measurements and abstract quality aspects. Product factors have measures and instruments to operationalise quality by measurements from manual inspection and tool analysis. The base model uses the ISO 25010 quality attributes, which we refine by 200 factors and 600 measures for Java and C# systems. We found in several empirical validations that the assessment results fit to the expectations of experts for the corresponding systems. The empirical analyses also showed that several of the correlations are statistically significant and that the maintainability part of the base model has the highest correlation, which fits to the fact that this part is the most comprehensive. Although we still see room for extending and improving the base model, it shows a high correspondence with expert opinions and hence is able to form the basis for repeatable and understandable quality assessments in practice.

Reliable predictions are essential for managing software projects with respect to cost and quality. Several studies have shown that hybrid prediction models combining causal models with Monte Carlo simulation are especially successful in addressing the needs and constraints of today's software industry: They deal with limited measurement data and, additionally, make use of expert knowledge. Moreover, instead of providing merely point estimates, they support the handling of estimation uncertainty, e.g., estimating the probability of falling below or exceeding a specific threshold. Although existing methods do well in terms of handling uncertainty of information, we can show that they leave uncertainty coming from imperfect modeling largely unaddressed. One of the consequences is that they probably provide over-confident uncertainty estimates. This paper presents a possible solution by integrating bootstrapping into the existing methods. In order to evaluate whether this solution does not only theoretically improve the estimates but also has a practical impact on the quality of the results, we evaluated the solution in an empirical study using data from more than sixty projects and six estimation models from different domains and application areas. The results indicate that the uncertainty estimates of currently used models are not realistic and can be significantly improved by the proposed solution.

Companies increasingly recognize that software and IT play a significant role for their current and future business strategies. Therefore, it is important to align IT/software-related strategies with the business goals across the organization. Currently, little experience exists regarding how to effectively create this missing business-IT link. For this purpose, the GQM+Strategies®approach was developed to support companies in aligning IT/software-related strategies with business goals through measurement. This paper focuses on facilitating the approach for aligning IT/software projects with an organization's higher-level goals. Lessons learned from applying the approach in the context of the Japanese Information-technology Promotion Agency (IPA), specifically its Software Engineering Center (SEC), are presented. The transparent documentation of goals and strategies, and the collection of key performance indicators were helpful for effectively aligning the projects with overall organizational goals and strategies as well as for evaluating the degree of alignment and the risk of misalignment.

GQM+Strategies is a measurement approach that builds on the well-tested GQM approach to planning and implementing software measurement. Although GQM has proven itself useful in a variety of industrial settings, one recognized weakness is the difficulty for GQM users to link software measurement goals to higher-level goals of the organization in which the software is being developed. This linkage is important, as it helps to justify software measurement efforts and allows measurement data to contribute to higher-level decisions. GQM+strategies provides mechanisms for explicitly linking software measurement goals, to higher-level goals for the software organization, and further to goals and strategies at the level of the entire business.

Cost estimation is a very crucial field for software developing companies. The acceptance of an estimation technique is highly dependent on estimation accuracy. Often, this accuracy is only determined after an initial application. Possible further steps for improving the underlying estimation model typically do not influence the decision on whether to discard the technique or deploy it. In addition, most estimation techniques do not explicitly support the evolution of the underlying estimation model in an iterative manner. This increases the risk of overlooking some important cost drivers or data inconsistencies. This paper presents an enhanced process for developing a CoBRA® cost estimation model by systematically including iterative analysis and feedback cycles, and its evaluation in a software development unit of Oki Electric Industry Co., Ltd., Japan. During the model improvement cycles, estimation accuracy was improved from an initial 120% down to 14%. In addition, lessons learned with the iterative development approach are described.

Cost estimation is a vital task in most important software project decisions such as resource allocation and bidding. Analogy-based cost estimation is particularly transparent, as it relies on historical information from similar past projects, whereby similarities are determined by comparing the projects' key attributes and features. However, one crucial aspect of the analogy-based method is not yet fully accounted for: the different impact or weighting of a project's various features. Current approaches either try to find the dominant features or require experts to weight the features. Neither of these yields optimal estimation performance. Therefore, we propose to allocate separate weights to each project feature and to find the optimal weights by extensive search. We test this approach on several real-world data sets and measure the improvements with commonly used quality metrics. We find that this method 1) increases estimation accuracy and reliability, 2) reduces the model's volatility and, thus, is likely to increase its acceptance in practice, and 3) indicates upper limits for analogy-based estimation quality as measured by standard metrics.