The 2016 United States presidential election was marked by the abuse of targeted advertising on Facebook. Concerned with the risk of the same kind of abuse to happen in the 2018 Brazilian elections, we designed and deployed an independent auditing system to monitor political ads on Meta in Brazil. To do that we first adapted a browser plugin to gather ads from the timeline of volunteers using Facebook. We managed to convince more than 2,000 volunteers to help our project and install our tool. Then, we use a Convolution Neural Network (CNN) to detect political Meta ads using word embeddings. To evaluate our approach, we manually label a data collection of 10k ads as political or non-political and then we provide an in-depth evaluation of proposed approach for identifying political ads by comparing it with classic supervised machine learning methods. Finally, we deployed a real system that shows the ads identified as related to politics during the 2018 National Brazilian elections. We also investigated early electoral advertisement before the 2020 local Brazilian elections using our model on unsponsored content (regular posts in groups and pages). We noticed that not all political ads we detected were present in the Meta Ad Library for political ads on 2018. Additionally, we found possible early electoral advertisements in 2020, which is forbidden in Brazil. Our results emphasize the importance of enforcement mechanisms for declaring political ads and the need for independent auditing platforms.

The growth of social networks has raised concerns about the misuse of these platforms by disinformation campaigns, social bots, and coordinated activities. Among these platforms, WhatsApp has become a focal point for this abuse, particularly in Brazil, one of the countries with the highest use of the platform. Despite acknowledging the presence of coordinated campaigns and implementing restrictions on the number of messages forwarded per user, the platform continues to be abused. Due to its private nature and the difficulty of collecting information, little is known about these campaigns and the messages they disseminate. Given this context, our study investigates the presence of coordinated activities on WhatsApp in Brazil, identifying their content and purpose, especially how these messages relate to recent Brazilian political events. To answer these questions, we analyzed 13 million messages from 1,444 political groups over seven months from July 2022 to January 2023. Using network analysis, our findings suggest a significant prevalence of coordinated activity in the propagation of news messages, 26% of which originate from misinformation sites. Furthermore, we found that images play a key role in coordinated activity, accounting for 15% of messages, which are also used to mislead. Finally, coordinated accounts were used to organize collective actions, including attacks and protests against election results.

Instant messaging platforms have become an important means of communication in our world. According to WhatsApp, more than 100 billion messages are sent daily through the app. Communication on these platforms has allowed individuals to express themselves in other types of media, rather than simple text, including audio, videos, images, and, more recently, stickers. This new multimedia format, in particular, emerged with messaging apps and gained considerable popularity among users, promoting new forms of interactions. Stickers range from static images of memes and emojis to animated images similar to GIFs, often used in humorous contexts. However, in the Brazilian context of WhatsApp, they are transcending their role as a mere form of humor to become an important element in political strategy. In this regard, we investigate how stickers are used, revealing unique characteristics that these media bring to public WhatsApp groups and, more specifically, the political use of this new media format. Furthermore, we found evidence of sticker abuse on WhatsApp, where users attack political opponents and spread hate speech and offensive content in public groups without any moderation. To investigate this phenomenon, we collected a large sample of messages from public political WhatsApp groups in Brazil and analyzed the sticker messages shared in this context. Warning! This paper contains images and terms that may be offensive to some audiences.

Online messaging platforms are key communication tools but are vulnerable to fake news and conspiracy theories. Mainstream platforms such as Facebook are increasing content moderation of harmful and conspiratorial content. In response, users from fringe communities are migrating to alternative platforms like Telegram. These platforms offer more freedom and less intervention. Currently, Telegram is one of the leading messaging platforms hosting fringe communities. Despite the popularity, as a research community, we lack knowledge of how content spreads over this network. Motivated by the importance and impact of messaging platforms on society, we aim to measure the information propagation within fringe communities on the Telegram network, focusing on how public groups and channels exchange messages. We collect and explore about 140 million messages from 9,000 channels and groups on Telegram. We examine message forwarding and the lifetime of the messages from different aspects. Among other things, we find inequality in content creation; 6% of the users are responsible for 90% of forwarded messages. We also discover that while the forwarding feature considerably amplifies the reach of messages, the spread of content within our dataset remains largely localized. Additionally, we find that 5% of the channels are responsible for 40% of the forwarded messages in the entire dataset. Finally, our lifetime analysis shows that messages disseminated in groups with numerous active users exhibit significantly longer lifespans compared to those circulated in channels.

WhatsApp provides a fertile ground for the large-scale dissemination of information, particularly in countries like Brazil and India. Given its increasing popularity and use for political discussions, it is paramount to ensure that WhatsApp groups are adequately protected from attackers who aim to disrupt the activity of WhatsApp groups. Motivated by this, in this work, we characterize two types of attacks that may disrupt WhatsApp groups. We look into the flooding attack, where an attacker shares a large number of usually duplicate messages within a short period, and the hijacking attack, where attackers aim to obtain complete control of the group. We collect a large dataset of 19M messages shared in 1.6K WhatsApp public political groups from Brazil and analyze them to identify and characterize flooding and hijacking attacks. Among other things, we find that approximately 7% of the groups receive flooding attacks, which are usually short-lived (usually less than four minutes), and groups can receive multiple flooding attacks, even within the same day. Also, we find that most flooding attacks are executed using stickers (62% of all flooding attacks) and that, in most cases, attackers use both flooding and hijacking attacks to obtain complete control of the WhatsApp groups. Our work aims to raise user awareness about such attacks on WhatsApp and emphasizes the need to develop effective moderation tools to assist group administrators in preventing or mitigating such attacks.

WhatsApp has evolved into a popular communication tool, facilitating the exchange of billions of multimedia messages globally. With its large public groups and forwarding features, the platform has enabled messages to go viral, rapidly disseminating across the WhatsApp network. This has also brought WhatsApp to a central position in spreading misinformation campaigns, prompting the company to implement measures to counter bulk message dissemination, such as limiting simultaneous forwards and flagging viral content. Despite these measures, there remains a gap in our understanding of how forwarded messages function within this ecosystem and the effectiveness of the restrictions in containing the spread of viral content. In this study, we analyze approximately 10 million messages from 1,101 public WhatsApp groups dedicated to political discussion in Brazil, focusing on forwarded content. We investigate the structure of message forwarding, assess the reach of Forwarded Many Times (FTM) labeling mechanism, and evaluate the platform's ability to detect and flag duplicated media. Our findings reveal that forwarded messages constitute a substantial portion of the content shared in public WhatsApp groups. Moreover, we discover that the measures implemented by WhatsApp to restrict the dissemination of such messages can be easily circumvented, allowing users to intentionally bypass the architecture of the system and share media beyond the imposed limits. Notably, we identify that 59% of duplicated content flagged as FMT by WhatsApp does not receive the corresponding flag and find evidences of misinformation circulating virally in those groups. This research provides valuable insights into the dynamics of forwarded messages on WhatsApp and highlights the need for more effective strategies to combat the spread of viral content within the platform.

Conspiracy theories are widely propagated on social media. Among various social media services, YouTube is one of the most influential sources of news and entertainment. This paper seeks to develop a dataset, YOUNICON, to enable researchers to perform conspiracy theory detection as well as classification of videos with conspiracy theories into different topics. YOUNICON is a dataset with a large collection of videos from suspicious channels that were identified to contain conspiracy theories in a previous study. Overall, YOUNICON will enable researchers to study trends in conspiracy theories and understand how individuals can interact with the conspiracy theory producing community or channel. Our data is available at: https://doi.org/10.5281/zenodo.7466262.

A key gear in most misinformation ecosystems is the deployment of fake news web sites that publish news in a similar fashion to how news articles are put out by credible sources. The content offered by these sites is disseminated in a complex process that may involve automation, exploitation of message apps and social network algorithms, political bias, and targeted ads to reach large and niche audiences. Due to this high complexity and the rapidly evolving nature of the problem, we are just beginning to understand patterns in the various misinformation ecosystems on the Web. In this work, we offer a first step towards understanding network properties, including data from DNS records, domain registration, TLS certificates, and hosting infrastructure of Brazilian web sites associated with the dissemination of misinformation content on digital platforms. Our findings, in addition to providing a better understanding of the misinformation ecosystem in Brazil, also reveal a novel set of features useful to distinguish low credibility web sites from others.

Despite the valuable social interactions that online media promote, these systems provide space for speech that would be potentially detrimental to different groups of people. The moderation of content imposed by many social media has motivated the emergence of a new social system for free speech named Gab, which lacks moderation of content. This article characterizes and compares moderated textual data from Twitter with a set of unmoderated data from Gab. In particular, we analyze distinguishing characteristics of moderated and unmoderated content in terms of linguistic features, evaluate hate speech and its different forms in both environments. Our work shows that unmoderated content presents different psycholinguistic features, more negative sentiment and higher toxicity. Our findings support that unmoderated environments may have proportionally more online hate speech. We hope our analysis and findings contribute to the debate about hate speech and benefit systems aiming at deploying hate speech detection approaches.

As Internet users increasingly rely on social media sites to receive news, they are faced with a bewildering number of news media choices. For example, thousands of Facebook pages today are registered and categorized as some form of news media outlets. This situation boosted the so-called independent journalism, also known as alternative news media. Identifying and characterizing all the news pages that play an important role in news dissemination is key for understanding the news ecosystems of a country. In this work, we propose a graph-based semi-supervised method to measure the political bias of pages on most countries and show the political split of the alternative media, mainstream media, and public figures pages. We validate our method using the publicly available U.S. dataset and then apply it to Brazilian pages, where we found a larger number of right-wing pages in general, except for alternative news media.

WhatsApp is a free messaging app with more than one billion active monthly users which has become one of the main communication platforms in many countries, including Saudi Arabia, Germany, and Brazil. In addition to allowing the direct exchange of messages among pairs of users, the app also enables group conversations, where multiple people can interact with one another. A number of recent studies have shown that WhatsApp groups play an important role as an information dissemination platform, especially during important social mobilization events. In this paper, we build upon those prior efforts by taking a first look into the use of audio messages in WhatsApp groups, a type of content that is becoming increasingly important in the platform. We present a methodology to analyze audio messages shared in WhatsApp groups, characterizing content properties (e.g, topics and language characteristics), their propagation dynamics and the impact of different types of audios (e.g., speech versus music) on such dynamics.

The 2016 United States presidential election was marked by the abuse of targeted advertising on Facebook. Concerned with the risk of the same kind of abuse to happen in the 2018 Brazilian elections, we designed and deployed an independent auditing system to monitor political ads on Facebook in Brazil. To do that we first adapted a browser plugin to gather ads from the timeline of volunteers using Facebook. We managed to convince more than 2000 volunteers to help our project and install our tool. Then, we use a Convolution Neural Network (CNN) to detect political Facebook ads using word embeddings. To evaluate our approach, we manually label a data collection of 10k ads as political or non-political and then we provide an in-depth evaluation of proposed approach for identifying political ads by comparing it with classic supervised machine learning methods. Finally, we deployed a real system that shows the ads identified as related to politics. We noticed that not all political ads we detected were present in the Facebook Ad Library for political ads. Our results emphasize the importance of enforcement mechanisms for declaring political ads and the need for independent auditing platforms.

Measuring the affinity to a particular culture has been an active area of research. Countries and their residents can be characterized by many cultural aspects, such as clothing, music, art and food. As one of the central aspects, the cuisine of a country can reflect one of the dominant aspects of its culture. As such, the number of people interested in a typical national dish can be used to estimate the prevalence of that culture inside the host region. In this study, we measure the global spread of Brazilian culture across countries by exploring Facebook user’s preferences for typical Brazilian dishes through the Facebook Advertising Platform. To decide which dish will be considered typical from Brazil, we made use of spatial analysis to understand the distribution of interests around the world and to quantify how typical the dish is in Brazil and among Brazilian immigrants. This methodology can be generalized to other countries to infer cultural elements that emigrants usually take to and preserve in the countries they migrate to. Also, the interest in Brazilian typical dishes can be used to characterize countries in terms of Brazilian cultural exposition. While evaluating the cultural distance between Brazil and the countries with more Brazilian immigrants, we explore several measures of distance to compare these in the context of affinity to Brazilian cuisine. Our results revealed that these cultural distance measures can complement other metrics of distance applied to gravity-type models, for example, in order to explain flows of people between countries.

WhatsApp has revolutionized the way people communicate and interact. It is not only cheaper than the traditional Short Message Service (SMS) communication but it also brings a new form of mobile communication: the group chats. Such groups are great forums for collective discussions on a variety of topics. In particular, in events of great social mobilization, such as strikes and electoral campaigns, WhatsApp group chats are very attractive as they facilitate information exchange among interested people. Yet, recent events have raised concerns about the spreading of misinformation in WhatsApp. In this work, we analyze information dissemination within WhatsApp, focusing on publicly accessible political-oriented groups, collecting all shared messages during major social events in Brazil: a national truck drivers' strike and the Brazilian presidential campaign. We analyze the types of content shared within such groups as well as the network structures that emerge from user interactions within and cross-groups. We then deepen our analysis by identifying the presence of misinformation among the shared images using labels provided by journalists and by a proposed automatic procedure based on Google searches. We identify the most important sources of the fake images and analyze how they propagate across WhatsApp groups and from/to other Web platforms.

Sentiment analysis has become a very important tool for analysis of social media data. There are several methods developed, covering distinct aspects of the problem and disparate strategies. However, no single technique fits well in all cases or for all data sources. Supervised approaches may be able to adapt to specific situations, but require manually labeled training, which is very cumbersome and expensive to acquire, mainly for a new application. In this context, we propose to combine several popular and effective state‐of‐the‐practice sentiment analysis methods by means of an unsupervised bootstrapped strategy. One of our main goals is to reduce the large variability (low stability) of the unsupervised methods across different domains. The experimental results demonstrate that our combined method (aka, 10SENT) improves the effectiveness of the classification task, considering thirteen different data sets. Also, it tackles the key problem of cross‐domain low stability and produces the best (or close to best) results in almost all considered contexts, without any additional costs (e.g., manual labeling). Finally, we also investigate a transfer learning approach for sentiment analysis to gather additional (unsupervised) information for the proposed approach, and we show the potential of this technique to improve our results.

The moderation of content in many social media systems, such as Twitter and Facebook, motivated the emergence of a new social network system that promotes free speech, named Gab. Soon after that, Gab has been removed from Google Play Store for violating the company's hate speech policy and it has been rejected by Apple for similar reasons. In this paper we characterize Gab, aiming at understanding who are the users who joined it and what kind of content they share in this system. Our findings show that Gab is a very politically oriented system that hosts banned users from other social networks, some of them due to possible cases of hate speech and association with extremism. We provide the first measurement of news dissemination inside a right-leaning echo chamber, investigating a social media where readers are rarely exposed to content that cuts across ideological lines, but rather are fed with content that reinforces their current political or social views.

Online social networks (OSNs) such as Twitter and Facebook constitute an open space for developers to create sophisticated machines that imitate human users by automating their social network activities. The existence of socialbots is so powerful that bots can be transformed to influential users. Previous studies use fundamental functions such as posting a tweet or creating links to a specific target group to investigate the ability to infiltrate of these accounts. Our study analyzes the role of automated chatting in bots' infiltration. Our analysis is compared with the state of the art of this kind and reveals that the chat functionally was able to improve Klout and follow ratio on about 24% and 123% respectively. Also, the advanced communications skills contribute to more message interactions between socialbots and other accounts. Based on our empirical experimental study the conversational socialbots infiltrate more successfully in the Twitter-sphere.

Social media is considered a democratic space in which people connect and interact with each other regardless of their gender, race, or any other demographic factor. Despite numerous efforts that explore demographic factors in social media, it is still unclear whether social media perpetuates old inequalities from the offline world. In this paper, we attempt to identify gender and race of Twitter users located in U.S. using advanced image processing algorithms from Face++. Then, we investigate how different demographic groups (i.e. male/female, Asian/Black/White) connect with other. We quantify to what extent one group follow and interact with each other and the extent to which these connections and interactions reflect in inequalities in Twitter. Our analysis shows that users identified as White and male tend to attain higher positions in Twitter, in terms of the number of followers and number of times in user's lists. We hope our effort can stimulate the development of new theories of demographic information in the online space.

In this paper we build and characterize a large-scale dataset of internet radio streams. More than 25 million snapshots of more than 75 thousand different radio stations were collected from the SHOUTcast service between December 2016 and April 2017. We characterized several attributes of the dataset, such as audience and music genre distributions among radio stations, advertisement and seasonal content dynamics, as well as bit rates and media formats of the radio streams. Finally, we analyzed to which extent these features affect audience size. We hope these and the other findings of our study may provide valuable information for content personalization and better advertisement placement in internet radio streams.

Recently, there have been considerable efforts to use online data to investigate international migration. These efforts show that Web data are valuable for estimating migration rates and are relatively easy to obtain. However, existing studies have only investigated flows of people along migration corridors, i.e. between pairs of countries. In our work, we use data about “places lived” from millions of Google+ users in order to study migration `clusters', i.e. groups of countries in which individuals have lived sequentially. For the first time, we consider information about more than two countries people have lived in. We argue that these data are very valuable because this type of information is not available in traditional demographic sources which record country-to-country migration flows independent of each other. We show that migration clusters of country triads cannot be identified using information about bilateral flows alone. To demonstrate the additional insights that can be gained by using data about migration clusters, we first develop a model that tries to predict the prevalence of a given triad using only data about its constituent pairs. We then inspect the groups of three countries which are more or less prominent, compared to what we would expect based on bilateral flows alone. Next, we identify a set of features such as a shared language or colonial ties that explain which triple of country pairs are more or less likely to be clustered when looking at country triples. Then we select and contrast a few cases of clusters that provide some qualitative information about what our data set shows. The type of data that we use is potentially available for a number of social media services. We hope that this first study about migration clusters will stimulate the use of Web data for the development of new theories of international migration that could not be tested appropriately before.

The increasing use of smartphones to access social media platforms opens a new wave of applications that explore sentiment analysis in the mobile environment. However, there are various existing sentiment analysis methods and it is unclear which of them are deployable in the mobile environment. This paper provides the first of a kind study in which we compare the performance of 17 sentence-level sentiment analysis methods in the mobile environment. To do that, we adapted these sentence-level methods to run on Android OS and then we measure their performance in terms of memory usage, CPU usage, and battery consumption. Our findings unveil sentence-level methods that require almost no adaptations and run relatively fast as well as methods that could not be deployed due to excessive use of memory. We hope our effort provides a guide to developers and researchers interested in exploring sentiment analysis as part of a mobile application and can help new applications to be executed without the dependency of a server-side API.

In this paper we address the problem of automatically learning to classify the sentiment of short messages/reviews by exploiting information derived from meta-level features i.e., features derived primarily from the original bag-of-words representation. We propose new meta-level features especially designed for the sentiment analysis of short messages such as: (i) information derived from the sentiment distribution among the k nearest neighbors of a given short test document x, (ii) the distribution of distances of x to their neighbors and (iii) the document polarity of these neighbors given by unsupervised lexical-based methods. Our approach is also capable of exploiting information from the neighborhood of document x regarding (highly noisy) data obtained from 1.6 million Twitter messages with emoticons. The set of proposed features is capable of transforming the original feature space into a new one, potentially smaller and more informed. Experiments performed with a substantial number of datasets (nineteen) demonstrate that the effectiveness of the proposed sentiment-based meta-level features is not only superior to the traditional bag-of-word representation (by up to 16%) but is also superior in most cases to state-of-art meta-level features previously proposed in the literature for text classification tasks that do not take into account some idiosyncrasies of sentiment analysis. Our proposal is also largely superior to the best lexicon-based methods as well as to supervised combinations of them. In fact, the proposed approach is the only one to produce the best results in all tested datasets in all scenarios.

Online Social Networks (OSNs) such as Twitter and Facebook have become a significant testing ground for Artificial Intelligence developers who build programs, known as socialbots, that imitate actual users by automating their social-network activities such as forming social links and posting content. Particularly, Twitter users have shown difficulties in distinguishing these socialbots from the human users in their social graphs. Frequently, legitimate users engage in conversations with socialbots. More impressively, socialbots are effective in acquiring human users as followers and exercising influence within them. While the success of socialbots is certainly a remarkable achievement for AI practitioners, their proliferation in the Twitter-sphere opens many possibilities for cybercrime. The proliferation of socialbots in the Twitter-sphere motivates us to assess the characteristics or strategies that make socialbots most likely to succeed. In this direction, we created 120 socialbot accounts in Twitter, which have a profile, follow other users, and generate tweets either by reposting messages that others have posted or by creating their own synthetic tweets. Then, we employ a 2k factorial design experiment in order to quantify the infiltration effectiveness of different socialbot strategies. Our analysis is the first of a kind, and reveals what strategies make socialbots successful in the Twitter-sphere.

Influential people have an important role in the process of information diffusion. However, there are several ways to be influential, for example, to be the most popular or the first that adopts a new idea. In this paper we present a methodology to find trendsetters in information networks according to a specific topic of interest. Trendsetters are people that adopt and spread new ideas influencing other people before these ideas become popular. At the same time, not all early adopters are trendsetters because only few of them have the ability of propagating their ideas by their social contacts through word-of-mouth. Differently from other influence measures, a trendsetter is not necessarily popular or famous, but the one whose ideas spread over the graph successfully. Other metrics such as node in-degree or even standard Pagerank focus only in the static topology of the network. We propose a ranking strategy that focuses on the ability of some users to push new ideas that will be successful in the future. To that end, we combine temporal attributes of nodes and edges of the network with a Pagerank based algorithm to find the trendsetters for a given topic. To test our algorithm we conduct innovative experiments over a large Twitter dataset. We show that nodes with high in-degree tend to arrive late for new trends, while users in the top of our ranking tend to be early adopters that also influence their social contacts to adopt the new trend.

Finding topic experts on microblogging sites with millions of users, such as Twitter, is a hard and challenging problem. In this paper, we propose and investigate a new methodology for discovering topic experts in the popular Twitter social network. Our methodology relies on the wisdom of the Twitter crowds -- it leverages Twitter Lists, which are often carefully curated by individual users to include experts on topics that interest them and whose meta-data (List names and descriptions) provides valuable semantic cues to the experts' domain of expertise. We mined List information to build Cognos, a system for finding topic experts in Twitter. Detailed experimental evaluation based on a real-world deployment shows that: (a) Cognos infers a user's expertise more accurately and comprehensively than state-of-the-art systems that rely on the user's bio or tweet content, (b) Cognos scales well due to built-in mechanisms to efficiently update its experts' database with new users, and (c) Despite relying only on a single feature, namely crowdsourced Lists, Cognos yields results comparable to, if not better than, those given by the official Twitter experts search engine for a wide range of queries in user tests. Our study highlights Lists as a potentially valuable source of information for future content or expert search systems in Twitter.

Online Location Based Social Networks (LBSNs), which combine social network features with geographic information sharing, are becoming increasingly popular. One such application is Foursquare, which doubled its user population in less than six months. Among other features, Foursquare allows users to leave tips (i.e., reviews or recommendations) at specific venues as well as to give feedback on previously posted tips by adding them to their to-do lists or marking them as done. In this paper, we analyze how Foursquare users exploit these three features - tips, dones and to-dos - uncovering different behavior profiles. Our study reveals the existence of very active and influential users, some of which are famous businesses and brands, that seem engaged in posting tips at a large variety of venues while also receiving a great amount of user feedback on them. We also provide evidence of spamming, showing the existence of users that post tips whose contents are unrelated to the nature or domain of the venue where the tips were left.

Recently, Twitter has emerged as a popular platform for discovering real-time information on the Web, such as news stories and people's reaction to them. Like the Web, Twitter has become a target for link farming, where users, especially spammers, try to acquire large numbers of follower links in the social network. Acquiring followers not only increases the size of a user's direct audience, but also contributes to the perceived influence of the user, which in turn impacts the ranking of the user's tweets by search engines.

Understanding content popularity growth is of great importance to Internet service providers, content creators and online marketers. In this work, we characterize the growth patterns of video popularity on the currently most popular video sharing application, namely YouTube. Using newly provided data by the application, we analyze how the popularity of individual videos evolves since the video's upload time. Moreover, addressing a key aspect that has been mostly overlooked by previous work, we characterize the types of the referrers that most often attracted users to each video, aiming at shedding some light into the mechanisms (e.g., searching or external linking) that often drive users towards a video, and thus contribute to popularity growth. Our analyses are performed separately for three video datasets, namely, videos that appear in the YouTube top lists, videos removed from the system due to copyright violation, and videos selected according to random queries submitted to YouTube's search engine. Our results show that popularity growth patterns depend on the video dataset. In particular, copyright protected videos tend to get most of their views much earlier in their lifetimes, often exhibiting a popularity growth characterized by a viral epidemic-like propagation process. In contrast, videos in the top lists tend to experience sudden significant bursts of popularity. We also show that not only search but also other YouTube internal mechanisms play important roles to attract users to videos in all three datasets.

A number of online video social networks, out of which YouTube is the most popular, provides features that allow users to post a video as a response to a discussion topic. These features open opportunities for users to introduce polluted content, or simply pollution, into the system. For instance, spammers may post an unrelated video as response to a popular one aiming at increasing the likelihood of the response being viewed by a larger number of users. Moreover, opportunistic users--promoters--may try to gain visibility to a specific video by posting a large number of (potentially unrelated) responses to boost the rank of the responded video, making it appear in the top lists maintained by the system. Content pollution may jeopardize the trust of users on the system, thus compromising its success in promoting social interactions. In spite of that, the available literature is very limited in providing a deep understanding of this problem.