Giordano d'Aloisio

University of L'Aquila
DISIM Department
via Vetoio
67100 L'Aquila, Italy
I’m a Postdoctoral Researcher at the University of L’Aquila in Italy where I obtained the European PhD in Information and Communication Technology in 2025 under the supervision of Prof. Antinisca Di Marco.
From Febraury to August 2024 I was a visiting researcher at the University College London (UCL) as a member of the SOLAR research group working with Prof. Federica Sarro.
From 2019 to 2024, I was a member of the Territori Aperti project where I was responsible of the Data Integration activity.
My research is mostly focused on quality aspects of learning-based systems, with a particular attention on software bias and fairness.
Research Interests
- Software Fairness
- Software Sustainability
- Empirical Software Engineering
- Search-Based Software Engineering
- Human Aspects in Computer Science
Selected publications
- ISTTowards early detection of algorithmic bias from dataset’s bias symptoms: An empirical studyGiordano d’Aloisio, Claudio Di Sipio, Antinisca Di Marco, and Davide Di RuscioInformation and Software Technology, 2025
Context The adoption of AI software has made fairness auditing critical, especially in areas where biased decisions can have serious consequences. This process involves identifying sensitive variables and calculating fairness metrics, which depend on accurately identifying these variables and predictions from a baseline model. Model training is the most computationally demanding phase in AI development. Consequently, recent research has shifted to early bias assessment to detect bias signals before intensive model training begins. Objective This paper presents an empirical study to evaluate how dataset characteristics, ie, bias symptoms, can assist in the early identification of variables that may lead to bias in the system. The aim of this study is to avoid training a machine learning model before assessing-and, in case, mitigating-its bias, thus increasing the sustainability of the development process. Method We first identify a bias symptoms dataset, employing 24 datasets from diverse application domains commonly used in fairness auditing. Through extensive empirical analysis, we investigate the ability of these bias symptoms to predict variables associated with bias under three fairness definitions. Results Our results demonstrate that bias symptoms are effective in supporting early predictions of bias-inducing variables under specific fairness definitions. Additionally, we investigate the correlation between those symptoms and specific fairness metrics. Conclusion These findings offer valuable insights for practitioners and researchers, encouraging further exploration in developing methods for proactive bias mitigation involving bias symptoms.
@article{d2025towards, title = {Towards early detection of algorithmic bias from dataset’s bias symptoms: An empirical study}, author = {d’Aloisio, Giordano and Di Sipio, Claudio and Di Marco, Antinisca and Di Ruscio, Davide}, journal = {Information and Software Technology}, pages = {107905}, year = {2025}, publisher = {Elsevier}, doi = {https://doi.org/10.1016/j.infsof.2025.107905} }
- SoSymHow fair are we? From conceptualization to automated assessment of fairness definitionsGiordano d’Aloisio, Claudio Di Sipio, Antinisca Di Marco, and Davide Di RuscioSoftware and Systems Modeling, 2025
Fairness is a critical concept in ethics and social domains, but it is also a challenging property to engineer in software systems. With the increasing use of machine learning in software systems, researchers have been developing techniques to automatically assess the fairness of software systems. Nonetheless, a significant proportion of these techniques rely upon pre-established fairness definitions, metrics, and criteria, which may fail to encompass the wide-ranging needs and preferences of users and stakeholders. To overcome this limitation, we propose a novel approach, called MODNESS, that enables users to customize and define their fairness concepts using a dedicated modeling environment. Our approach guides the user through the definition of new fairness concepts also in emerging domains, and the specification and composition of metrics for its evaluation. Ultimately, MODNESS generates the source code to implement fair assessment based on these custom definitions. In addition, we elucidate the process we followed to collect and analyze relevant literature on fairness assessment in software engineering (SE). We compare MODNESS with the selected approaches and evaluate how they support the distinguishing features identified by our study. Our findings reveal that i) most of the current approaches do not support user-defined fairness concepts; ii) our approach can cover two additional application domains not addressed by currently available tools, i.e., mitigating bias in recommender systems for software engineering and Arduino software component recommendations; iii) MODNESS demonstrates the capability to overcome the limitations of the only two other Model-Driven Engineering-based approaches for fairness assessment.
@article{d2024fair, title = {How fair are we? From conceptualization to automated assessment of fairness definitions}, author = {d'Aloisio, Giordano and Di Sipio, Claudio and Di Marco, Antinisca and Di Ruscio, Davide}, journal = {Software and Systems Modeling}, pages = {1--27}, year = {2025}, publisher = {Springer Berlin Heidelberg}, doi = {https://doi.org/10.1007/s10270-025-01277-2} }
- SANEROn the Compression of Language Models for Code: An Empirical Study on CodeBERTGiordano d’Aloisio, Luca Traini, Federica Sarro, and Antinisca Di MarcoIn IEEE/ACM International Conference on Software Analysis, Evolution, and Reengineering, 2025
Language models have proven successful across a wide range of software engineering tasks, but their significant computational costs often hinder their practical adoption. To address this challenge, researchers have begun applying various compression strategies to improve the efficiency of language models for code. These strategies aim to optimize inference latency and memory usage, though often at the cost of reduced model effectiveness. However, there is still a significant gap in understanding how these strategies influence the efficiency and effectiveness of language models for code. Here, we empirically investigate the impact of three well-known compression strategies – knowledge distillation, quantization, and pruning – across three different classes of software engineering tasks: vulnerability detection, code summarization, and code search. Our findings reveal that the impact of these strategies varies greatly depending on the task and the specific compression method employed. Practitioners and researchers can use these insights to make informed decisions when selecting the most appropriate compression strategy, balancing both efficiency and effectiveness based on their specific needs.
@inproceedings{daloisio_compression_2024, title = {On the {Compression} of {Language} {Models} for {Code}: {An} {Empirical} {Study} on {CodeBERT}}, copyright = {All rights reserved}, shorttitle = {On the {Compression} of {Language} {Models} for {Code}}, booktitle = {{IEEE}/{ACM} {International} {Conference} on {Software} {Analysis}, {Evolution}, and {Reengineering}}, urldate = {2025-03-11}, publisher = {arXiv}, author = {d'Aloisio, Giordano and Traini, Luca and Sarro, Federica and Marco, Antinisca Di}, year = {2025}, keywords = {Computer Science - Artificial Intelligence, Computer Science - Performance, Computer Science - Software Engineering}, file = {Preprint PDF:/Users/giord/Zotero/storage/H59UCM8I/d'Aloisio et al. - 2024 - On the Compression of Language Models for Code An.pdf:application/pdf;Snapshot:/Users/giord/Zotero/storage/BP5EMX5Y/2412.html:text/html} }
- JSSUncovering gender gap in academia: A comprehensive analysis within the software engineering communityAndrea D’Angelo, Giordano d’Aloisio, Francesca Marzi, Antinisca Di Marco, and Giovanni StiloJournal of Systems and Software, 2024
Gender gap in education has gained considerable attention in recent years, as it carries profound implications for the academic community. However, while the problem has been tackled from a student perspective, research is still lacking from an academic point of view. In this work, our main objective is to address this unexplored area by shedding light on the intricate dynamics of gender gap within the Software Engineering (SE) community. To this aim, we first review how the problem of gender gap in the SE community and in academia has been addressed by the literature so far. Results show that men in SE build more tightly-knit clusters but less global co-authorship relations than women, but the networks do not exhibit homophily. Concerning academic promotions, the Software Engineering community presents a higher bias in promotions to Associate Professors and a smaller bias in promotions to Full Professors than the overall Informatics community.
@article{DANGELO2024112162, title = {Uncovering gender gap in academia: A comprehensive analysis within the software engineering community}, journal = {Journal of Systems and Software}, pages = {112162}, year = {2024}, issn = {0164-1212}, doi = {https://doi.org/10.1016/j.jss.2024.112162}, url = {https://www.sciencedirect.com/science/article/pii/S0164121224002073}, author = {D’Angelo, Andrea and d’Aloisio, Giordano and Marzi, Francesca and {Di Marco}, Antinisca and Stilo, Giovanni}, keywords = {Gender gap, Gender bias, Academia, Italy, Informatics, Software engineering} }
- FASEDemocratizing Quality-Based Machine Learning Development through Extended Feature ModelsGiordano d’Aloisio, Antinisca Di Marco, and Giovanni StiloIn Fundamental Approaches to Software Engineering, 2023
ML systems have become an essential tool for experts of many domains, data scientists and researchers, allowing them to find answers to many complex business questions starting from raw datasets. Nevertheless, the development of ML systems able to satisfy the stakeholders’ needs requires an appropriate amount of knowledge about the ML domain. Over the years, several solutions have been proposed to automate the development of ML systems. However, an approach taking into account the new quality concerns needed by ML systems (like fairness, interpretability, privacy, and others) is still missing.
@inproceedings{daloisio_democratizing_2023, address = {Cham}, series = {Lecture {Notes} in {Computer} {Science}}, title = {Democratizing {Quality}-{Based} {Machine} {Learning} {Development} through {Extended} {Feature} {Models}}, copyright = {All rights reserved}, isbn = {978-3-031-30826-0}, doi = {https://doi.org/10.1007/978-3-031-30826-0_5}, language = {en}, booktitle = {Fundamental {Approaches} to {Software} {Engineering}}, publisher = {Springer Nature Switzerland}, author = {d’Aloisio, Giordano and Di Marco, Antinisca and Stilo, Giovanni}, editor = {Lambers, Leen and Uchitel, Sebastián}, year = {2023}, keywords = {/unread, Feature Models, Low-code development, Machine Learning System, Software Product Line, Software Quality}, pages = {88--110}, }
- IP&MDebiaser for Multiple Variables to enhance fairness in classification tasksGiordano d’Aloisio, Andrea D’Angelo, Antinisca Di Marco, and Giovanni StiloInformation Processing & Management, 2023
Nowadays assuring that search and recommendation systems are fair and do not apply discrimination among any kind of population has become of paramount importance. This is also highlighted by some of the sustainable development goals proposed by the United Nations. Those systems typically rely on machine learning algorithms that solve the classification task. Although the problem of fairness has been widely addressed in binary classification, unfortunately, the fairness of multi-class classification problem needs to be further investigated lacking well-established solutions. For the aforementioned reasons, in this paper, we present the Debiaser for Multiple Variables (DEMV), an approach able to mitigate unbalanced groups bias (i.e., bias caused by an unequal distribution of instances in the population) in both binary and multi-class classification problems with multiple sensitive variables. The proposed method is compared, under several conditions, with a set of well-established baselines using different categories of classifiers. At first we conduct a specific study to understand which is the best generation strategies and their impact on DEMV’s ability to improve fairness. Then, we evaluate our method on a heterogeneous set of datasets and we show how it overcomes the established algorithms of the literature in the multi-class classification setting and in the binary classification setting when more than two sensitive variables are involved. Finally, based on the conducted experiments, we discuss strengths and weaknesses of our method and of the other baselines.
@article{daloisio_debiaser_2023, title = {Debiaser for {Multiple} {Variables} to enhance fairness in classification tasks}, volume = {60}, copyright = {All rights reserved}, issn = {0306-4573}, url = {https://www.sciencedirect.com/science/article/pii/S0306457322003272}, doi = {https://doi.org/10.1016/j.ipm.2022.103226}, language = {en}, number = {2}, urldate = {2022-12-22}, journal = {Information Processing & Management}, author = {d’Aloisio, Giordano and D’Angelo, Andrea and Di Marco, Antinisca and Stilo, Giovanni}, year = {2023}, keywords = {Machine learning, Multi-class classification, Preprocessing algorithm, Bias and Fairness, Equality}, pages = {103226} }