Typology
operator:
and / or
Language
operator:
and / or
Date
operator:
and / or
Rights
operator:
and / or
2023
Conference article
Restricted
Semantic enrichment of explanations of AI models for healthcare
Corbucci L., Monreale A., Panigutti C., Natilli M., Smiraglio S., Pedreschi D.
Explaining AI-based clinical decision support systems is crucial to enhancing clinician trust in those powerful systems. Unfortunately, current explanations provided by eXplainable Artificial Intelligence techniques are not easily understandable by experts outside of AI. As a consequence, the enrichment of explanations with relevant clinical information concerning the health status of a patient is fundamental to increasing human experts' ability to assess the reliability of AI decisions. Therefore, in this paper, we propose a methodology to enable clinical reasoning by semantically enriching AI explanations. Starting with a medical AI explanation based only on the input features provided to the algorithm, our methodology leverages medical ontologies and NLP embedding techniques to link relevant information present in the patient's clinical notes to the original explanation. Our experiments, involving a human expert, highlight promising performance in correctly identifying relevant information about the diseases of the patients.Source: DS 2023: 26th International Conference on Discovery Science, pp. 216–229, Porto, Portugal, 09-11/10/2023
DOI: 10.1007/978-3-031-45275-8_15
Project(s): TAILOR
, HumanE-AI-Net , XAI , SoBigData-PlusPlus
Metrics:
Corbucci L., Monreale A., Panigutti C., Natilli M., Smiraglio S., Pedreschi D.
Explaining AI-based clinical decision support systems is crucial to enhancing clinician trust in those powerful systems. Unfortunately, current explanations provided by eXplainable Artificial Intelligence techniques are not easily understandable by experts outside of AI. As a consequence, the enrichment of explanations with relevant clinical information concerning the health status of a patient is fundamental to increasing human experts' ability to assess the reliability of AI decisions. Therefore, in this paper, we propose a methodology to enable clinical reasoning by semantically enriching AI explanations. Starting with a medical AI explanation based only on the input features provided to the algorithm, our methodology leverages medical ontologies and NLP embedding techniques to link relevant information present in the patient's clinical notes to the original explanation. Our experiments, involving a human expert, highlight promising performance in correctly identifying relevant information about the diseases of the patients.Source: DS 2023: 26th International Conference on Discovery Science, pp. 216–229, Porto, Portugal, 09-11/10/2023
DOI: 10.1007/978-3-031-45275-8_15
Project(s): TAILOR
, HumanE-AI-Net , XAI , SoBigData-PlusPlus Metrics:
See at:
doi.org 
| link.springer.com | CNR ExploRA
2022
Journal article
Open Access
The long-tail effect of the COVID-19 lockdown on Italians' quality of life, sleep and physical activity
Natilli M., Rossi A., Trecroci A., Cavaggioni L., Merati G., Formenti D.
From March 2020 to May 2021, several lockdown periods caused by the COVID-19 pandemic have limited people's usual activities and mobility in Italy, as well as around the world. These unprecedented confnement measures dramatically modifed citizens' daily lifestyles and behaviours. However, with the advent of summer 2021 and thanks to the vaccination campaign that signifcantly prevents serious illness and death, and reduces the risk of contagion, all the Italian regions fnally returned to regular behaviours and routines. Anyhow, it is unclear if there is a long-tail efect on people's quality of life, sleep- and physical activity-related behaviours. Thanks to the dataset described in this paper, it will be possible to obtain accurate insights of the changes induced by the lockdown period in the Italians' health that will permit to provide practical suggestions at local, regional, and state institutions and companies to improve infrastructures and services that could be benefcial to Italians' well being.Source: Scientific data (2022). doi:10.1038/s41597-022-01376-5
DOI: 10.1038/s41597-022-01376-5
Project(s): SoBigData-PlusPlus
Metrics:
Natilli M., Rossi A., Trecroci A., Cavaggioni L., Merati G., Formenti D.
From March 2020 to May 2021, several lockdown periods caused by the COVID-19 pandemic have limited people's usual activities and mobility in Italy, as well as around the world. These unprecedented confnement measures dramatically modifed citizens' daily lifestyles and behaviours. However, with the advent of summer 2021 and thanks to the vaccination campaign that signifcantly prevents serious illness and death, and reduces the risk of contagion, all the Italian regions fnally returned to regular behaviours and routines. Anyhow, it is unclear if there is a long-tail efect on people's quality of life, sleep- and physical activity-related behaviours. Thanks to the dataset described in this paper, it will be possible to obtain accurate insights of the changes induced by the lockdown period in the Italians' health that will permit to provide practical suggestions at local, regional, and state institutions and companies to improve infrastructures and services that could be benefcial to Italians' well being.Source: Scientific data (2022). doi:10.1038/s41597-022-01376-5
DOI: 10.1038/s41597-022-01376-5
Project(s): SoBigData-PlusPlus
Metrics:
See at:
ISTI Repository 
| www.nature.com | CNR ExploRA
2022
Conference article
Open Access
SoBigData RI: european integrated infrastructure for social mining and big data analytics
Trasarti R., Grossi V., Natilli M., Rapisarda B.
SoBigData RI has the ambition to support the rising demand for cross-disciplinary research and innovation on the multiple aspects of social complexity from combined data and model-driven perspectives and the increasing importance of ethics and data scientists' responsibility as pillars of trustworthy use of Big Data and analytical technology. Digital traces of human activities offer a considerable opportunity to scrutinize the ground truth of individual and collective behaviour at an unprecedented detail and on a global scale. This increasing wealth of data is a chance to understand social complexity, provided we can rely on social mining, i.e., adequate means for accessing big social data and models for extracting knowledge from them. SoBigData RI, with its tools and services, empowers researchers and innovators through a platform for the design and execution of large-scale social mining experiments, open to users with diverse backgrounds, accessible on the cloud (aligned with EOSC), and also exploiting supercomputing facilities. Pushing the FAIR (Findable, Accessible, Interoperable) and FACT (Fair, Accountable, Confidential, and Transparent) principles will render social mining experiments more efficiently designed, adjusted, and repeatable by domain experts that are not data scientists. SoBigData RI moves forward from the simple awareness of ethical and legal challenges in social mining to the development of concrete tools that operationalize ethics with value-sensitive design, incorporating values and norms for privacy protection, fairness, transparency, and pluralism. SoBigData RI is the result of two H2020 grants (g.a. n.654024 and 871042), and it is part of the ESFRI 2021 Roadmap.Source: SEBD 2022 - The 30th Italian Symposium on Advanced Database Systems, pp. 117–124, Tirrenia (PI), Italy, 19-22/06/2022
Project(s): SoBigData-PlusPlus
Trasarti R., Grossi V., Natilli M., Rapisarda B.
SoBigData RI has the ambition to support the rising demand for cross-disciplinary research and innovation on the multiple aspects of social complexity from combined data and model-driven perspectives and the increasing importance of ethics and data scientists' responsibility as pillars of trustworthy use of Big Data and analytical technology. Digital traces of human activities offer a considerable opportunity to scrutinize the ground truth of individual and collective behaviour at an unprecedented detail and on a global scale. This increasing wealth of data is a chance to understand social complexity, provided we can rely on social mining, i.e., adequate means for accessing big social data and models for extracting knowledge from them. SoBigData RI, with its tools and services, empowers researchers and innovators through a platform for the design and execution of large-scale social mining experiments, open to users with diverse backgrounds, accessible on the cloud (aligned with EOSC), and also exploiting supercomputing facilities. Pushing the FAIR (Findable, Accessible, Interoperable) and FACT (Fair, Accountable, Confidential, and Transparent) principles will render social mining experiments more efficiently designed, adjusted, and repeatable by domain experts that are not data scientists. SoBigData RI moves forward from the simple awareness of ethical and legal challenges in social mining to the development of concrete tools that operationalize ethics with value-sensitive design, incorporating values and norms for privacy protection, fairness, transparency, and pluralism. SoBigData RI is the result of two H2020 grants (g.a. n.654024 and 871042), and it is part of the ESFRI 2021 Roadmap.Source: SEBD 2022 - The 30th Italian Symposium on Advanced Database Systems, pp. 117–124, Tirrenia (PI), Italy, 19-22/06/2022
Project(s): SoBigData-PlusPlus
See at:
ceur-ws.org | ISTI Repository | CNR ExploRA
2019
Contribution to book
Open Access
Analysis and visualization of performance indicators in university admission tests
Natilli M., Fadda D., Rinzivillo S., Pedreschi D., Licari F.
This paper presents an analytical platform for evaluation of the performance and anomaly detection of tests for admission to public universities in Italy. Each test is personalized for each student and is composed of a series of questions, classified on different domains (e.g. maths, science, logic, etc.). Since each test is unique for composition, it is crucial to guarantee a similar level of difficulty for all the tests in a session. For this reason, to each question, it is assigned a level of difficulty from a domain expert. Thus, the general difficultness of a test depends on the correct classification of each item. We propose two approaches to detect outliers. A visualization-based approach using dynamic filter and responsive visual widgets. A data mining approach to evaluate the performance of the different questions for five years. We used clustering to group the questions according to a set of performance indicators to provide labeling of the data-driven level of difficulty. The measured level is compared with the a priori assigned by experts. The misclassifications are then highlighted to the expert, who will be able to refine the ques- tion or the classification. Sequential pattern mining is used to check if biases are present in the composition of the tests and their performance. This analysis is meant to exclude overlaps or direct dependencies among questions. Analyzing co-occurrences we are able to state that the compo- sition of each test is fair and uniform for all the students, even on several sessions. The analytical results are presented to the expert through a visual web application that loads the analytical data and indicators and composes an interactive dashboard. The user may explore the patterns and models extracted by filtering and changing thresholds and analytical parameters.Source: Formal Methods. FM 2019 International Workshops, edited by Emil Sekerinski et al..., pp. 186–199, 2019
DOI: 10.1007/978-3-030-54994-7_14
Metrics:
Natilli M., Fadda D., Rinzivillo S., Pedreschi D., Licari F.
This paper presents an analytical platform for evaluation of the performance and anomaly detection of tests for admission to public universities in Italy. Each test is personalized for each student and is composed of a series of questions, classified on different domains (e.g. maths, science, logic, etc.). Since each test is unique for composition, it is crucial to guarantee a similar level of difficulty for all the tests in a session. For this reason, to each question, it is assigned a level of difficulty from a domain expert. Thus, the general difficultness of a test depends on the correct classification of each item. We propose two approaches to detect outliers. A visualization-based approach using dynamic filter and responsive visual widgets. A data mining approach to evaluate the performance of the different questions for five years. We used clustering to group the questions according to a set of performance indicators to provide labeling of the data-driven level of difficulty. The measured level is compared with the a priori assigned by experts. The misclassifications are then highlighted to the expert, who will be able to refine the ques- tion or the classification. Sequential pattern mining is used to check if biases are present in the composition of the tests and their performance. This analysis is meant to exclude overlaps or direct dependencies among questions. Analyzing co-occurrences we are able to state that the compo- sition of each test is fair and uniform for all the students, even on several sessions. The analytical results are presented to the expert through a visual web application that loads the analytical data and indicators and composes an interactive dashboard. The user may explore the patterns and models extracted by filtering and changing thresholds and analytical parameters.Source: Formal Methods. FM 2019 International Workshops, edited by Emil Sekerinski et al..., pp. 186–199, 2019
DOI: 10.1007/978-3-030-54994-7_14
Metrics:
See at:
ISTI Repository | doi.org | link.springer.com | CNR ExploRA
2019
Conference article
Open Access
A visual analytics platform to measure performance on university entrance tests
Boncoraglio D., Deri F., Distefano F., Fadda D., Filippi G., Forte G., Licari F., Natilli M., Pedreschi D., Rinzivillo S.
Data visualization dashboards provide an efficient approach that helps to improve the ability to understand the information behind complex databases. It is possible with such tools to create new insights, to represent keys indicators of the activity, to communicate (in real-time) snapshots of the state of the work. In this paper, we present a visual analytics platform created for the exploration and analysis of performance data on entrance tests taken by Italian students when entering the university career. The data is provided by CISIA (Consorzio Interuniversitario Sistemi Integrati per l'Accesso), a non-profit consortium formed exclusively by public universities. With this platform, it is possible to explore the performance of the students along different dimensions, such as gender, high school of provenience, type of test and so on.Source: 27th Italian Symposium on Advanced Database Systems, Castiglione della Pescaia, Grosseto, Italy (Grosseto), Italy, 16-19 June 2019
Boncoraglio D., Deri F., Distefano F., Fadda D., Filippi G., Forte G., Licari F., Natilli M., Pedreschi D., Rinzivillo S.
Data visualization dashboards provide an efficient approach that helps to improve the ability to understand the information behind complex databases. It is possible with such tools to create new insights, to represent keys indicators of the activity, to communicate (in real-time) snapshots of the state of the work. In this paper, we present a visual analytics platform created for the exploration and analysis of performance data on entrance tests taken by Italian students when entering the university career. The data is provided by CISIA (Consorzio Interuniversitario Sistemi Integrati per l'Accesso), a non-profit consortium formed exclusively by public universities. With this platform, it is possible to explore the performance of the students along different dimensions, such as gender, high school of provenience, type of test and so on.Source: 27th Italian Symposium on Advanced Database Systems, Castiglione della Pescaia, Grosseto, Italy (Grosseto), Italy, 16-19 June 2019
See at:
ceur-ws.org | ISTI Repository | CNR ExploRA
2019
Conference article
Closed Access
Exploring students eating habits through individual profiling and clustering analysis
Natilli M., Monreale A., Guidotti R., Pappalardo L.
Individual well-being strongly depends on food habits, therefore it is important to educate the general population, and especially young people, to the importance of a healthy and balanced diet. To this end, understanding the real eating habits of people becomes fundamental for a better and more effective intervention to improve the students' diet. In this paper we present two exploratory analyses based on centroid-based clustering that have the goal of understanding the food habits of university students. The first clustering analysis simply exploits the information about the students' food consumption of specific food categories, while the second exploratory analysis includes the temporal dimension in order to capture the information about when the students consume specific foods. The second approach enables the study of the impact of the time of consumption on the choice of the food.Source: PAP 2018 - The 2nd International Workshop on Personal Analytics and Privacy, pp. 156–171, Dublin, Ireland, 10-14 September 2018
DOI: 10.1007/978-3-030-13463-1_12
Project(s): SoBigData
Metrics:
Natilli M., Monreale A., Guidotti R., Pappalardo L.
Individual well-being strongly depends on food habits, therefore it is important to educate the general population, and especially young people, to the importance of a healthy and balanced diet. To this end, understanding the real eating habits of people becomes fundamental for a better and more effective intervention to improve the students' diet. In this paper we present two exploratory analyses based on centroid-based clustering that have the goal of understanding the food habits of university students. The first clustering analysis simply exploits the information about the students' food consumption of specific food categories, while the second exploratory analysis includes the temporal dimension in order to capture the information about when the students consume specific foods. The second approach enables the study of the impact of the time of consumption on the choice of the food.Source: PAP 2018 - The 2nd International Workshop on Personal Analytics and Privacy, pp. 156–171, Dublin, Ireland, 10-14 September 2018
DOI: 10.1007/978-3-030-13463-1_12
Project(s): SoBigData
Metrics:
See at:
doi.org | link.springer.com | CNR ExploRA
2016
Report
Restricted
Database of female scientists at the 6 targeted Departments of UNIPI
Romano V., Natilli M., Fadda D., Rossetti G., Giannotti F.
The experience gained with the several funded European projects allows us to collect data on female careers but also to identify the context (at institutional level) as a crucial factor in defining the phenomenon of gender equality. The usual approach is to perform a survey or to ask the administration in order to understand how many women are employed at the different levels of the institution at a certain time. The institution obtains a snapshot of the gender equality or, if the study is repeated regularly, a sequence of snapshots that allows gender researchers to perform comparisons and better understand the trends. The aim of the Women Scientific Career Database is to integrate the study of gender equality in the structure of the administration of the institution, in order to build a permanent gender monitor that is automatically updated by the administration. This new approach allows a real time analysis of the gender equality within the institution and, as data are continuously updated, makes it easier to verify how different strategies, laws or regulations can modify the status of gender equality. In order to better understand how the career of a researcher evolves within the institution through years, a lot of different events have to be monitored, like the type of contract and its evolution, and scientific production. The analysis provides statistics aggregated at university level, at department level and personal level in order to give a global picture of the university status and to show how different departments present different behaviors with respect to the gender inequalities. The personal level aggregation wants instead to show how real women scientist can have a successful career. This report describes the realization of the Women Scientific Career Database, the data model, the acquisition procedures and the implementation of first family of indicators and their rendering through a navigable web interface.Source: Project report, TRIGGER, Deliverable D1.8, 2016
Project(s): TRIGGER
Romano V., Natilli M., Fadda D., Rossetti G., Giannotti F.
The experience gained with the several funded European projects allows us to collect data on female careers but also to identify the context (at institutional level) as a crucial factor in defining the phenomenon of gender equality. The usual approach is to perform a survey or to ask the administration in order to understand how many women are employed at the different levels of the institution at a certain time. The institution obtains a snapshot of the gender equality or, if the study is repeated regularly, a sequence of snapshots that allows gender researchers to perform comparisons and better understand the trends. The aim of the Women Scientific Career Database is to integrate the study of gender equality in the structure of the administration of the institution, in order to build a permanent gender monitor that is automatically updated by the administration. This new approach allows a real time analysis of the gender equality within the institution and, as data are continuously updated, makes it easier to verify how different strategies, laws or regulations can modify the status of gender equality. In order to better understand how the career of a researcher evolves within the institution through years, a lot of different events have to be monitored, like the type of contract and its evolution, and scientific production. The analysis provides statistics aggregated at university level, at department level and personal level in order to give a global picture of the university status and to show how different departments present different behaviors with respect to the gender inequalities. The personal level aggregation wants instead to show how real women scientist can have a successful career. This report describes the realization of the Women Scientific Career Database, the data model, the acquisition procedures and the implementation of first family of indicators and their rendering through a navigable web interface.Source: Project report, TRIGGER, Deliverable D1.8, 2016
Project(s): TRIGGER
See at:
triggerproject.eu | CNR ExploRA