Paolo Giorgini
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Paolo Giorgini has a total of 87 co-authors. Here are the top co-authors by number of shared publications:
- Elda Paja
IT University of Copenhagen, Copenhagen, Denmark
7 shared publications - Fabiano Dalpiaz
Utrecht University
6 shared publications - John Mylopoulos
University of Ottawa, Ottawa, Canada
6 shared publications - Marco Robol
University of Trento
6 shared publications - Haralambos Mouratidis
Institute for Analytics and Data Science, University of Essex, Colchester, United Kingdom
5 shared publications - Marco Roveri
University of Trento
4 shared publications
Latest Publications
- 17 October 2026
- Published by Springer Nature
- p. 377-388
Engineering Autonomous LLM-Based Desktop Agents: A Modular Approach
- 17 October 2026
- Published by Springer Nature
- p. 377-388
Engineering Autonomous LLM-Based Desktop Agents: A Modular Approach
- Open Access
- 12 October 2026
- Published by Springer Nature in Requirements Engineering
- Vol. 31 (2) , 20
A systematic mapping of goal-oriented requirements engineering: reflections, impact, and the way forward
Goal-oriented Requirements Engineering describes the body of work which focuses on goals as the driving force motivating and capturing requirements, often using graphical goal models. A proliferation of work in this area inspired the study: “Goal-oriented requirements engineering: an extended systematic mapping study” first available in 2017, covering 246 publications in GORE up to 2015. In the current article, we have the pleasure of reflecting on this original mapping study. Using bibliometrics, we examine the study’s visibility, although such a quantitative view gives us only an approximation of real impact. The bibliometric view shows us that most citations are relatively recent: 85% of the citations have occurred since 2020. Unsurprisingly, the work has primarily been cited in requirements engineering venues, but also related venues such as the Conceptual Modeling conference (ER) and the Information and Software Technology Journal (IST). The work has been cited by a wide variety of authors. We estimate that roughly 40% of the publications citing the original SLR actually focus on goal models (have some variant of “goal” in the title), but that only roughly 53% of citing papers which are not themselves systematic reviews emphasize some form of empirical evaluation, 10% lower than the presence of empirical studies found in the original mapping study. Turning to trends and future directions, we reflect on the limitations of GORE uptake in industry, along with recommendations for continued use in education. We highlight new developments in using Large Language Models (LLMs) to generate goal models, including limitations, and reflect on what opportunities may be lost via full automation of goal model generation. We reflect on the potential role of GORE in the rising field of intent-driven engineering with AI, and the disconnect between much of this work and requirements-focused goal-oriented foundations. Finally, we consider the rising popularity of agentic AI, describing conceptual challenges and opportunities for GORE in terms of monitoring, capturing, and driving agentic system design. Overall, our reflection emphasizes that in the rise of LLMs and Agentic AI, the expressive power of GORE has great potential, and we invite the research community to explore and harness goal modeling for this purpose.
- 12 October 2026
- Published by Springer Nature in Requirements Engineering
- Vol. 31 (2) , 20
Open AccessA systematic mapping of goal-oriented requirements engineering: reflections, impact, and the way forward
Goal-oriented Requirements Engineering describes the body of work which focuses on goals as the driving force motivating and capturing requirements, often using graphical goal models. A proliferation of work in this area inspired the study: “Goal-oriented requirements engineering: an extended systematic mapping study” first available in 2017, covering 246 publications in GORE up to 2015. In the current article, we have the pleasure of reflecting on this original mapping study. Using bibliometrics, we examine the study’s visibility, although such a quantitative view gives us only an approximation of real impact. The bibliometric view shows us that most citations are relatively recent: 85% of the citations have occurred since 2020. Unsurprisingly, the work has primarily been cited in requirements engineering venues, but also related venues such as the Conceptual Modeling conference (ER) and the Information and Software Technology Journal (IST). The work has been cited by a wide variety of authors. We estimate that roughly 40% of the publications citing the original SLR actually focus on goal models (have some variant of “goal” in the title), but that only roughly 53% of citing papers which are not themselves systematic reviews emphasize some form of empirical evaluation, 10% lower than the presence of empirical studies found in the original mapping study. Turning to trends and future directions, we reflect on the limitations of GORE uptake in industry, along with recommendations for continued use in education. We highlight new developments in using Large Language Models (LLMs) to generate goal models, including limitations, and reflect on what opportunities may be lost via full automation of goal model generation. We reflect on the potential role of GORE in the rising field of intent-driven engineering with AI, and the disconnect between much of this work and requirements-focused goal-oriented foundations. Finally, we consider the rising popularity of agentic AI, describing conceptual challenges and opportunities for GORE in terms of monitoring, capturing, and driving agentic system design. Overall, our reflection emphasizes that in the rise of LLMs and Agentic AI, the expressive power of GORE has great potential, and we invite the research community to explore and harness goal modeling for this purpose.
- 5 October 2026
- Published by Springer Nature
- p. 417-427
Tropos$^{+}$: Tropos Extended in the Days of Large Language Models- 5 October 2026
- Published by Springer Nature
- p. 417-427
Tropos$^{+}$: Tropos Extended in the Days of Large Language Models- 29 September 2026
- Published by Springer Nature
- p. 75-90
From Text to Knowledge: An Ontology-Aware Triple Extraction Framework with LLMs
- 29 September 2026
- Published by Springer Nature
- p. 75-90
From Text to Knowledge: An Ontology-Aware Triple Extraction Framework with LLMs
- Open Access
- 30 April 2026
- Published by Elsevier in Information and Software Technology
- Vol. 196
Consent under control with ProPrivacy: Business process compliance verification for GDPR-consent requirements
- 30 April 2026
- Published by Elsevier in Information and Software Technology
- Vol. 196
Open AccessConsent under control with ProPrivacy: Business process compliance verification for GDPR-consent requirements
- Open Access
- 28 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Translational Engineering in Health and Medicine
- Vol. 13, 61-74
Multi-Branch CNN-LSTM Fusion Network-Driven System With BERT Semantic Evaluator for Radiology Reporting in Emergency Head CTs
The high volume of emergency room patients often necessitates head CT examinations to rule out ischemic, hemorrhagic, or other organic pathologies. A system that enhances the diagnostic efficacy of head CT imaging in emergency settings through structured reporting would significantly improve clinical decision making. Currently, no AI solutions address this need. Thus, our research aims to develop an automatic radiology reporting system by directly analyzing brain anomalies in head CT data. We propose a multi-branch CNN-LSTM fusion network-driven system for enhanced radiology reporting in emergency settings. We preprocessed head CT scans by resizing all slices, selecting those with significant variability, and applying PCA to retain 95% of the original data variance, ultimately saving the most representative five slices for each scan. We linked the reports to their respective slice IDs, divided them into individual captions, and preprocessed each. We performed an 80-20 split of the dataset for ten times, with 15% of the training set used for validation. Our model utilizes a pretrained VGG16, processing groups of five slices simultaneously, and features multiple end-to-end LSTM branches, each specialized in predicting one caption, subsequently combined to form the ordered reports after a BERT-based semantic evaluation. Our system demonstrates effectiveness and stability, with the postprocessing stage refining the syntax of the generated descriptions. However, there remains an opportunity to empower the evaluation framework to more accurately assess the clinical relevance of the automatically-written reports. Part of future work will include transitioning to 3D and developing an improved version based on vision-language models.
- 28 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Translational Engineering in Health and Medicine
- Vol. 13, 61-74
Open AccessMulti-Branch CNN-LSTM Fusion Network-Driven System With BERT Semantic Evaluator for Radiology Reporting in Emergency Head CTs
The high volume of emergency room patients often necessitates head CT examinations to rule out ischemic, hemorrhagic, or other organic pathologies. A system that enhances the diagnostic efficacy of head CT imaging in emergency settings through structured reporting would significantly improve clinical decision making. Currently, no AI solutions address this need. Thus, our research aims to develop an automatic radiology reporting system by directly analyzing brain anomalies in head CT data. We propose a multi-branch CNN-LSTM fusion network-driven system for enhanced radiology reporting in emergency settings. We preprocessed head CT scans by resizing all slices, selecting those with significant variability, and applying PCA to retain 95% of the original data variance, ultimately saving the most representative five slices for each scan. We linked the reports to their respective slice IDs, divided them into individual captions, and preprocessed each. We performed an 80-20 split of the dataset for ten times, with 15% of the training set used for validation. Our model utilizes a pretrained VGG16, processing groups of five slices simultaneously, and features multiple end-to-end LSTM branches, each specialized in predicting one caption, subsequently combined to form the ordered reports after a BERT-based semantic evaluation. Our system demonstrates effectiveness and stability, with the postprocessing stage refining the syntax of the generated descriptions. However, there remains an opportunity to empower the evaluation framework to more accurately assess the clinical relevance of the automatically-written reports. Part of future work will include transitioning to 3D and developing an improved version based on vision-language models.
- 20 November 2024
- Published by Springer Nature
- p. 190-208
LLM-Driven Knowledge Extraction in Temporal and Description Logics
- 20 November 2024
- Published by Springer Nature
- p. 190-208
LLM-Driven Knowledge Extraction in Temporal and Description Logics
- Open Access
- 14 November 2024
- Published by Springer Nature in Annals of Mathematics and Artificial Intelligence
- Vol. 93 (4) , 569-588
A multi-algorithm pathfinding method: Exploiting performance variations for enhanced efficiency
- 14 November 2024
- Published by Springer Nature in Annals of Mathematics and Artificial Intelligence
- Vol. 93 (4) , 569-588
Open AccessA multi-algorithm pathfinding method: Exploiting performance variations for enhanced efficiency
- Open Access
- 24 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 598-602
MOOD: Mindfulness fOr sOftware Developers
Peopleware, which includes anything related to the role of people in Software Development (SD), has been arousing an increasing interest from both the software industry and research community. This interest is due to the current economic system that demands high-quality software products with a short time to market, staying on the budget. This exposes software developers to the risk of experiencing stress, burnout, and reduced motivation, leading, in turn, to reduced job performance, low-quality SD-related artifacts, and increased turnover. Mindfulness represents a promising intervention that might let developers do their best at work, limiting or even preventing the previously mentioned negative outcomes. This paper presents MOOD (Mindfulness fOr sOftware Developers), a research project whose overarching goal is to customize a well-known and validated group-based intervention program, Mindfulness-Based Stress Reduction (MBSR), in the context of SD-related tasks and assess whether it helps developers to improve their well-being and performance, as well as the quality of the SD-related artifacts they produce.
- 24 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 598-602
Open AccessMOOD: Mindfulness fOr sOftware Developers
Peopleware, which includes anything related to the role of people in Software Development (SD), has been arousing an increasing interest from both the software industry and research community. This interest is due to the current economic system that demands high-quality software products with a short time to market, staying on the budget. This exposes software developers to the risk of experiencing stress, burnout, and reduced motivation, leading, in turn, to reduced job performance, low-quality SD-related artifacts, and increased turnover. Mindfulness represents a promising intervention that might let developers do their best at work, limiting or even preventing the previously mentioned negative outcomes. This paper presents MOOD (Mindfulness fOr sOftware Developers), a research project whose overarching goal is to customize a well-known and validated group-based intervention program, Mindfulness-Based Stress Reduction (MBSR), in the context of SD-related tasks and assess whether it helps developers to improve their well-being and performance, as well as the quality of the SD-related artifacts they produce.
- Open Access
Implementing BDI Continual Temporal Planning for Robotic Agents
Making autonomous agents effective in real-life applications requires the ability to decide at run-time and a high degree of adaptability to unpredictable and uncontrollable events. Reacting to events is still a fundamental ability for an agent, but it has to be boosted up with proactive behaviors that allow the agent to explore alternatives and decide at run-time for optimal solutions. This calls for a continuous planning as part of the deliberation process that makes an agent able to reconsider plans on the base of temporal constraints and changes of the environment. Online planning literature offers several approaches used to select the next action on the base of a partial exploration of the solution space. In this paper, we propose a BDI continuous temporal planning framework, where interleave planning and execution loop is used to integrate online planning with the BDI control-loop. The framework has been implemented with the ROS2 robotic framework and planning algorithms offered by JavaFF.
Open AccessImplementing BDI Continual Temporal Planning for Robotic Agents
Making autonomous agents effective in real-life applications requires the ability to decide at run-time and a high degree of adaptability to unpredictable and uncontrollable events. Reacting to events is still a fundamental ability for an agent, but it has to be boosted up with proactive behaviors that allow the agent to explore alternatives and decide at run-time for optimal solutions. This calls for a continuous planning as part of the deliberation process that makes an agent able to reconsider plans on the base of temporal constraints and changes of the environment. Online planning literature offers several approaches used to select the next action on the base of a partial exploration of the solution space. In this paper, we propose a BDI continuous temporal planning framework, where interleave planning and execution loop is used to integrate online planning with the BDI control-loop. The framework has been implemented with the ROS2 robotic framework and planning algorithms offered by JavaFF.