The recent development of decentralised and interoperable social networks (such as the "fediverse") creates new challenges for content moderators. This is because millions of posts generated on one server can easily "spread" to another, even if the recipient server has very different moderation policies. An obvious solution would be to leverage moderation tools to automatically tag (and filter) posts that contravene moderation policies, e.g. related to toxic speech. Recent work has exploited the conversational context of a post to improve this automatic tagging, e.g. using the replies to a post to help classify if it contains toxic speech. This has shown particular potential in environments with large training sets that contain complete conversations. This, however, creates challenges in a decentralised context, as a single conversation may be fragmented across multiple servers. Thus, each server only has a partial view of an entire conversation because conversations are often federated across servers in a non-synchronized fashion. To address this, we propose a decentralised conversation-aware content moderation approach suitable for the fediverse. Our approach employs a graph deep learning model (GraphNLI) trained locally on each server. The model exploits local data to train a model that combines post and conversational information captured through random walks to detect toxicity. We evaluate our approach with data from Pleroma, a major decentralised and interoperable micro-blogging network containing 2 million conversations. Our model effectively detects toxicity on larger instances, exclusively trained using their local post information (0.8837 macro-F1). Yet, we show that this approach does not perform well on smaller instances that do not possess sufficient local training data. Thus, in cases where a server contains insufficient data, we strategically retrieve information (posts or model parameters) from other servers to reconstruct larger conversations and improve results. With this, we show that we can attain a macro-F1 of 0.8826. Our approach has considerable scope to improve moderation in decentralised and interoperable social networks such as Pleroma or Mastodon.

There have been numerous recent attempts to “decentralize” social media platforms, loosely referred to as Web3. Such ideas, often underpinned by blockchain solutions, offer decentralized equivalents of well-known services (e.g., forums, social networks, video sharing sites, microblogs). One particularly challenging function to implement in such a design is content moderation, due to the lack of central control. Consequently, they often rely on user-controlled moderation, whereby each user must create their own personal block list to filter out content they do not wish to see. This paper presents a first study of user-controlled moderation on one exemplar Web3 social microblogging platform called memo.cash. Based on a dataset covering 391K posts, we study the factors that lead users to “mute” each other. We find that the most crucial factor is the platform action count, rather than the presence of things like hate speech. We also show that the followership network plays a pivotal role in determining their visibility on the platform, further influencing their muting behavior. This leads us to design tooling to automate the muting process on a per-user basis. We model this as a recommendation problem, and experiment with a number of state-of-the-art recommender engines. We show that our system can generate effective personalized mute lists for users.

As an alternative to Twitter and other centralized social networks, the Fediverse is growing in popularity. The recent, and polemical, takeover of Twitter by Elon Musk has exacerbated this trend. The Fediverse includes a growing number of decentralized social networks, such as Pleroma or Mastodon, that share the same subscription protocol (ActivityPub). Each of these decentralized social networks is composed of independent instances that are run by different administrators. Users, however, can interact with other users across the Fediverse regardless of the instance they are signed up to. The growing user base of the Fediverse creates key challenges for the administrators, who may experience a growing burden. In this paper, we explore how large that overhead is, and whether there are solutions to alleviate the burden. We study the overhead of moderation on the administrators. We observe a diversity of administrator strategies, with evidence that administrators on larger instances struggle to find sufficient resources. We then propose a tool, WatchGen, to semi-automate the process.

There has been growing interest in the so-called “Web3” movement. This loosely refers to a mix of decentralized technologies, often underpinned by blockchain technologies. Among these, Web3 social media platforms have begun to emerge. These store all social interaction data (e.g., posts) on a public ledger, removing the need for centralized data ownership and management. But this comes at a cost, which some argue is prohibitively expensive. As an exemplar within this growing ecosytem, we explore memo.cash, a microblogging service built on the Bitcoin Cash (BCH) blockchain. We gather data for 24K users, 317K posts, 2.57M user actions, which have facilitated $6.75M worth of transactions. A particularly unique feature is that users must pay BCH tokens for each interaction (e.g., posting, following). We study how this may impact the social makeup of the platform. We therefore study memo.cash as both a social network and a transaction platform.

WhatsApp is a popular messaging app used by over a billion users around the globe. Due to this popularity, understanding misbehavior on WhatsApp is an important issue. The sending of unwanted junk messages by unknown contacts via WhatsApp remains understudied by researchers, in part because of the end-to-end encryption offered by the platform. We address this gap by studying junk messaging on a multilingual dataset of 2.6M messages sent to 5K public WhatsApp groups in India. We characterise both junk content and senders. We find that nearly 1 in 10 messages is unwanted content sent by junk senders, and a number of unique strategies are employed to reflect challenges faced on WhatsApp, e.g., the need to change phone numbers regularly. We finally experiment with on-device classification to automate the detection of junk, whilst respecting end-to-end encryption.

The era of live-broadcast is back but with two major changes. First, unlike traditional TV broadcasts, content is now streamed over the Internet enabling it to reach a wider audience. Second, due to various user-generated content platforms it has become possible for anyone to get involved, streaming their own content to the world. This emerging trend of going live usually happens via social platforms, where users perform live social broadcasts predominantly from their mobile devices, allowing their friends (and the general public) to engage with the stream in real-time. With the growing popularity of such platforms, the burden on the current Internet infrastructure is therefore expected to multiply. With this in mind, we explore one such prominent platform - Facebook Live. We gather 3TB of data, representing one month of global activity and explore the characteristics of live social broadcast. From this, we derive simple yet effective principles which can decrease the network burden. We then dissect global and hyper-local properties of the video while on-air, by capturing the geography of the broadcasters or the users who produce the video and the viewers or the users who interact with it. Finally, we study the social engagement while the video is live and distinguish the key aspects when the same video goes on-demand. A common theme throughout the paper is that, despite its name, many attributes of Facebook Live deviate from both the concepts of live and broadcast.