A5011228873- Complex Network Analysis Techniques
- Opinion Dynamics and Social Influence
- Misinformation and Its Impacts
- Spam and Phishing Detection
- Social Media and Politics
- Agriculture Sustainability and Environmental Impact
- Environmental Impact and Sustainability
- Protein Structure and Dynamics
- Energy and Environment Impacts
- Network Security and Intrusion Detection
- Complex Systems and Time Series Analysis
- Advanced Malware Detection Techniques
- Evolutionary Game Theory and Cooperation
- Web Data Mining and Analysis
- Topic Modeling
- Media Influence and Politics
- Caching and Content Delivery
- Food Waste Reduction and Sustainability
- Advanced Text Analysis Techniques
- Stochastic processes and statistical mechanics
- Theoretical and Computational Physics
- Game Theory and Applications
- Sustainable Supply Chain Management
- Bioinformatics and Genomic Networks
- Enzyme Structure and Function
Indiana University Bloomington
2016-2025
University of Rome Tor Vergata
2025
Food and Agriculture Organization of the United Nations
2008-2024
United Nations Industrial Development Organization
2019-2023
Azienda Sanitaria Locale Roma 3
2023
ASL Roma
2023
Indiana University
2007-2021
Politecnico di Milano
2021
Scuola Internazionale Superiore di Studi Avanzati
1996-2011
Biocom
2005-2007
The Turing test aimed to recognize the behavior of a human from that computer algorithm. Such challenge is more relevant than ever in today's social media context, where limited attention and technology constrain expressive power humans, while incentives abound develop software agents mimicking humans. These bots interact, often unnoticed, with real people ecosystems, but their abundance uncertain. While many are benign, one can design harmful goals persuading, smearing, or deceiving. Here...
Increasing evidence suggests that a growing amount of social media content is generated by autonomous entities known as bots. In this work we present framework to detect such on Twitter. We leverage more than thousand features extracted from public data and meta-data about users: friends, tweet sentiment, network patterns, activity time series. benchmark the classification using publicly available dataset Twitter This training enriched manually annotated collection active users include both...
We refine the information available through IPCC AR5 with regard to recent trends in global GHG emissions from agriculture, forestry and other land uses (AFOLU), including emission updates 2012. Using all three AFOLU datasets employed for analysis AR5, rather than just one as done WGIII Summary Policy Makers, our analyses point a down-revision of shares total anthropogenic emissions, while providing important additional on subsectoral trends. Our findings confirm that share declined over...
The wide adoption of social media has increased the competition among ideas for our finite attention. We employ a parsimonious agent-based model to study whether such may affect popularity different memes, diversity information we are exposed and fading collective interests specific topics. Agents share messages on network but can only pay attention portion they receive. In emerging dynamics diffusion, few memes go viral while most do not. predictions consistent with empirical data from...
The widespread adoption of social media for political communication creates unprecedented opportunities to monitor the opinions large numbers politically active individuals in real time. However, without a way distinguish between users opposing alignments, conflicting signals at individual level may, aggregate, obscure partisan differences opinion that are important strategy. In this article we describe several methods predicting alignment Twitter based on content and structure their run-up...
Abstract The massive spread of digital misinformation has been identified as a major threat to democracies. Communication, cognitive, social, and computer scientists are studying the complex causes for viral diffusion misinformation, while online platforms beginning deploy countermeasures. Little systematic, data-based evidence published guide these efforts. Here we analyze 14 million messages spreading 400 thousand articles on Twitter during ten months in 2016 2017. We find that social bots...
In this study we investigate how social media shape the networked public sphere and facilitate communication between communities with different political orientations. We examine two networks of on Twitter, comprised more than 250,000 tweets from six weeks leading up to 2010 U.S. congressional midterm elections. Using a combination network clustering algorithms manually-annotated data demonstrate that retweets exhibits highly segregated partisan structure, extremely limited connectivity...
We study astroturf political campaigns on microblogging platforms: politically-motivated individuals and organizations that use multiple centrally-controlled accounts to create the appearance of widespread support for a candidate or opinion. describe machine learning framework combines topological, content-based crowdsourced features information diffusion networks Twitter detect early stages viral spreading misinformation. present promising preliminary results with better than 96% accuracy...
Online social media are complementing and in some cases replacing person-to-person interaction redefining the diffusion of information. In particular, microblogs have become crucial grounds on which public relations, marketing, political battles fought. We demonstrate a web service that tracks memes Twitter helps detect astroturfing, smear campaigns, other misinformation context U.S. elections. also present abusive behaviors uncovered by our service. Our is based an extensible framework will...
While most online social media accounts are controlled by humans, these platforms also host automated agents called bots or sybil accounts. Recent literature reported on cases of imitating humans to manipulate discussions, alter the popularity users, pollute content and spread misinformation, even perform terrorist propaganda recruitment actions. Here we present BotOrNot, a publicly-available service that leverages more than one thousand features evaluate extent which Twitter account...
Traditional fact checking by expert journalists cannot keep up with the enormous volume of information that is now generated online. Computational may significantly enhance our ability to evaluate veracity dubious information. Here we show complexities human can be approximated quite well finding shortest path between concept nodes under properly defined semantic proximity metrics on knowledge graphs. Framed as a network problem this approach feasible efficient computational techniques. We...
We introduce a graph-generating model aimed at representing the evolution of protein interaction networks. The is based on hypothesis by duplication and divergence genes which produce proteins. obtained graphs have multifractal properties recovering absence characteristic connectivity as found in real data error tolerance to random or targeted damage very good agreement with behavior network analyses. proposed first step identification evolutionary dynamics leading development functions interactions.
The availability, access, utilization and stability of food supply over time are the four pillars security which support nutrition outcomes. Addressing issues raised globally around these remains a challenge. Food Agriculture Organization United Nations (FAO) 2017 report “The future agriculture: trends challenges” outlined challenges will have to be addressed in order for sustainable agricultural services cost-effectively meet growing demand world population. In this study, we systematically...
A Sleeping Beauty (SB) in science refers to a paper whose importance is not recognized for several years after publication. Its citation history exhibits long hibernation period followed by sudden spike of popularity. Previous studies suggest relative scarcity SBs. The reliability this conclusion is, however, heavily dependent on identification methods based arbitrary threshold parameters sleeping time and number citations, applied small or monodisciplinary bibliographic datasets. Here we...
From politicians and nation states to terrorist groups, numerous organizations reportedly conduct explicit campaigns influence opinions on social media, posing a risk freedom of expression. Thus, there is need identify eliminate "influence bots" - realistic, automated identities that illicitly shape discussions sites like Twitter Facebook before they get too influential.
We examine partisan differences in the behavior, communication patterns and social interactions of more than 18,000 politically-active Twitter users to produce evidence that points changing levels engagement with American online political landscape. Analysis a network defined by activity these proximity 2010 midterm congressional elections reveals highly segregated, well clustered community structure. Using cluster membership as high-fidelity (87% accuracy) proxy for affiliation, we...
We investigate the impact of community structure on information diffusion with linear threshold model. Our results demonstrate that modular may have counterintuitive effects when social reinforcement is present. show strong communities can facilitate global by enhancing local, intracommunity spreading. Using both analytic approaches and numerical simulations, we existence an optimal network modularity, where requires minimal number early adopters.Received 8 January...
Online popularity has an enormous impact on opinions, culture, policy, and profits. We provide a quantitative, large scale, temporal analysis of the dynamics online content in two massive model systems: Wikipedia entire country's Web space. find that are characterized by bursts, displaying characteristic features critical systems such as fat-tailed distributions magnitude interevent time. propose minimal combining classic preferential increase mechanism with occurrence random shifts due to...