Gait and posture analysis play a critical role in the early diagnosis of neurological disorders, particularly Parkinson’s disease. Although numerous machine learning (ML) approaches have been developed to support this task, many lack integration with complementary clinical biomarkers and offer limited transparency in their decision-making processes. This lack of interpretability hinders their adoption in clinical practice. To address these limitations, we propose EDAM (Explainable Diagnosis Recommender), a decision support system that leverages gait and posture data alongside clinical biomarkers, incorporating Explainable Artificial Intelligence (XAI) techniques to enhance interpretability. EDAM not only achieves high predictive accuracy but also generates interpretable visual and textual explanations, enabling clinicians to understand the underlying rationale of each diagnosis. We evaluate EDAM across two primary tasks: (i) differentiating healthy individuals from those with Parkinson’s disease, and (ii) classifying abnormal gait patterns commonly associated with early-stage neurological conditions. To ensure robust evaluation, we constructed one of the most comprehensive datasets to date by aggregating and harmonizing multiple publicly available sources. The resulting dataset comprises 557 features and 7,303 labeled instances, encompassing a broad spectrum of gait characteristics and clinical indicators. Experimental results demonstrate that EDAM delivers high performance in both classification tasks, highlighting its potential for early detection of neurological disorders. Beyond quantitative metrics, we conducted a qualitative assessment to evaluate EDAM’s clinical applicability and interpretability. A semi-structured interview with an experienced physiotherapist specializing in neurorehabilitation affirmed EDAM’s transparency, practical utility in gait assessment, and its promise as a tool for supporting early-stage diagnostic decision-making.

Artificial Intelligence (AI) is transforming many domains, including software engineering. AI-based tools are gaining popularity and are increasingly being integrated into software development workflows, automating complex tasks such as code writing and reviewing. When it comes to coding tasks, some evidence suggests that tools like Copilot boost developers’ productivity (e.g., developers can handle a larger number of pull requests per week). However, it remains unclear whether this comes at the expense of code ownership (i.e., the developer’ ability to argue about their implementation choices). To partially address this gap, we present an experiment aimed at investigating the impact of AI-based assistants on developers’ productivity and behavior in the context of code writing (e.g., developing a program from scratch or evolving an existing code). Our focus is on the interplay between productivity and code ownership. We asked 69 participants (34 BSc and 13 MSc students, 8 researchers, and 14 professional developers) to perform two code writing tasks, one with the support of AI and one without. Then, we compared the two treatments in terms of: (i) time spent on the coding task and percentage of the task completeness—both being productivity proxies; and (ii) ability of the participants to answer questions about the code they implemented—code ownership proxy. While the time-based analyses did not provide strong evidence on the impact of AI on the investigated dependent variables, participants using AI achieved a much higher task completeness (>2× in terms of median), confirming a positive impact on their productivity. However, such a boost did not come for free. Indeed, we also observed a loss in code ownership when participants used AI, with lower ability to answer technical questions (–12.5%).

Voice-based virtual assistants enable hands-free operation, allowing users to perform tasks, access information, and control smart home devices through simple voice commands. Their growing ubiquity in smartphones, smart speakers, and other devices led to the flourish of more and more apps taking advantage of a Voice User Interface (VUI). VUI testing is far from trivial due to the wide variability in human speech (e.g., different accents, dialects, speech patterns), and the fact that users can express the same command in numerous ways, using different (but semantically equivalent) wordings and phrases. For this reason, techniques have been proposed to support VUI testing. The basic idea behind these specialized approaches is to generate paraphrases for the set of voice commands for which developers implemented support in the VUI. Preliminary results from a recent study suggest that specialized models can outperform a general-purpose LLM (ChatGPT). However, a simple prompt and interaction strategy with ChatGPT has been adopted. In other words, it is still unknown whether optimizing the LLM usage allows to obtain better results. In this paper, we aim to thoroughly study to what extent LLMs (ChatGPT, specifically) can be adopted to test VUIs. We focused on optimizing the used prompt and the interaction with the model. Our results show that an optimized use of LLMs results in new state-of-the-art performance for VUI testing in terms of number of correct and bug-revealing paraphrases. While introducing the generated paraphrases into the Voice Interaction Models of the skills allows to fix some bugs, we observe that many bugs remain, and some are even introduced by the generated paraphrases. Our results call for specialized approaches for fixing bugs in VUIs.

Mobility impairments reduce the ability of patients to perform daily activities. Physio-therapeutic exercises are essential to address such limitations. Correctly executing exercises is crucial to achieving good results; therefore, a physiotherapist’s presence is essential to guide the execution of rehabilitation movements. In this context, the potential of combining advanced sensors with artificial intelligence techniques can represent a real breakthrough for home rehabilitation, as remote monitoring systems allow patients to avoid moving from home, making rehabilitation sessions less stressful. In this paper, we propose Virtual-Physio, a virtual assistant for remote rehabilitation integrated into a home-deployable low-cost physiotherapy monitoring system 2Vita-B Physical . Virtual-Physio supports patients during exercises by providing real-time feedback and automatically evaluating session performance through machine learning techniques. This allows physiotherapists to focus their attention on cases requiring professional intervention, thereby improving scalability and efficiency in care delivery. Results show that (i) the proposed machine learning approach can effectively perform automated evaluations of rehabilitative movements, and (ii) participants reported high levels of satisfaction and acceptance, as assessed through a dedicated post-session usability questionnaire. These findings indicate the system’s potential to enhance patient engagement and support home rehabilitation.

Code comprehension is vital for software development. Still, unreadable code remains a significant issue, costing substantial time and money losses. While tools to identify code exhibiting a low readability exist, take actions to improve such a quality aspect is far from trivial. To support developers in such a task, Vitale et al. introduced at ASE’23 an approach using a transformer model (T5) fine-tuned on code commits in which developers explicitly stated their goal to improve code readability. The authors reported that their model is able to generate readability-improving changes being identical to those implemented by developers (exact matches) in 21 ∼ 80 ∼ 1 ∼ 9,500 inspected LLMs’ change recommendations. While all models do improve code readability, our results clearly show that fine-tuning using mined readability-improving commits does not help and, instead, results in sensibly poorer readability recommendations as compared to LLMs used in a zero-shot setting, with GPT-4.1 being the one achieving the best results.

This paper presents the Nova Tellus project, a pioneering platform that integrates blockchain technology and artificial intelligence within a unified information system to advance smart agricultural practices. Built on a private Hyperledger Fabric blockchain infrastructure, Nova Tellus enables secure, transparent collection, storage, and sharing of data from IoT devices deployed in agricultural environments. An AI module trained on a comprehensive dataset, including field observations, satellite imagery, and climate data, supports phenological phase classification and adversity detection across multiple crop types, providing data-driven insights for farm decision-making. Results show that the integrated blockchain-AI system effectively combines secure data provenance with robust analytical capabilities, enhancing the credibility and traceability of AI-generated insights. The AI component achieved strong performance, such as an F1-Score of 0.89 for olive phenological classification and 0.70 for grapevine adversity detection, demonstrating its ability to deliver reliable, real-time decision support. Overall, Nova Tellus highlights the substantial potential of combining blockchain and AI within agricultural information systems to enable transparent, intelligent, and sustainable farming practices.

Large Language Models (LLMs) are increasingly relevant in Software Engineering research and practice, with Automated Bug Fixing (ABF) being one of their key applications. ABF involves transforming a buggy method into its fixed equivalent. A common preprocessing step in ABF involves removing comments from code prior to training. However, we hypothesize that comments may play a critical role in fixing certain types of bugs by providing valuable design and implementation insights. In this study, we investigate how the presence or absence of comments, both during training and at inference time, impacts the bug-fixing capabilities of LLMs. We conduct an empirical evaluation comparing two model families, each evaluated under all combinations of training and inference conditions (with and without comments), and thereby revisiting the common practice of removing comments during training. To address the limited availability of comments in state-of-the-art datasets, we use an LLM to automatically generate comments for methods lacking them. Our findings show that comments improve ABF accuracy by up to threefold when present in both phases, while training with comments does not degrade performance when instances lack them. Additionally, an interpretability analysis identifies that comments detailing method implementation are particularly effective in aiding LLMs to fix bugs accurately.

AI-assisted tools support developers in performing cognitively demanding tasks such as bug detection and code readability assessment. Despite the advancements in the technical characteristics of these tools, little is known about how developers mentally model them and how mismatches affect trust, control, and adoption. We conducted six co-design workshops with 58 developers to elicit their mental models about AI-assisted bug detection and readability features. It emerged that developers conceive bug detection tools as bug detectives, which warn users only in cases of critical issues, guaranteeing transparency, actionable feedback, and confidence cues. Readability assessment tools, on the other hand, are envisioned as quality coaches, which provide contextual, personalized, and progressive guidance. Trust, in both tasks, depends on the clarity of explanations, timing, and user control. A set of design principles for Human-Centered AI in IDEs has been distilled, aiming to balance disruption with support, conciseness with depth, and automation with human agency.

Modern video games are extremely complex software systems and, as such, they might suffer from several types of post-release issues. A particularly insidious issue is constituted by drops in the frame rate ( i.e. , stuttering events), which might have a negative impact on the user experience. Stuttering events are frequently documented in the million of hours of gameplay videos shared by players on platforms such as Twitch or YouTube. From the developers’ perspective, these videos represent a free source of documented “testing activities”. However, especially for popular games, the quantity and length of these videos make impractical their manual inspection. We introduce HASTE, an approach for the automatic detection of stuttering events in gameplay videos that can be exploited to generate candidate bug reports. HASTE firstly splits a given video into visually coherent slices, with the goal of filtering-out those that not representing actual gameplay ( e.g ., navigating the game settings). Then, it identifies the subset of pixels in the video frames which actually show the game in action excluding additional elements on screen such as the logo of the YouTube channel, on-screen chats etc. In this way, HASTE can exploit state-of-the-art image similarity metrics to identify candidate stuttering events, namely subsequent frames being almost identical in the pixels depicting the game. We evaluate the different steps behind HASTE on a total of 105 videos showing that it can correctly extract video slices with a 76% precision, and can correctly identify the slices related to gameplay with a recall and precision higher than 77%. Overall, HASTE achieves 71% recall and 89% precision for the identification of stuttering events in gameplay videos.

The gaming industry has witnessed remarkable growth in recent years, attracting millions of people who engage in its products both as a hobby and for professional purposes (e.g., e-sports). Video games are software products that have a unique and fundamental requirement: They must be engaging. Previous research introduced approaches aimed at measuring engagement, some of which specifically designed for video games. Such approaches could be useful for practitioners since they can be adopted on the large collection of gameplay videos daily published on platforms such as Twitch and YouTube to allow developers to monitor players' engagement and detect areas in which it is low. Such specialized approaches have been evaluated on datasets in which the engagement was manually assessed by external evaluators based on the face of the player (we call it perceived engagement). We still do not know whether such approaches can capture the real engagement of players. Also, it is unclear to what extent practitioners would be willing to adopt such approaches in practice. In this paper, we provide two contributions. First, we ran an experiment with human 40 players aimed at defining a dataset of gameplay sessions in which participants self-reported their real engagement after every minute. We captured both their face and the gameplay. Based on this data, we compared state-of-the-art machine learning-based approaches to detect lowly engaging sessions. Our results show that the best model correctly classifies engagement in 74.7% of the cases and ranks video games in terms of their real engagement very similarly to how players would rank them (Spearman. = 0.833). Second, to assess the practicality of adopting such approaches in an industrial setting, we conducted two semi-structured interviews with senior game developers, who provided generally positive feedback and interesting insights for future developments.

Background and Objective: Mobility impairments reduce the ability of patients to complete daily activities. Physio-therapeutic exercises help patients address such limitations or even completely recover. Correctly executing exercises is crucial to achieving good results; therefore, a physiotherapist's presence is essential to guide the execution of rehabilitation movements. In this context, the potential of combining advanced sensors with artificial intelligence techniques can represent a real breakthrough for home rehabilitation, as remote monitoring systems allow patients to avoid moving from home, making rehabilitation sessions less burdensome.Methods: In this paper, we introduce Virtual-Physio, a virtual assistant for remote rehabilitation integrated into a home-deployable low-cost physiotherapy monitoring system called 2Vita-B Physical. Virtual-Physio automatically (i) provides real-time online feedback to the patients while they perform the rehabilitation exercises, and (ii) evaluates a whole exercise session, so that a physiotherapist can focus only on cases that deserve more attention.Results: We experimented with Virtual-Physio on 51 individuals whose performances were also evaluated by an equipe of physiotherapists as a reference. The results (i) highlight good patient acceptability for the virtual assistant, and (ii) show that the proposed machine learning approach can effectively perform an automated evaluation of rehabilitative movements.Conclusions: We believe that the findings presented in this study provide a valid foundation for future improvements to home rehabilitation supported by the integration of motion capture and Machine Learning.

Unreadable code could be a breeding ground for errors. Thus, previous work defined approaches based on machine learning to automatically assess code readability that can warn developers when some code artifacts (e.g., classes) become unreadable. Given datasets of code snippets manually evaluated by several developers in terms of their perceived readability, such approaches (i) establish a snippet-level ground truth, and (ii) train a binary (readable/unreadable) or a ternary (readable/neutral/unreadable) code readability classifier. Given this procedure, all existing approaches neglect the subjectiveness of code readability, i.e., the possible different developer-specific nuances in the code readability perception. In this paper, we aim to understand to what extent it is possible to assess code readability as subjectively perceived by developers through a personalized code readability assessment approach. This problem is significantly more challenging than the snippet-level classification problem: We assume that, in a realistic scenario, a given developer is keen to provide only a few code readability evaluations, thus less data is available. For this reason, we adopt an LLM with few-shot learning to achieve our goal. Our results, however, show that such an approach achieves worse results than a state-of-the-art feature-based model that is trained to work at the snippet-level. We tried to understand why this happens by looking more closely at the quality of the available code readability datasets and assessed the consistency of the inter-developer evaluations. We observed that up to a third of the evaluations are self-contradictory. Our negative results call for new and more reliable code readability datasets.

Context: Performance regressions negatively impact execution time and memory usage of software systems. Nevertheless, there is a lack of systematic methods to evaluate the effectiveness of performance test suites. Performance mutation testing, which introduces intentional defects (mutants) to measure and enhance fault-detection capabilities, is promising but underexplored. A key challenge is understanding if generated mutants accurately reflect real-world performance issues. Goal: This study evaluates and extends mutation operators for performance testing. Its objectives include (i) collecting existing performance mutation operators, (ii) introducing new operators from real-world code changes that impact performance, and (iii) evaluating these operators on real-world systems to see if they effectively degrade performance. Method: To this aim, we will (i) review the literature to identify performance mutation operators, (ii) conduct a mining study to extract patterns of code changes linked to performance regressions, (iii) propose new mutation operators based on these patterns, and (iv) apply and evaluate the operators to assess their effectiveness in exposing performance degradations. Expected Outcomes: We aim to provide an enriched set of mutation operators for performance testing, helping developers and researchers identify harmful coding practices and design better strategies to detect and prevent performance regressions.

Software containerization, for which Docker is the reference tool, is widely adopted in modern software engineering. The performance of the Docker build process in terms of image size and build time is crucial to developers. While previous work and Docker itself provide best practices to keep the images small and fast to build, we conjecture that developers might adopt undocumented practices. In this paper, we present an empirical study in which we aim (i) to mine the practices adopted by developers for improving the image size and build time, and (ii) to measure the impact of such practices. As for the mining study, we manually analyzed a total of 1,026 commits from open-source projects in which developers declared they wanted to improve the image size or build time. We categorize such changes and define a taxonomy of 46 optimization strategies, including practices such as removing temporary files (e.g., package manager cache) or improving the structure of the Dockerfile (e.g., using multi-stage build). Such a taxonomy reveals some previously undocumented techniques, providing valuable insights for developers. As for the measurement study, we empirically assess the actual improvement in image size and build time (over 20 builds) of the most frequent change patterns observed in the mining study. Our results show that changing the base image has the best results in terms of image size, but it negatively affects the build time. On the other hand, we observed no change pattern that significantly reduces the build time. Our study provides interesting insights for both tool makers who want to support practitioners in improving Dockerfile build performance and practitioners themselves, who can better decide how to optimize their Dockerfiles.

Context. AI-based development tools, such as GitHub Copilot, are transforming the software development process by offering real-time code suggestions. These tools promise to improve the productivity by reducing cognitive load and speeding up task completion. Previous exploratory studies, however, show that developers sometimes perceive the automatic suggestions as intrusive. As a result, they feel like their productivity decreased. Theory. We propose two theories on the impact of automatic suggestions on frustration and productivity. First, we hypothesize that experienced developers are frustrated from automatic suggestions (mostly from irrelevant ones), and this also negatively impacts their productivity. Second, we conjecture that novice developers benefit from automatic suggestions, which reduce the frustration caused from being stuck on a technical problem and thus increase their productivity. Objective. We plan to conduct a quasi-experimental study to test our theories. The empirical evidence we will collect will allow us to either corroborate or reject our theories. Method. We will involve at least 32 developers, both experts and novices. We will ask each of them to complete two software development tasks, one with automatic suggestions enabled and one with them disabled, allowing for within-subject comparisons. We will measure independent and dependent variables by monitoring developers' actions through an IDE plugin and screen recording. Besides, we will collect physiological data through a wearable device. We will use statistical hypothesis tests to study the effects of the treatments (i.e., automatic suggestions enabled/disabled) on the outcomes (frustration and productivity).

Ensuring high software quality in development cycles with frequent updates is critical, especially in Agile and CI/CD environments. Just-In-Time Software Defect Prediction (JIT-SDP) has emerged as a promising solution for finding bugs early, as it enables immediate identification of changes prone to defects. JIT-SPD models based on Machine Learning focus primarily on project- and change-specific features, such as number of lines added and number of files modified in the change. Recent research has started to investigate developer-related features for defect prediction. However, these studies overlook information about developers’ work habits and cross-project activities. In this paper, we try to fill this gap by introducing a set of developer-centric features for JIT-SDP, which span through temporal aspects (when do developers usually make commits?), change-related aspects (how do developers usually make commits?), and project-related aspects (how are the contributions distributed among different repositories?). We conducted an empirical evaluation to understand if such features allow to improve ML-based JIT-SPD models and evaluated the importance of developer-centric features on the performance of the model. Our results show that integrating developer-centric features improves model performance. We observed a +15.48% precision and +10.47% recall in a within-project evaluation and +14.59% precision and even +85.83% recall in cross-project evaluation.

The growing interest in treating emotional disorders has led to the exploration of virtual reality (VR) as a tool for cognitive rehabilitation. This paper present an extensive evaluation of the 2ViTA-B Cognitive system, an innovative software and hardware system designed to enhance psychological and emotional engagement through customizable stimuli sequences in both standard and VR modes. A controlled study with 16 participants was conducted to assess the effectiveness of the system in enhancing emotional well-being and cognitive function, as well as its impact on self-assessed affectivity using well-established questionnaires. Additionally, the study evaluated the system usability, a critical factor for its overall effectiveness. Results indicate that 2ViTA-B Cognitive significantly enhances mood and cognitive performance, with VR providing a particularly immersive and impactful experience. These findings suggest promising directions for the use of VR in therapeutic practices aimed at improving mental well-being. The paper also discusses the potential benefits and limitations of VR as a therapeutic tool and highlights areas for future research.

The video game industry has experienced a continuous growth in the last decades. In such a competitive market, it is fundamental to ensure a great gaming experience to the player avoiding, for example, bugs. However, video game testing is an extremely challenging activity, especially considering the extensive number of gaming scenarios that modern video games support (e.g., 3D worlds to explore). Thus, more often than not, numerous bugs are discovered only once the game is released and played by millions of users. For this reason, recent work in the literature suggested to exploit gameplay videos to support developers in identifying possible bugs missed during testing: given the large amount of gameplays posted every day on streaming platforms (> 2M hours), these gameplays are likely to document failures experienced by the player. Empirical evidence show the ability of these techniques to identify parts of the gameplay in which the failure was experienced. However, it could still be difficult for game developers to reproduce the bug. In this paper, we propose the idea of developing a technique able to automate this process, providing the game developer with all actions performed by the player to reach the faulty state shown in the gameplay. We present a simple approach which leverages the on-screen controls overlay available in some gameplay videos. We show that such an approach can replicate 47.2% of gameplays in our preliminary study run on a racing game. We discuss the strong limitations of this first attempt, listing directions for future work we plan to pursue in order to overcome them.