Manuel Pratelli
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Manuel Pratelli has a total of 18 co-authors. Here are the top co-authors by number of shared publications:
- Marinella Petrocchi
IMT School for Advanced Studies Lucca
12 shared publications - Fabio Saracco
Centro Studi e Ricerche "Enrico Fermi" Roma Italy
5 shared publications - Fabio Pinelli
IMT School for Advanced Studies Lucca
4 shared publications - John Bianchi
IMT School for Advanced Studies Lucca, Lucca, Italy
3 shared publications Rocco De Nicola
IMT Scuola Alti Studi Lucca, Lucca, Italy
3 shared publications- Rocco De Nicola
IMT School for Advanced Studies Lucca
3 shared publications
Latest Publications
- Open Access
- 25 May 2026
- Published by Association for Computing Machinery (ACM)
- p. 75-78
News Source Profiling from Articles text: A Benchmark Study
- 25 May 2026
- Published by Association for Computing Machinery (ACM)
- p. 75-78
Open AccessNews Source Profiling from Articles text: A Benchmark Study
- Open Access
- 21 October 2025
- Published by IOS Press
- p. 944-951
Evaluating the Simulation of Human Personality-Driven Susceptibility to Misinformation with LLMs
Recent advances in large language models (LLMs) promise to facilitate the creation of synthetic agents that simulate realistic human behaviors, offering social sciences an ethical and scalable alternative to experiments with individuals. Whether such agents can faithfully capture psychological differences driven by personality traits, however, remains an open question. We evaluate the capacity of LLM agents, conditioned on Big-Five profiles, to reproduce personality-based variation in susceptibility to misinformation, focusing on news discernment, the ability to judge true headlines as true and false headlines as false. We observe that LLM agents are able to simulate human-like associations between personality traits and susceptibility to misinformation for socially desirable traits (Agreeableness, Conscientiousness, Open-Mindedness), but diverge for less desirable traits. The results underscore both the promise and the limits of personality-aligned LLMs for online human behavioral simulation.
- 21 October 2025
- Published by IOS Press
- p. 944-951
Open AccessEvaluating the Simulation of Human Personality-Driven Susceptibility to Misinformation with LLMs
Recent advances in large language models (LLMs) promise to facilitate the creation of synthetic agents that simulate realistic human behaviors, offering social sciences an ethical and scalable alternative to experiments with individuals. Whether such agents can faithfully capture psychological differences driven by personality traits, however, remains an open question. We evaluate the capacity of LLM agents, conditioned on Big-Five profiles, to reproduce personality-based variation in susceptibility to misinformation, focusing on news discernment, the ability to judge true headlines as true and false headlines as false. We observe that LLM agents are able to simulate human-like associations between personality traits and susceptibility to misinformation for socially desirable traits (Agreeableness, Conscientiousness, Open-Mindedness), but diverge for less desirable traits. The results underscore both the promise and the limits of personality-aligned LLMs for online human behavioral simulation.
- Open Access
On the efficacy of old features for the detection of new bots
For more than a decade now, academicians and online platform administrators have been studying solutions to the problem of bot detection. Bots are computer algorithms whose use is far from being benign: malicious bots are purposely created to distribute spam, sponsor public characters and, ultimately, induce a bias within the public opinion. To fight the bot invasion on our online ecosystem, several approaches have been implemented, mostly based on (supervised and unsupervised) classifiers, which adopt the most varied account features, from the simplest to the most expensive ones to be extracted from the raw data obtainable through the Twitter public APIs. In this exploratory study, using Twitter as a benchmark, we compare the performances of four state-of-art feature sets in detecting novel bots: one of the output scores of the popular bot detector Botometer, which considers more than 1,000 features of an account to take a decision; two feature sets based on the account profile and timeline; and the information about the Twitter client from which the user tweets. The results of our analysis, conducted on six recently released datasets of Twitter accounts, hint at the possible use of general-purpose classifiers and cheap-to-compute account features for the detection of evolved bots.
Open AccessOn the efficacy of old features for the detection of new bots
For more than a decade now, academicians and online platform administrators have been studying solutions to the problem of bot detection. Bots are computer algorithms whose use is far from being benign: malicious bots are purposely created to distribute spam, sponsor public characters and, ultimately, induce a bias within the public opinion. To fight the bot invasion on our online ecosystem, several approaches have been implemented, mostly based on (supervised and unsupervised) classifiers, which adopt the most varied account features, from the simplest to the most expensive ones to be extracted from the raw data obtainable through the Twitter public APIs. In this exploratory study, using Twitter as a benchmark, we compare the performances of four state-of-art feature sets in detecting novel bots: one of the output scores of the popular bot detector Botometer, which considers more than 1,000 features of an account to take a decision; two feature sets based on the account profile and timeline; and the information about the Twitter client from which the user tweets. The results of our analysis, conducted on six recently released datasets of Twitter accounts, hint at the possible use of general-purpose classifiers and cheap-to-compute account features for the detection of evolved bots.
- Open Access
- 19 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 179-188
Evaluation of Reliability Criteria for News Publishers with Large Language Models
- 19 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 179-188
Open AccessEvaluation of Reliability Criteria for News Publishers with Large Language Models
- 3 April 2025
- Published by Springer Nature
- p. 407-412
TROPIC – Trustworthiness Rating of Online Publishers Through Online Interactions Calculation
- 3 April 2025
- Published by Springer Nature
- p. 407-412
TROPIC – Trustworthiness Rating of Online Publishers Through Online Interactions Calculation
- 9 October 2024
- Published by Springer Nature
- p. 13-26
Systems Security Modeling and Analysis at IMT Lucca
- 9 October 2024
- Published by Springer Nature
- p. 13-26
Systems Security Modeling and Analysis at IMT Lucca
- Open Access
- 21 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 139-148
Unveiling News Publishers Trustworthiness Through Social Interactions
With the primary goal of raising readers’ awareness of misinformation phenomena, extensive efforts have been made by both academic institutions and independent organizations to develop methodologies for assessing the trustworthiness of online news publishers. Unfortunately, existing approaches are costly and face critical scalability challenges. This study presents a novel framework for assessing the trustworthiness of online news publishers using user interactions on social media platforms. The proposed methodology provides a versatile solution that serves the dual purpose of i) identifying verifiable online publishers and ii) automatically performing an initial estimation of the trustworthiness of previously unclassified online news outlets.
- 21 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 139-148
Open AccessUnveiling News Publishers Trustworthiness Through Social Interactions
With the primary goal of raising readers’ awareness of misinformation phenomena, extensive efforts have been made by both academic institutions and independent organizations to develop methodologies for assessing the trustworthiness of online news publishers. Unfortunately, existing approaches are costly and face critical scalability challenges. This study presents a novel framework for assessing the trustworthiness of online news publishers using user interactions on social media platforms. The proposed methodology provides a versatile solution that serves the dual purpose of i) identifying verifiable online publishers and ii) automatically performing an initial estimation of the trustworthiness of previously unclassified online news outlets.
- Open Access
- 25 April 2024
- Published by Oxford University Press (OUP) in PNAS Nexus
- Vol. 3  (5) , pgae177
Entropy-based detection of Twitter echo chambers
Echo chambers, i.e. clusters of users exposed to news and opinions in line with their previous beliefs, were observed in many online debates on social platforms. We propose a completely unbiased entropy-based method for detecting echo chambers. The method is completely agnostic to the nature of the data. In the Italian Twitter debate about the Covid-19 vaccination, we find a limited presence of users in echo chambers (about 0.35% of all users). Nevertheless, their impact on the formation of a common discourse is strong, as users in echo chambers are responsible for nearly a third of the retweets in the original dataset. Moreover, in the case study observed, echo chambers appear to be a receptacle for disinformative content.
- 25 April 2024
- Published by Oxford University Press (OUP) in PNAS Nexus
- Vol. 3  (5) , pgae177
Open AccessEntropy-based detection of Twitter echo chambers
Echo chambers, i.e. clusters of users exposed to news and opinions in line with their previous beliefs, were observed in many online debates on social platforms. We propose a completely unbiased entropy-based method for detecting echo chambers. The method is completely agnostic to the nature of the data. In the Italian Twitter debate about the Covid-19 vaccination, we find a limited presence of users in echo chambers (about 0.35% of all users). Nevertheless, their impact on the formation of a common discourse is strong, as users in echo chambers are responsible for nearly a third of the retweets in the original dataset. Moreover, in the case study observed, echo chambers appear to be a receptacle for disinformative content.
- Open Access
- 8 April 2024
- Published by Association for Computing Machinery (ACM)
- p. 671-678
Evaluating Trustworthiness of Online News Publishers via Article Classification
The proliferation of low-quality online information in today's era has underscored the need for robust and automatic mechanisms to evaluate the trustworthiness of online news publishers. In this paper, we analyse the trustworthiness of online news media outlets by leveraging a dataset of 4033 news stories from 40 different sources. We aim to infer the trustworthiness level of the source based on the classification of individual articles' content. The trust labels are obtained from NewsGuard, a journalistic organization that evaluates news sources using well-established editorial and publishing criteria. The results indicate that the classification model is highly effective in classifying the trustworthiness levels of the news articles. This research has practical applications in alerting readers to potentially untrustworthy news sources, assisting journalistic organizations in evaluating new or unfamiliar media outlets and supporting the selection of articles for their trustworthiness assessment.
- 8 April 2024
- Published by Association for Computing Machinery (ACM)
- p. 671-678
Open AccessEvaluating Trustworthiness of Online News Publishers via Article Classification
The proliferation of low-quality online information in today's era has underscored the need for robust and automatic mechanisms to evaluate the trustworthiness of online news publishers. In this paper, we analyse the trustworthiness of online news media outlets by leveraging a dataset of 4033 news stories from 40 different sources. We aim to infer the trustworthiness level of the source based on the classification of individual articles' content. The trust labels are obtained from NewsGuard, a journalistic organization that evaluates news sources using well-established editorial and publishing criteria. The results indicate that the classification model is highly effective in classifying the trustworthiness levels of the news articles. This research has practical applications in alerting readers to potentially untrustworthy news sources, assisting journalistic organizations in evaluating new or unfamiliar media outlets and supporting the selection of articles for their trustworthiness assessment.
- Open Access
- 26 March 2024
- Published by Springer Nature in EPJ Data Science
- Vol. 13  (1) , 1-23
Online disinformation in the 2020 U.S. election: swing vs. safe states
For U.S. presidential elections, most states use the so-called winner-take-all system, in which the state’s presidential electors are awarded to the winning political party in the state after a popular vote phase, regardless of the actual margin of victory. Therefore, election campaigns are especially intense in states where there is no clear direction on which party will be the winning party. These states are often referred to as swing states. To measure the impact of such an election law on the campaigns, we analyze the Twitter activity surrounding the 2020 US preelection debate, with a particular focus on the spread of disinformation. We find that about 88% of the online traffic was associated with swing states. In addition, the sharing of links to unreliable news sources is significantly more prevalent in tweets associated with swing states: in this case, untrustworthy tweets are predominantly generated by automated accounts. Furthermore, we observe that the debate is mostly led by two main communities, one with a predominantly Republican affiliation and the other with accounts of different political orientations. Most of the disinformation comes from the former.
- 26 March 2024
- Published by Springer Nature in EPJ Data Science
- Vol. 13  (1) , 1-23
Open AccessOnline disinformation in the 2020 U.S. election: swing vs. safe states
For U.S. presidential elections, most states use the so-called winner-take-all system, in which the state’s presidential electors are awarded to the winning political party in the state after a popular vote phase, regardless of the actual margin of victory. Therefore, election campaigns are especially intense in states where there is no clear direction on which party will be the winning party. These states are often referred to as swing states. To measure the impact of such an election law on the campaigns, we analyze the Twitter activity surrounding the 2020 US preelection debate, with a particular focus on the spread of disinformation. We find that about 88% of the online traffic was associated with swing states. In addition, the sharing of links to unreliable news sources is significantly more prevalent in tweets associated with swing states: in this case, untrustworthy tweets are predominantly generated by automated accounts. Furthermore, we observe that the debate is mostly led by two main communities, one with a predominantly Republican affiliation and the other with accounts of different political orientations. Most of the disinformation comes from the former.