Michele Coscia
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Michele Coscia has a total of 50 co-authors. Here are the top co-authors by number of shared publications:
Fosca Giannotti
KDDLab, ISTI-CNR
4 shared publicationsDiego Pennacchioli
ISTI, CNR, Via G. Moruzzi, 1, Pisa, Italy
3 shared publicationsDino Pedreschi
KDDLab, University of Pisa
3 shared publications- Fosca Giannotti
Scuola Normale Superiore
3 shared publications - Dino Pedreschi
University of Pisa, Pisa, Italy
2 shared publications - Giulio Rossetti
Institute of Information Science and Technologies “A. Faedo” (ISTI), National Research Council (CNR), 56127 Pisa, Italy
2 shared publications
Latest Publications
- Open Access
- 6 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 4407-4418
Multilayer Network Backboning
- 6 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 4407-4418
Open AccessMultilayer Network Backboning
- Open Access
- 6 October 2026
- Published by Springer Nature in npj Complexity
The economic complexity of the Roman Empire as inferred from Latin inscriptions
- 6 October 2026
- Published by Springer Nature in npj Complexity
Open AccessThe economic complexity of the Roman Empire as inferred from Latin inscriptions
- Open Access
- 29 April 2026
- Published by The Royal Society in Royal Society Open Science
- Vol. 13 (4)
Evaluating fraud detection algorithms in a decentralized scenario
Financial fraud is an umbrella term including a vast number of illegal activities. These activities involve a significant fraction of the global economy. Traditional investigation techniques are labour-intensive and cannot scale to match the size of the issue. Machine learning has provided effective tools which deliver high accuracy in identifying transactions that could be involved in fraudulent activities. In this paper, we point out that the state-of-the-art in financial fraud detection has been applied to the unrealistic scenario of an omniscient centralized global authority which has access to all bank transactions globally. We propose a more realistic evaluation scenario, one made of two steps: first, the bank flags its own transactions using exclusively information it possesses; then only flagged transactions from all banks are analysed by the governmental authority for potential prosecution. We find that, in such a realistic scenario, the effectiveness of the state-of-the-art method for financial fraud detection decreases. Moreover, we show that in this decentralized scenario, it pays off to use simpler methods than the state-of-the-art, depending on the specific objective function the system wants to ensure.
- 29 April 2026
- Published by The Royal Society in Royal Society Open Science
- Vol. 13 (4)
Open AccessEvaluating fraud detection algorithms in a decentralized scenario
Financial fraud is an umbrella term including a vast number of illegal activities. These activities involve a significant fraction of the global economy. Traditional investigation techniques are labour-intensive and cannot scale to match the size of the issue. Machine learning has provided effective tools which deliver high accuracy in identifying transactions that could be involved in fraudulent activities. In this paper, we point out that the state-of-the-art in financial fraud detection has been applied to the unrealistic scenario of an omniscient centralized global authority which has access to all bank transactions globally. We propose a more realistic evaluation scenario, one made of two steps: first, the bank flags its own transactions using exclusively information it possesses; then only flagged transactions from all banks are analysed by the governmental authority for potential prosecution. We find that, in such a realistic scenario, the effectiveness of the state-of-the-art method for financial fraud detection decreases. Moreover, we show that in this decentralized scenario, it pays off to use simpler methods than the state-of-the-art, depending on the specific objective function the system wants to ensure.
- Open Access
- 23 February 2026
- Published by Springer Nature in Applied Network Science
- Vol. 11 (1)
Disconnect between the public face and the voting behavior of political representatives
- 23 February 2026
- Published by Springer Nature in Applied Network Science
- Vol. 11 (1)
Open AccessDisconnect between the public face and the voting behavior of political representatives
- Open Access
- 1 January 2026
- Published by Wiley in Complexity
- Vol. 2026 (1)
Comparing the Italian and Danish Music Industries With Network Analysis
Making music is a collaborative effort, and understanding artistic endeavors with a network science lens has been a prolific field of study. Recently, a new approach produced a novel database of Italian music and analyzed it with innovative network analysis techniques to find music eras, a data‐driven genre classification and a specialization index based on how bands share artists to publish records. In this work, we expand the previous work by adding data from Denmark and recording music labels’ information. We are interested in investigating whether the music eras and genre classifications found in previous work were universal, or whether they are driven by context‐specific dynamics in a country. We find that only the most superficial structural statistics are preserved between Denmark and Italy. However, music eras, genre classifications, and artist specialization are actually different between the two countries. This shows how, to get a more complete picture of contemporary music production, the view from a single country is not enough.
- 1 January 2026
- Published by Wiley in Complexity
- Vol. 2026 (1)
Open AccessComparing the Italian and Danish Music Industries With Network Analysis
Making music is a collaborative effort, and understanding artistic endeavors with a network science lens has been a prolific field of study. Recently, a new approach produced a novel database of Italian music and analyzed it with innovative network analysis techniques to find music eras, a data‐driven genre classification and a specialization index based on how bands share artists to publish records. In this work, we expand the previous work by adding data from Denmark and recording music labels’ information. We are interested in investigating whether the music eras and genre classifications found in previous work were universal, or whether they are driven by context‐specific dynamics in a country. We find that only the most superficial structural statistics are preserved between Denmark and Italy. However, music eras, genre classifications, and artist specialization are actually different between the two countries. This shows how, to get a more complete picture of contemporary music production, the view from a single country is not enough.
- Open Access
- 24 September 2025
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 20 (9) , e0328210
Estimating affective polarization on a social network
Concerns about polarization and hate speech on social media are widespread. Affective polarization, i.e., hostility among partisans, is crucial in this regard as it links political disagreements to hostile language online. However, only a few methods are available to measure how affectively polarized an online debate is, and the existing approaches do not investigate jointly two defining features of affective polarization: hostility and social distance. To address this methodological gap, we propose a network-based measure of affective polarization that combines both aspects – which allows them to be studied independently. We show that our measure accurately captures the relation between the level of disagreement and the hostility expressed towards others (affective component) and whom individuals choose to interact with or avoid (social distance component). Applying our measure to a large-scale Twitter data set on COVID-19, we find that affective polarization was low in February 2020 and increased to high levels as more users joined the Twitter discussion in the following months.
- 24 September 2025
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 20 (9) , e0328210
Open AccessEstimating affective polarization on a social network
Concerns about polarization and hate speech on social media are widespread. Affective polarization, i.e., hostility among partisans, is crucial in this regard as it links political disagreements to hostile language online. However, only a few methods are available to measure how affectively polarized an online debate is, and the existing approaches do not investigate jointly two defining features of affective polarization: hostility and social distance. To address this methodological gap, we propose a network-based measure of affective polarization that combines both aspects – which allows them to be studied independently. We show that our measure accurately captures the relation between the level of disagreement and the hostility expressed towards others (affective component) and whom individuals choose to interact with or avoid (social distance component). Applying our measure to a large-scale Twitter data set on COVID-19, we find that affective polarization was low in February 2020 and increased to high levels as more users joined the Twitter discussion in the following months.
- Open Access
The Economic Complexity of the Roman Empire
Economic complexity is a powerful tool to estimate the productive capabilities and future growth of modern economies. Little is known of how economic complexity evolves over long periods in history. In this paper, we use archaeological evidence from the Roman Empire in the form of short texts preserved on a durable material (i.e. inscriptions) to estimate the economic complexity of the various provinces of the empire. By connecting the occupations listed in the text of inscriptions with the location in which the inscribed objects were found we can estimate that the most complex areas during the first four centuries of the Roman Empire have a remarkable and statistically significant overlap with the most complex countries today. While we lack an explanation for the reason of the preservation of economic complexity through the ages, this evidence provides a suggestion about how difficult the development of economic capabilities might be.
Open AccessThe Economic Complexity of the Roman Empire
Economic complexity is a powerful tool to estimate the productive capabilities and future growth of modern economies. Little is known of how economic complexity evolves over long periods in history. In this paper, we use archaeological evidence from the Roman Empire in the form of short texts preserved on a durable material (i.e. inscriptions) to estimate the economic complexity of the various provinces of the empire. By connecting the occupations listed in the text of inscriptions with the location in which the inscribed objects were found we can estimate that the most complex areas during the first four centuries of the Roman Empire have a remarkable and statistically significant overlap with the most complex countries today. While we lack an explanation for the reason of the preservation of economic complexity through the ages, this evidence provides a suggestion about how difficult the development of economic capabilities might be.
- 23 January 2025
- Published by Springer Nature in Journal of Archaeological Method and Theory
- Vol. 32 (1) , 1-36
“A Network of Mutualities of Being”: Socio-material Archaeological Networks and Biological Ties at Çatalhöyük
- 23 January 2025
- Published by Springer Nature in Journal of Archaeological Method and Theory
- Vol. 32 (1) , 1-36
“A Network of Mutualities of Being”: Socio-material Archaeological Networks and Biological Ties at Çatalhöyük
- Open Access
- 5 September 2024
- Published by Springer Nature in Applied Network Science
- Vol. 9 (1) , 1-20
Node attribute analysis for cultural data analytics: a case study on Italian XX–XXI century music
- 5 September 2024
- Published by Springer Nature in Applied Network Science
- Vol. 9 (1) , 1-20
Open AccessNode attribute analysis for cultural data analytics: a case study on Italian XX–XXI century music
- 1 May 2024
- Published by World Scientific Pub Co Pte Ltd in Advances in Complex Systems
- Vol. 27 (03)
TRACES OF UNEQUAL ENTRY REQUIREMENT FOR ILLUSTRIOUS PEOPLE ON WIKIPEDIA BASED ON THEIR GENDER
Wikipedia is a widely used tool people use to gather knowledge about the world, causing it to have a vast impact on the way individuals perceive the reality they live in. It is then of paramount importance that the picture of the world Wikipedia provides is accurate. We cannot afford such an important tool to eschew inclusiveness or a fair representation of reality: an inaccurate picture of the world in such a tool can be used to claim unjust and unfair positions — such as that women are inferior to men — as if they were facts, because they are enshrined on an encyclopedia. In this paper, we study issues of fair gender representations for people in history noted by multiple language editions of Wikipedia: are women underrepresented on Wikipedia? We do so via a combination of natural language processing and network science. Our results indicate that there is indeed a higher bar for women to have their own biographical page on Wikipedia: women are only included when they have more significant connections than men to the rest of the network. There are visible effects of the initiatives Wikipedia is taking to fix this issue, showing that the gap is narrowing, which validates our interpretation of the data.
- 1 May 2024
- Published by World Scientific Pub Co Pte Ltd in Advances in Complex Systems
- Vol. 27 (03)
TRACES OF UNEQUAL ENTRY REQUIREMENT FOR ILLUSTRIOUS PEOPLE ON WIKIPEDIA BASED ON THEIR GENDER
Wikipedia is a widely used tool people use to gather knowledge about the world, causing it to have a vast impact on the way individuals perceive the reality they live in. It is then of paramount importance that the picture of the world Wikipedia provides is accurate. We cannot afford such an important tool to eschew inclusiveness or a fair representation of reality: an inaccurate picture of the world in such a tool can be used to claim unjust and unfair positions — such as that women are inferior to men — as if they were facts, because they are enshrined on an encyclopedia. In this paper, we study issues of fair gender representations for people in history noted by multiple language editions of Wikipedia: are women underrepresented on Wikipedia? We do so via a combination of natural language processing and network science. Our results indicate that there is indeed a higher bar for women to have their own biographical page on Wikipedia: women are only included when they have more significant connections than men to the rest of the network. There are visible effects of the initiatives Wikipedia is taking to fix this issue, showing that the gap is narrowing, which validates our interpretation of the data.