Bruno Bertaccini
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Bruno Bertaccini has a total of 94 co-authors. Here are the top co-authors by number of shared publications:
- Silvia Bacci
University of Florence, IT
15 shared publications - Alessandra Petrucci
University of Florence, IT
5 shared publications - Federico Crescenzi
Department of Economics, Engineering, Society and Business Administration, University of Tuscia, Viterbo, Italy
3 shared publications - Matilde Bini
European University of Rome
3 shared publications - Stefano Masini
Department of Statistics, Computer Science, Applications “G. Parenti” University of Florence Florence Italy
3 shared publications Alessandra Petrucci
2 shared publications
Latest Publications
- Open Access
- 1 July 2026
- Published by Wiley in International Journal of Tourism Research
- Vol. 28 (4)
A Visitor‐Centric Recommender System for Florentine Museums: A Foundation for Sustainable Tourism Management
The paper presents the development of a recommender system designed to capture visitors' preferences during museum visits in Florence. The system is trained on data from the entry records of the FirenzeCard , the official museum pass of the city offering tourists access to the Florentine museums. The system utilises an ensemble of deep learning models based on a two‐tower architecture, in which separate representations of visitors and museums are learned independently and then combined to generate predictions. It incorporates structured and unstructured data and contextual and visitor features. Results show that the integration of new components into the system architecture significantly enhances the recommender system's ability to accurately predict the top‐ K museums a visitor is likely to visit next. This aspect makes it one foundational layer of a larger decision‐making framework: a reinforcement learning agent to dynamically manage visitor flows, alleviate crowding in overexposed cultural sites and balance urban resources.
- 1 July 2026
- Published by Wiley in International Journal of Tourism Research
- Vol. 28 (4)
Open AccessA Visitor‐Centric Recommender System for Florentine Museums: A Foundation for Sustainable Tourism Management
The paper presents the development of a recommender system designed to capture visitors' preferences during museum visits in Florence. The system is trained on data from the entry records of the FirenzeCard , the official museum pass of the city offering tourists access to the Florentine museums. The system utilises an ensemble of deep learning models based on a two‐tower architecture, in which separate representations of visitors and museums are learned independently and then combined to generate predictions. It incorporates structured and unstructured data and contextual and visitor features. Results show that the integration of new components into the system architecture significantly enhances the recommender system's ability to accurately predict the top‐ K museums a visitor is likely to visit next. This aspect makes it one foundational layer of a larger decision‐making framework: a reinforcement learning agent to dynamically manage visitor flows, alleviate crowding in overexposed cultural sites and balance urban resources.
- Open Access
- 1 April 2026
- Published by Springer Nature in Italian Economic Journal
- p. 1-28
The Path to Sustainability Consciousness: The Role of Subjective Well-Being and Civic Engagement
This study examines how subjective well-being and civic engagement are statistically associated with Sustainability Consciousness, a multidimensional construct that encompasses knowingness, attitudes, and behaviours across environmental, social, and economic dimensions. Using structural equation modelling on data from 1040 Italian youngs aged 12–25, we find that both factors play a significant role in shaping sustainability consciousness, particularly through the impact of personal well-being dimensions such as satisfaction with free time and health, as well as engagement with collective movements such as Fridays for the Future. Our findings underscore the importance of integrating psychological well-being and civic participation into sustainability education and policy, thereby reinforcing the role of both individual and collective agency in fostering sustainable mindsets.
- 1 April 2026
- Published by Springer Nature in Italian Economic Journal
- p. 1-28
Open AccessThe Path to Sustainability Consciousness: The Role of Subjective Well-Being and Civic Engagement
This study examines how subjective well-being and civic engagement are statistically associated with Sustainability Consciousness, a multidimensional construct that encompasses knowingness, attitudes, and behaviours across environmental, social, and economic dimensions. Using structural equation modelling on data from 1040 Italian youngs aged 12–25, we find that both factors play a significant role in shaping sustainability consciousness, particularly through the impact of personal well-being dimensions such as satisfaction with free time and health, as well as engagement with collective movements such as Fridays for the Future. Our findings underscore the importance of integrating psychological well-being and civic participation into sustainability education and policy, thereby reinforcing the role of both individual and collective agency in fostering sustainable mindsets.
- 10 February 2026
- Published by Taylor & Francis in Technometrics
- Vol. 68 (3) , 1-15
Time Series Models for Cultural Heritage Preservation: The Case of the Brunelleschi’s Dome
- 10 February 2026
- Published by Taylor & Francis in Technometrics
- Vol. 68 (3) , 1-15
Time Series Models for Cultural Heritage Preservation: The Case of the Brunelleschi’s Dome
- Open Access
- 1 January 2026
- Published by Firenze University Press
- p. 297-302
Mitigating academic dropout through ML: The case study of the University of Florence
In Italy, student dropout represents one of the most critical challenges facing the higher education system. According to the ANVUR 2023 Report, the dropout rate between the first and second year of bachelor’s degree programs has shown a significant increase for students enrolled in the academic year 2020/21, rising to 14.5%, thereby returning to levels observed a decade ago. Dropout risks, however, are not limited to the first year: 20.4% of students abandon their studies within three years of enrollment in bachelor’s programs, increasing to 24.2% after six years. At the University of Florence, first-year dropout rates for bachelor’s programs exceed the national average, although a slight decline has been observed over the last four years (from 27.1% in 2020/21 to 25.3% in 2023/24). Contributing factors include insufficient academic preparation, lack of motivation, economic difficulties, and inadequate guidance, all resulting in a misalignment between student expectations and the academic experience. Dropout entails not only a substantial cost in terms of public and private resources, but also negatively affects human capital development and delays entry into the labour market. Given this context, the adoption of effective preventive policies—such as targeted pre-admission guidance and ongoing academic tutoring—is increasingly urgent. This study proposes a decision-support tool for university governance, aimed at the early identification of students at risk of dropping out. A machine learning model has been developed based on an ensemble of XGBoost, well-suited for handling complex problems with strong variable interactions, and TabNet, which provides advanced neural capabilities and selective attention mechanisms. Trained on enrollment cohorts at the University of Florence from 2020/21 to 2022/23, the model has demonstrated exceptionally high accuracy (approximately 97%) in predicting what has been observed in the 2023/24 academic year for all types of course program.
- 1 January 2026
- Published by Firenze University Press
- p. 297-302
Open AccessMitigating academic dropout through ML: The case study of the University of Florence
In Italy, student dropout represents one of the most critical challenges facing the higher education system. According to the ANVUR 2023 Report, the dropout rate between the first and second year of bachelor’s degree programs has shown a significant increase for students enrolled in the academic year 2020/21, rising to 14.5%, thereby returning to levels observed a decade ago. Dropout risks, however, are not limited to the first year: 20.4% of students abandon their studies within three years of enrollment in bachelor’s programs, increasing to 24.2% after six years. At the University of Florence, first-year dropout rates for bachelor’s programs exceed the national average, although a slight decline has been observed over the last four years (from 27.1% in 2020/21 to 25.3% in 2023/24). Contributing factors include insufficient academic preparation, lack of motivation, economic difficulties, and inadequate guidance, all resulting in a misalignment between student expectations and the academic experience. Dropout entails not only a substantial cost in terms of public and private resources, but also negatively affects human capital development and delays entry into the labour market. Given this context, the adoption of effective preventive policies—such as targeted pre-admission guidance and ongoing academic tutoring—is increasingly urgent. This study proposes a decision-support tool for university governance, aimed at the early identification of students at risk of dropping out. A machine learning model has been developed based on an ensemble of XGBoost, well-suited for handling complex problems with strong variable interactions, and TabNet, which provides advanced neural capabilities and selective attention mechanisms. Trained on enrollment cohorts at the University of Florence from 2020/21 to 2022/23, the model has demonstrated exceptionally high accuracy (approximately 97%) in predicting what has been observed in the 2023/24 academic year for all types of course program.
- Open Access
- 29 April 2024
- Published by Springer Nature in Quality & Quantity
- Vol. 58 (5) , 4751-4778
Measuring sustainability consciousness in Italy
Sustainability can be conceived as an ideal state in which society, environment and economy can thrive without harming each other in the present, future and global space. In this paper, we validate the Italian version of the sustainability consciousness questionnaire developed by a Swedish research group. The psychometric construct aims to measure individual sustainability consciousness through a combination of knowingness, attitudes and behaviour. The validation process is based on the hierarchical confirmatory factor analysis model and includes both the long and short forms of the questionnaire proposed in the original study. Specifically, we propose two alternative approaches to validate the short form, one based on the same subset of items as the original proposal and another based on a data-driven strategy that resulted in a different subset of items. Once validated on a national level, the sustainability consciousness questionnaire becomes a valuable tool for researchers and policymakers to assess the sustainability consciousness levels within a national or local population regarding this crucial issue, enabling the design of more precisely targeted policies.
- 29 April 2024
- Published by Springer Nature in Quality & Quantity
- Vol. 58 (5) , 4751-4778
Open AccessMeasuring sustainability consciousness in Italy
Sustainability can be conceived as an ideal state in which society, environment and economy can thrive without harming each other in the present, future and global space. In this paper, we validate the Italian version of the sustainability consciousness questionnaire developed by a Swedish research group. The psychometric construct aims to measure individual sustainability consciousness through a combination of knowingness, attitudes and behaviour. The validation process is based on the hierarchical confirmatory factor analysis model and includes both the long and short forms of the questionnaire proposed in the original study. Specifically, we propose two alternative approaches to validate the short form, one based on the same subset of items as the original proposal and another based on a data-driven strategy that resulted in a different subset of items. Once validated on a national level, the sustainability consciousness questionnaire becomes a valuable tool for researchers and policymakers to assess the sustainability consciousness levels within a national or local population regarding this crucial issue, enabling the design of more precisely targeted policies.
- Open Access
- 6 February 2024
- Published by Springer Nature in Data Mining and Knowledge Discovery
- Vol. 38 (3) , 1440-1465
Revealing the structural behaviour of Brunelleschi’s Dome with machine learning techniques
The Brunelleschi’s Dome is one of the most iconic symbols of the Renaissance and is among the largest masonry domes ever constructed. Since the late 17th century, first masonry cracks appeared on the Dome, giving the start to a monitoring activity. In modern times, since 1988 a monitoring system comprised of 166 electronic sensors, including deformometers and thermometers, has been in operation, providing a valuable source of real-time data on the monument’s health status. With the deformometers taking measurements at least four times per day, a vast amount of data is now available to explore the potential of the latest Artificial Intelligence and Machine Learning techniques in the field of historical-architectural heritage conservation. The objective of this contribution is twofold. Firstly, for the first time ever, we aim to unveil the overall structural behaviour of the Dome as a whole, as well as that of its specific sections (known as webs). We achieve this by evaluating the effectiveness of certain dimensionality reduction techniques on the extensive daily detections generated by the monitoring system, while also accounting for fluctuations in temperature over time. Secondly, we estimate a number of recurrent and convolutional neural network models to verify their capability for medium- and long-term prediction of the structural evolution of the Dome. We believe this contribution is an important step forward in the protection and preservation of historical buildings, showing the utility of machine learning in a context in which these are still little used.
- 6 February 2024
- Published by Springer Nature in Data Mining and Knowledge Discovery
- Vol. 38 (3) , 1440-1465
Open AccessRevealing the structural behaviour of Brunelleschi’s Dome with machine learning techniques
The Brunelleschi’s Dome is one of the most iconic symbols of the Renaissance and is among the largest masonry domes ever constructed. Since the late 17th century, first masonry cracks appeared on the Dome, giving the start to a monitoring activity. In modern times, since 1988 a monitoring system comprised of 166 electronic sensors, including deformometers and thermometers, has been in operation, providing a valuable source of real-time data on the monument’s health status. With the deformometers taking measurements at least four times per day, a vast amount of data is now available to explore the potential of the latest Artificial Intelligence and Machine Learning techniques in the field of historical-architectural heritage conservation. The objective of this contribution is twofold. Firstly, for the first time ever, we aim to unveil the overall structural behaviour of the Dome as a whole, as well as that of its specific sections (known as webs). We achieve this by evaluating the effectiveness of certain dimensionality reduction techniques on the extensive daily detections generated by the monitoring system, while also accounting for fluctuations in temperature over time. Secondly, we estimate a number of recurrent and convolutional neural network models to verify their capability for medium- and long-term prediction of the structural evolution of the Dome. We believe this contribution is an important step forward in the protection and preservation of historical buildings, showing the utility of machine learning in a context in which these are still little used.
- Open Access
Proso Millet (Panicum miliaceum L.) as Alternative Source of Starch and Phenolic Compounds: A Study on Twenty-Five Worldwide Accessions
Proso millet has been proposed as an effective anti-diabetic food thanks to the combined action of polyphenols and starch. This study aimed to characterize the chemical composition of twenty-five accessions, in order to enhance this cereal as an alternative to available starch for food applications or to propose new food ingredients with health benefits. Proso millet contained a high percentage of starch, reaching values of 58.51%. The amylose content showed high variability, with values ranging from 1.36 to 42.70%, and significantly higher values were recorded for the white accessions than for the colored ones. High-resistant starch content (13.41–26.07%) was also found. The HPLC-MS analysis showed the same phenolic pattern in all the samples. Cinnamic acids are the most abundant compounds and significant differences in their total content were found (0.69 to 1.35 mg/g DW), while flavonoids were only detected in trace amounts. Statistical results showed significantly higher antiradical activity in the colored accessions than in the white ones.
Open AccessProso Millet (Panicum miliaceum L.) as Alternative Source of Starch and Phenolic Compounds: A Study on Twenty-Five Worldwide Accessions
Proso millet has been proposed as an effective anti-diabetic food thanks to the combined action of polyphenols and starch. This study aimed to characterize the chemical composition of twenty-five accessions, in order to enhance this cereal as an alternative to available starch for food applications or to propose new food ingredients with health benefits. Proso millet contained a high percentage of starch, reaching values of 58.51%. The amylose content showed high variability, with values ranging from 1.36 to 42.70%, and significantly higher values were recorded for the white accessions than for the colored ones. High-resistant starch content (13.41–26.07%) was also found. The HPLC-MS analysis showed the same phenolic pattern in all the samples. Cinnamic acids are the most abundant compounds and significant differences in their total content were found (0.69 to 1.35 mg/g DW), while flavonoids were only detected in trace amounts. Statistical results showed significantly higher antiradical activity in the colored accessions than in the white ones.
- 1 August 2023
- Published by Wolters Kluwer Health in Circulation: Genomic and Precision Medicine
- Vol. 16 (4) , 363-371
Long-Term Prevalence of Systolic Dysfunction in MYBPC3 Versus MYH7-Related Hypertrophic Cardiomyopathy
BACKGROUND: The 2 sarcomere genes most commonly associated with hypertrophic cardiomyopathy (HCM), MYBPC3 (myosin-binding protein C3) and MYH7 (β-myosin heavy chain), are indistinguishable at presentation, and genotype-phenotype correlations have been elusive. Based on molecular and pathophysiological differences, however, it is plausible to hypothesize a different behavior in myocardial performance, impacting lifetime changes in left ventricular (LV) function. METHODS: We reviewed the initial and final echocardiograms of 402 consecutive HCM patients with pathogenic or likely pathogenic MYBPC3 (n=251) or MYH7 (n=151) mutations, followed over 9±8 years. RESULTS: At presentation, MYBPC3 patients were less frequently obstructive (15% versus 26%; P =0.005) and had lower LV ejection fraction compared with MYH7 (66±8% versus 68±8%, respectively; P =0.03). Both HCM patients harboring MYBPC3 and MYH7 mutations exhibited a small but significant decline in LV systolic function during follow-up; however, new onset of severe LV systolic dysfunction (LV ejection fraction, P =0.013). Prevalence of grade II/III diastolic dysfunction at final evaluation was comparable between MYBPC3 and MYH7 patients ( P =0.509). In a Cox multivariable analysis, MYBPC3-positive status (hazard ratio, 2.53 [95% CI, 1.09–5.82]; P =0.029), age (hazard ratio, 1.03 [95% CI, 1.00–1.06]; P =0.027), and atrial fibrillation (hazard ratio, 2.39 [95% CI, 1.14–5.05]; P =0.020) were independent predictors of severe systolic dysfunction. No statistically significant differences occurred with regard to incidence of atrial fibrillation, heart failure, appropriate implanted cardioverter defibrillator shock, or cardiovascular death. CONCLUSIONS: MYBPC3-related HCM showed increased long-term prevalence of systolic dysfunction compared with MYH7, in spite of similar outcome. Such observations suggest different pathophysiology of clinical progression in the 2 subsets and may prove relevant for understanding of genotype-phenotype correlations in HCM.
- 1 August 2023
- Published by Wolters Kluwer Health in Circulation: Genomic and Precision Medicine
- Vol. 16 (4) , 363-371
Long-Term Prevalence of Systolic Dysfunction in MYBPC3 Versus MYH7-Related Hypertrophic Cardiomyopathy
BACKGROUND: The 2 sarcomere genes most commonly associated with hypertrophic cardiomyopathy (HCM), MYBPC3 (myosin-binding protein C3) and MYH7 (β-myosin heavy chain), are indistinguishable at presentation, and genotype-phenotype correlations have been elusive. Based on molecular and pathophysiological differences, however, it is plausible to hypothesize a different behavior in myocardial performance, impacting lifetime changes in left ventricular (LV) function. METHODS: We reviewed the initial and final echocardiograms of 402 consecutive HCM patients with pathogenic or likely pathogenic MYBPC3 (n=251) or MYH7 (n=151) mutations, followed over 9±8 years. RESULTS: At presentation, MYBPC3 patients were less frequently obstructive (15% versus 26%; P =0.005) and had lower LV ejection fraction compared with MYH7 (66±8% versus 68±8%, respectively; P =0.03). Both HCM patients harboring MYBPC3 and MYH7 mutations exhibited a small but significant decline in LV systolic function during follow-up; however, new onset of severe LV systolic dysfunction (LV ejection fraction, P =0.013). Prevalence of grade II/III diastolic dysfunction at final evaluation was comparable between MYBPC3 and MYH7 patients ( P =0.509). In a Cox multivariable analysis, MYBPC3-positive status (hazard ratio, 2.53 [95% CI, 1.09–5.82]; P =0.029), age (hazard ratio, 1.03 [95% CI, 1.00–1.06]; P =0.027), and atrial fibrillation (hazard ratio, 2.39 [95% CI, 1.14–5.05]; P =0.020) were independent predictors of severe systolic dysfunction. No statistically significant differences occurred with regard to incidence of atrial fibrillation, heart failure, appropriate implanted cardioverter defibrillator shock, or cardiovascular death. CONCLUSIONS: MYBPC3-related HCM showed increased long-term prevalence of systolic dysfunction compared with MYH7, in spite of similar outcome. Such observations suggest different pathophysiology of clinical progression in the 2 subsets and may prove relevant for understanding of genotype-phenotype correlations in HCM.
- Open Access
- 22 June 2023
- Published by Springer Nature in Scientometrics
- Vol. 128 (8) , 4269-4303
Insights from the co-authorship network of the Italian academic statisticians
Nowadays, new technologies have favored communication among scholars from different universities and countries, and huge amount of data and scientific works have become more and more accessible. This has led to an increase in the multidisciplinarity of research products, but often also to a more specialized level of knowledge of the scholars. Therefore, while belonging to the same disciplinary field, scholars may present different working styles and willingness to collaborate according to their specific topics of interest. This plays a particularly relevant role in Italy, where tenured scholars in academic institutions are classified in sub-fields that, in turn, may be aggregated for purposes of recruitment and career advancement. Aim of this contribution is to propose a methodological approach to understand if the work and collaborative style of academic scholars belonging to different sub-fields is really so similar as to justify their grouping. For illustrative purposes, we focus on the co-authorship network of Italian academic statisticians relying on the database of scientific works published since 1990 until 2021 and downloaded by SCOPUS. From this database, we obtain a network composed of 758 nodes and 1730 edges. Some network measures at node level representing the work and collaborative style of scholars (i.e., number of publications, degree, degree strength, some centrality indices, transitivity, and external-internal index) are explained through quantile regression models. Results provide policy makers with useful insights on which sub-fields present significant differences in terms of research interests and collaborative style, thus not justifying their aggregation for recruitment and career advancement purposes.
- 22 June 2023
- Published by Springer Nature in Scientometrics
- Vol. 128 (8) , 4269-4303
Open AccessInsights from the co-authorship network of the Italian academic statisticians
Nowadays, new technologies have favored communication among scholars from different universities and countries, and huge amount of data and scientific works have become more and more accessible. This has led to an increase in the multidisciplinarity of research products, but often also to a more specialized level of knowledge of the scholars. Therefore, while belonging to the same disciplinary field, scholars may present different working styles and willingness to collaborate according to their specific topics of interest. This plays a particularly relevant role in Italy, where tenured scholars in academic institutions are classified in sub-fields that, in turn, may be aggregated for purposes of recruitment and career advancement. Aim of this contribution is to propose a methodological approach to understand if the work and collaborative style of academic scholars belonging to different sub-fields is really so similar as to justify their grouping. For illustrative purposes, we focus on the co-authorship network of Italian academic statisticians relying on the database of scientific works published since 1990 until 2021 and downloaded by SCOPUS. From this database, we obtain a network composed of 758 nodes and 1730 edges. Some network measures at node level representing the work and collaborative style of scholars (i.e., number of publications, degree, degree strength, some centrality indices, transitivity, and external-internal index) are explained through quantile regression models. Results provide policy makers with useful insights on which sub-fields present significant differences in terms of research interests and collaborative style, thus not justifying their aggregation for recruitment and career advancement purposes.