arXiv subjects

Marinella Petrocchi

Publications and source records attributed to Marinella Petrocchi.

At least 37 records · Page 2Linked to original sources

arXiv2023

The paradigm-shift of social spambots: Evidence, theories, and tools for the arms race

Recent studies in social media spam and automation provide anecdotal argumentation of the rise of a new generation of spambots, so-called social spambots. Here, for the first time, we extensively study this novel phenomenon on Twitter and we provide quantitative evidence that a paradigm-shift exists in spambot design. First, we measure current Twitter's capabilities of detecting the new social spambots. Later, we assess the human performance in discriminating between genuine accounts, social spambots, and traditional spambots. Then, we benchmark several state-of-the-art techniques proposed by the academic literature. Results show that neither Twitter, nor humans, nor cutting-edge applications are currently capable of accurately detecting the new social spambots. Our results call for new approaches capable of turning the tide in the fight against this raising phenomenon. We conclude by reviewing the latest literature on spambots detection and we highlight an emerging common research trend based on the analysis of collective behaviors. Insights derived from both our extensive experimental campaign and survey shed light on the most promising directions of research and lay the foundations for the arms race against the novel social spambots. Finally, to foster research on this novel phenomenon, we make publicly available to the scientific community all the datasets used in this study.

cs.SI↗

arXiv2023

Swinging in the States: Does disinformation on Twitter mirror the US presidential election system?

For more than a decade scholars have been investigating the disinformation flow on social media contextually to societal events, like, e.g., elections. In this paper, we analyze the Twitter traffic related to the US 2020 pre-election debate and ask whether it mirrors the electoral system. The U.S. electoral system provides that, regardless of the actual vote gap, the premier candidate who received more votes in one state `takes' that state. Criticisms of this system have pointed out that election campaigns can be more intense in particular key states to achieve victory, so-called {\it swing states}. Our intuition is that election debate may cause more traffic on Twitter-and probably be more plagued by misinformation-when associated with swing states. The results mostly confirm the intuition. About 88\% of the entire traffic can be associated with swing states, and links to non-trustworthy news are shared far more in swing-related traffic than the same type of news in safe-related traffic. Considering traffic origin instead, non-trustworthy tweets generated by automated accounts, so-called social bots, are mostly associated with swing states. Our work sheds light on the role an electoral system plays in the evolution of online debates, with, in the spotlight, disinformation and social bots.

cs.SI↗

arXiv2022

Bow-Tie Structures of Twitter Discursive Communities

In the analysis of Twitter debate, the recent literature focused on discursive communities, i.e. clusters of accounts interacting among themselves via retweets. In the present work, we studied discursive communities in 8 different thematic Twitter datasets in various languages. Surprisingly, we observed that almost all discursive communities therein display a bow-tie structure during political or societal debates. Instead, they are absent when the argument of the discussion is different as sport events, as in the case of Euro2020 Turkish and Italian datasets. We furthermore analysed the quality of the content created in the various sectors of the different discursive communities, using the domain annotation from the fact-checking website Newsguard: we observe that, when the discursive community is affected by m/disinformation, the content with the lowest quality is the ones produced and shared in SCC and, in particular, a strong incidence of low- or non-reputable messages is present in the flow of retweets between the SCC and the OUT sectors. In this sense, in discursive communities affected by m/disinformation, the greatest part of the accounts has access to a great variety of contents, but whose quality is, in general, quite low; such a situation perfectly describes the phenomenon of infodemic, i.e. the access to "an excessive amount of information about a problem, which makes it difficult to identify a solution", according to WHO).

cs.SI↗

arXiv2022

A Structured Analysis of Journalistic Evaluations for News Source Reliability

In today's era of information disorder, many organizations are moving to verify the veracity of news published on the web and social media. In particular, some agencies are exploring the world of online media and, through a largely manual process, ranking the credibility and transparency of news sources around the world. In this paper, we evaluate two procedures for assessing the risk of online media exposing their readers to m/disinformation. The procedures have been dictated by NewsGuard and The Global Disinformation Index, two well-known organizations combating d/misinformation via practices of good journalism. Specifically, considering a fixed set of media outlets, we examine how many of them were rated equally by the two procedures, and which aspects led to disagreement in the assessment. The result of our analysis shows a good degree of agreement, which in our opinion has a double value: it fortifies the correctness of the procedures and lays the groundwork for their automation.

cs.CY↗

arXiv2021

Adversarial machine learning for protecting against online manipulation

Adversarial examples are inputs to a machine learning system that result in an incorrect output from that system. Attacks launched through this type of input can cause severe consequences: for example, in the field of image recognition, a stop signal can be misclassified as a speed limit indication.However, adversarial examples also represent the fuel for a flurry of research directions in different domains and applications. Here, we give an overview of how they can be profitably exploited as powerful tools to build stronger learning models, capable of better-withstanding attacks, for two crucial tasks: fake news and social bot detection.

cs.LG↗

arXiv2021

Flow of online misinformation during the peak of the COVID-19 pandemic in Italy

The COVID-19 pandemic has impacted on every human activity and, because of the urgency of finding the proper responses to such an unprecedented emergency, it generated a diffused societal debate. The online version of this discussion was not exempted by the presence of d/misinformation campaigns, but differently from what already witnessed in other debates, the COVID-19 -- intentional or not -- flow of false information put at severe risk the public health, reducing the effectiveness of governments' countermeasures. In the present manuscript, we study the effective impact of misinformation in the Italian societal debate on Twitter during the pandemic, focusing on the various discursive communities. In order to extract the discursive communities, we focus on verified users, i.e. accounts whose identity is officially certified by Twitter. We thus infer the various discursive communities based on how verified users are perceived by standard ones: if two verified accounts are considered as similar by non unverified ones, we link them in the network of certified accounts. We first observe that, beside being a mostly scientific subject, the COVID-19 discussion show a clear division in what results to be different political groups. At this point, by using a commonly available fact-checking software (NewsGuard), we assess the reputation of the pieces of news exchanged. We filter the network of retweets (i.e. users re-broadcasting the same elementary piece of information, or tweet) from random noise and check the presence of messages displaying an url. The impact of misinformation posts reaches the 22.1% in the right and center-right wing community and its contribution is even stronger in absolute numbers, due to the activity of this group: 96% of all non reputable urls shared by political groups come from this community.

cs.SI↗

arXiv2020

Improving Opinion Spam Detection by Cumulative Relative Frequency Distribution

Over the last years, online reviews became very important since they can influence the purchase decision of consumers and the reputation of businesses, therefore, the practice of writing fake reviews can have severe consequences on customers and service providers. Various approaches have been proposed for detecting opinion spam in online reviews, especially based on supervised classifiers. In this contribution, we start from a set of effective features used for classifying opinion spam and we re-engineered them, by considering the Cumulative Relative Frequency Distribution of each feature. By an experimental evaluation carried out on real data from Yelp.com, we show that the use of the distributional features is able to improve the performances of classifiers.

cs.SI↗

arXiv2019

Do you really follow them? Automatic detection of credulous Twitter users

Online Social Media represent a pervasive source of information able to reach a huge audience. Sadly, recent studies show how online social bots (automated, often malicious accounts, populating social networks and mimicking genuine users) are able to amplify the dissemination of (fake) information by orders of magnitude. Using Twitter as a benchmark, in this work we focus on what we define credulous users, i.e., human-operated accounts with a high percentage of bots among their followings. Being more exposed to the harmful activities of social bots, credulous users may run the risk of being more influenced than other users; even worse, although unknowingly, they could become spreaders of misleading information (e.g., by retweeting bots). We design and develop a supervised classifier to automatically recognize credulous users. The best tested configuration achieves an accuracy of 93.27% and AUC-ROC of 0.93, thus leading to positive and encouraging results.

cs.SI↗

arXiv2019

The role of bot squads in the political propaganda on Twitter

Social Media are nowadays the privileged channel for information spreading and news checking. Unexpectedly for most of the users, automated accounts, also known as social bots, contribute more and more to this process of news spreading. Using Twitter as a benchmark, we consider the traffic exchanged, over one month of observation, on a specific topic, namely the migration flux from Northern Africa to Italy. We measure the significant traffic of tweets only, by implementing an entropy-based null model that discounts the activity of users and the virality of tweets. Results show that social bots play a central role in the exchange of significant content. Indeed, not only the strongest hubs have a number of bots among their followers higher than expected, but furthermore a group of them, that can be assigned to the same political tendency, share a common set of bots as followers. The retwitting activity of such automated accounts amplifies the presence on the platform of the hubs' messages.

cs.SI↗

arXiv2019

Better Safe Than Sorry: An Adversarial Approach to Improve Social Bot Detection

The arm race between spambots and spambot-detectors is made of several cycles (or generations): a new wave of spambots is created (and new spam is spread), new spambot filters are derived and old spambots mutate (or evolve) to new species. Recently, with the diffusion of the adversarial learning approach, a new practice is emerging: to manipulate on purpose target samples in order to make stronger detection models. Here, we manipulate generations of Twitter social bots, to obtain - and study - their possible future evolutions, with the aim of eventually deriving more effective detection techniques. In detail, we propose and experiment with a novel genetic algorithm for the synthesis of online accounts. The algorithm allows to create synthetic evolved versions of current state-of-the-art social bots. Results demonstrate that synthetic bots really escape current detection techniques. However, they give all the needed elements to improve such techniques, making possible a proactive approach for the design of social bot detection systems.

cs.SI↗

arXiv2018

Who framed Roger Reindeer? De-censorship of Facebook posts by snippet classification

This paper considers online news censorship and it concentrates on censorship of identities. Obfuscating identities may occur for disparate reasons, from military to judiciary ones. In the majority of cases, this happens to protect individuals from being identified and persecuted by hostile people. However, being the collaborative web characterised by a redundancy of information, it is not unusual that the same fact is reported by multiple sources, which may not apply the same restriction policies in terms of censorship. Also, the proven aptitude of social network users to disclose personal information leads to the phenomenon that comments to news can reveal the data withheld in the news itself. This gives us a mean to figure out who the subject of the censored news is. We propose an adaptation of a text analysis approach to unveil censored identities. The approach is tested on a synthesised scenario, which however resembles a real use case. Leveraging a text analysis based on a context classifier trained over snippets from posts and comments of Facebook pages, we achieve promising results. Despite the quite constrained settings in which we operate -- such as considering only snippets of very short length -- our system successfully detects the censored name, choosing among 10 different candidate names, in more than 50\% of the investigated cases. This outperforms the results of two reference baselines. The findings reported in this paper, other than being supported by a thorough experimental methodology and interesting on their own, also pave the way for further investigation on the insidious issues of censorship on the web.

cs.CL↗

arXiv2017

A study on text-score disagreement in online reviews

In this paper, we focus on online reviews and employ artificial intelligence tools, taken from the cognitive computing field, to help understanding the relationships between the textual part of the review and the assigned numerical score. We move from the intuitions that 1) a set of textual reviews expressing different sentiments may feature the same score (and vice-versa); and 2) detecting and analyzing the mismatches between the review content and the actual score may benefit both service providers and consumers, by highlighting specific factors of satisfaction (and dissatisfaction) in texts. To prove the intuitions, we adopt sentiment analysis techniques and we concentrate on hotel reviews, to find polarity mismatches therein. In particular, we first train a text classifier with a set of annotated hotel reviews, taken from the Booking website. Then, we analyze a large dataset, with around 160k hotel reviews collected from Tripadvisor, with the aim of detecting a polarity mismatch, indicating if the textual content of the review is in line, or not, with the associated score. Using well established artificial intelligence techniques and analyzing in depth the reviews featuring a mismatch between the text polarity and the score, we find that -on a scale of five stars- those reviews ranked with middle scores include a mixture of positive and negative aspects. The approach proposed here, beside acting as a polarity detector, provides an effective selection of reviews -on an initial very large dataset- that may allow both consumers and providers to focus directly on the review subset featuring a text/score disagreement, which conveniently convey to the user a summary of positive and negative features of the review target.

cs.CL↗

arXiv2017

Mining Worse and Better Opinions. Unsupervised and Agnostic Aggregation of Online Reviews

In this paper, we propose a novel approach for aggregating online reviews, according to the opinions they express. Our methodology is unsupervised - due to the fact that it does not rely on pre-labeled reviews - and it is agnostic - since it does not make any assumption about the domain or the language of the review content. We measure the adherence of a review content to the domain terminology extracted from a review set. First, we demonstrate the informativeness of the adherence metric with respect to the score associated with a review. Then, we exploit the metric values to group reviews, according to the opinions they express. Our experimental campaign has been carried out on two large datasets collected from Booking and Amazon, respectively.

cs.SI↗

arXiv2017

Social Fingerprinting: detection of spambot groups through DNA-inspired behavioral modeling

Spambot detection in online social networks is a long-lasting challenge involving the study and design of detection techniques capable of efficiently identifying ever-evolving spammers. Recently, a new wave of social spambots has emerged, with advanced human-like characteristics that allow them to go undetected even by current state-of-the-art algorithms. In this paper, we show that efficient spambots detection can be achieved via an in-depth analysis of their collective behaviors exploiting the digital DNA technique for modeling the behaviors of social network users. Inspired by its biological counterpart, in the digital DNA representation the behavioral lifetime of a digital account is encoded in a sequence of characters. Then, we define a similarity measure for such digital DNA sequences. We build upon digital DNA and the similarity between groups of users to characterize both genuine accounts and spambots. Leveraging such characterization, we design the Social Fingerprinting technique, which is able to discriminate among spambots and genuine accounts in both a supervised and an unsupervised fashion. We finally evaluate the effectiveness of Social Fingerprinting and we compare it with three state-of-the-art detection algorithms. Among the peculiarities of our approach is the possibility to apply off-the-shelf DNA analysis techniques to study online users behaviors and to efficiently rely on a limited number of lightweight account characteristics.

cs.SI↗

arXiv2016

Semi-supervised knowledge extraction for detection of drugs and their effects

New Psychoactive Substances (NPS) are drugs that lay in a grey area of legislation, since they are not internationally and officially banned, possibly leading to their not prosecutable trade. The exacerbation of the phenomenon is that NPS can be easily sold and bought online. Here, we consider large corpora of textual posts, published on online forums specialized on drug discussions, plus a small set of known substances and associated effects, which we call seeds. We propose a semi-supervised approach to knowledge extraction, applied to the detection of drugs (comprising NPS) and effects from the corpora under investigation. Based on the very small set of initial seeds, the work highlights how a contrastive approach and context deduction are effective in detecting substances and effects from the corpora. Our promising results, which feature a F1 score close to 0.9, pave the way for shortening the detection time of new psychoactive substances, once these are discussed and advertised on the Internet.

cs.CL↗

arXiv2016

Spotting the diffusion of New Psychoactive Substances over the Internet

Online availability and diffusion of New Psychoactive Substances (NPS) represent an emerging threat to healthcare systems. In this work, we analyse drugs forums, online shops, and Twitter. By mining the data from these sources, it is possible to understand the dynamics of drugs diffusion and their endorsement, as well as timely detecting new substances. We propose a set of visual analytics tools to support analysts in tackling NPS spreading and provide a better insight about drugs market and analysis.

cs.CY↗

arXiv2016

A matter of words: NLP for quality evaluation of Wikipedia medical articles

Automatic quality evaluation of Web information is a task with many fields of applications and of great relevance, especially in critical domains like the medical one. We move from the intuition that the quality of content of medical Web documents is affected by features related with the specific domain. First, the usage of a specific vocabulary (Domain Informativeness); then, the adoption of specific codes (like those used in the infoboxes of Wikipedia articles) and the type of document (e.g., historical and technical ones). In this paper, we propose to leverage specific domain features to improve the results of the evaluation of Wikipedia medical articles. In particular, we evaluate the articles adopting an "actionable" model, whose features are related to the content of the articles, so that the model can also directly suggest strategies for improving a given article quality. We rely on Natural Language Processing (NLP) and dictionaries-based techniques in order to extract the bio-medical concepts in a text. We prove the effectiveness of our approach by classifying the medical articles of the Wikipedia Medicine Portal, which have been previously manually labeled by the Wiki Project team. The results of our experiments confirm that, by considering domain-oriented features, it is possible to obtain sensible improvements with respect to existing solutions, mainly for those articles that other approaches have less correctly classified. Other than being interesting by their own, the results call for further research in the area of domain specific features suitable for Web data quality assessment.

cs.IR↗

arXiv2016

DNA-inspired online behavioral modeling and its application to spambot detection

We propose a strikingly novel, simple, and effective approach to model online user behavior: we extract and analyze digital DNA sequences from user online actions and we use Twitter as a benchmark to test our proposal. We obtain an incisive and compact DNA-inspired characterization of user actions. Then, we apply standard DNA analysis techniques to discriminate between genuine and spambot accounts on Twitter. An experimental campaign supports our proposal, showing its effectiveness and viability. To the best of our knowledge, we are the first ones to identify and adapt DNA-inspired techniques to online user behavioral modeling. While Twitter spambot detection is a specific use case on a specific social media, our proposed methodology is platform and technology agnostic, hence paving the way for diverse behavioral characterization tasks.

cs.SI↗