Mental-health therapy involves a complex conversation flow in which patients and therapists continuously negotiate what should be talked about next. For example, therapists might try to shift the conversation's direction to keep the therapeutic process on track and avoid stagnation, or patients might push the discussion towards issues they want to focus on. How do such patient and therapist redirections relate to the development and quality of their relationship? To answer this question, we introduce a probabilistic measure of the extent to which a certain utterance immediately redirects the flow of the conversation, accounting for both the intention and the actual realization of such a change. We apply this new measure to characterize the development of patienttherapist relationships over multiple sessions in a very large, widely-used online therapy platform. Our analysis reveals that (1) patient control of the conversation's direction generally increases relative to that of the therapist as their relationship progresses; and (2) patients who Figure 1: Top: Examples of attempted redirection, both have less control in the first few sessions are realized (left) and unrealized (right). Bottom: Example significantly more likely to eventually express where redirection is not attempted.

Throughout a conversation, the way participants interact with each other is in constant flux: their tones may change, they may resort to different strategies to convey their points, or they might alter their interaction patterns. An understanding of these dynamics can complement that of the actual facts and opinions discussed, offering a more holistic view of the trajectory of the conversation: how it arrived at its current state and where it is likely heading. In this work, we introduce the task of summarizing the dynamics of conversations, by constructing a dataset of human-written summaries, and exploring several automated baselines. We evaluate whether such summaries can capture the trajectory of conversations via an established downstream task: forecasting whether an ongoing conversation will eventually derail into toxic behavior. We show that they help both humans and automated systems with this forecasting task. Humans make predictions three times faster, and with greater confidence, when reading the summaries than when reading the transcripts. Furthermore, automated forecasting systems are more accurate when constructing, and then predicting based on, summaries of conversation dynamics, compared to directly predicting on the transcripts.

Incivility remains a major challenge for online discussion platforms, to such an extent that even conversations between well-intentioned users can often derail into uncivil behavior. Traditionally, platforms have relied on moderators to -- with or without algorithmic assistance -- take corrective actions such as removing comments or banning users. In this work we propose a complementary paradigm that directly empowers users by proactively enhancing their awareness about existing tension in the conversation they are engaging in and actively guides them as they are drafting their replies to avoid further escalation. As a proof of concept for this paradigm, we design an algorithmic tool that provides such proactive information directly to users, and conduct a user study in a popular discussion platform. Through a mixed methods approach combining surveys with a randomized controlled experiment, we uncover qualitative and quantitative insights regarding how the participants utilize and react to this information. Most participants report finding this proactive paradigm valuable, noting that it helps them to identify tension that they may have otherwise missed and prompts them to further reflect on their own replies and to revise them. These effects are corroborated by a comparison of how the participants draft their reply when our tool warns them that their conversation is at risk of derailing into uncivil behavior versus in a control condition where the tool is disabled. These preliminary findings highlight the potential of this user-centered paradigm and point to concrete directions for future implementations.

To address the widespread problem of uncivil behavior, many online discussion platforms employ human moderators to take action against objectionable content, such as removing it or placing sanctions on its authors. This reactive paradigm of taking action against already-posted antisocial content is currently the most common form of moderation, and has accordingly underpinned many recent efforts at introducing automation into the moderation process. Comparatively less work has been done to understand other moderation paradigms -- such as proactively discouraging the emergence of antisocial behavior rather than reacting to it -- and the role algorithmic support can play in these paradigms. In this work, we investigate such a proactive framework for moderation in a case study of a collaborative setting: Wikipedia Talk Pages. We employ a mixed methods approach, combining qualitative and design components for a holistic analysis. Through interviews with moderators, we find that despite a lack of technical and social support, moderators already engage in a number of proactive moderation behaviors, such as preemptively intervening in conversations to keep them on track. Further, we explore how automation could assist with this existing proactive moderation workflow by building a prototype tool, presenting it to moderators, and examining how the assistance it provides might fit into their workflow. The resulting feedback uncovers both strengths and drawbacks of the prototype tool and suggests concrete steps towards further developing such assisting technology so it can most effectively support moderators in their existing proactive moderation workflow.

Aided by technology, people are increasingly able to communicate across geographical, cultural, and language barriers. This ability also results in new challenges, as interlocutors need to adapt their communication approaches to increasingly diverse circumstances. In this work, we take the first steps towards automatically assisting people in adjusting their language to a specific communication circumstance. As a case study, we focus on facilitating the accurate transmission of pragmatic intentions and introduce a methodology for suggesting paraphrases that achieve the intended level of politeness under a given communication circumstance. We demonstrate the feasibility of this approach by evaluating our method in two realistic communication scenarios and show that it can reduce the potential for misalignment between the speaker's intentions and the listener's perceptions in both cases.

Online discussions often derail into toxic exchanges between participants. Recent efforts mostly focused on detecting antisocial behavior after the fact, by analyzing single comments in isolation. To provide more timely notice to human moderators, a system needs to preemptively detect that a conversation is heading towards derailment before it actually turns toxic. This means modeling derailment as an emerging property of a conversation rather than as an isolated utterance-level event. Forecasting emerging conversational properties, however, poses several inherent modeling challenges. First, since conversations are dynamic, a forecasting model needs to capture the flow of the discussion, rather than properties of individual comments. Second, real conversations have an unknown horizon: they can end or derail at any time; thus a practical forecasting model needs to assess the risk in an online fashion, as the conversation develops. In this work we introduce a conversational forecasting model that learns an unsupervised representation of conversational dynamics and exploits it to predict future derailment as the conversation develops. By applying this model to two new diverse datasets of online conversations with labels for antisocial events, we show that it outperforms state-of-the-art systems at forecasting derailment.

One of the main challenges online social systems face is the prevalence of antisocial behavior, such as harassment and personal attacks. In this work, we introduce the task of predicting from the very start of a conversation whether it will get out of hand. As opposed to detecting undesirable behavior after the fact, this task aims to enable early, actionable prediction at a time when the conversation might still be salvaged. To this end, we develop a framework for capturing pragmatic devices---such as politeness strategies and rhetorical prompts---used to start a conversation, and analyze their relation to its future trajectory. Applying this framework in a controlled setting, we demonstrate the feasibility of detecting early warning signs of antisocial behavior in online discussions.

Online health communities are a valuable source of information for patients and physicians. However, such user-generated resources are often plagued by inaccuracies and misinformation. In this work we propose a method for automatically establishing the credibility of user-generated medical statements and the trustworthiness of their authors by exploiting linguistic cues and distant supervision from expert sources. To this end we introduce a probabilistic graphical model that jointly learns user trustworthiness, statement credibility, and language objectivity. We apply this methodology to the task of extracting rare or unknown side-effects of medical drugs --- this being one of the problems where large scale non-expert data has the potential to complement expert medical knowledge. We show that our method can reliably extract side-effects and filter out false statements, while identifying trustworthy users that are likely to contribute valuable medical information.

User contributions in the form of posts, comments, and votes are essential to the success of online communities. However, allowing user participation also invites undesirable behavior such as trolling. In this paper, we characterize antisocial behavior in three large online discussion communities by analyzing users who were banned from these communities. We find that such users tend to concentrate their efforts in a small number of threads, are more likely to post irrelevantly, and are more successful at garnering responses from other users. Studying the evolution of these users from the moment they join a community up to when they get banned, we find that not only do they write worse than other users over time, but they also become increasingly less tolerated by the community. Further, we discover that antisocial behavior is exacerbated when community feedback is overly harsh. Our analysis also reveals distinct groups of users with different levels of antisocial behavior that can change over time. We use these insights to identify antisocial users early on, a task of high practical importance to community maintainers.

Discourse on social media platforms is often plagued by acute polarization, with different camps promoting different perspectives on the issue at hand—compare, for example, the differences in the liberal and conservative discourse on the U.S. immigration debate. A large body of research has studied this phenomenon by focusing on the affiliation of groups and individuals. We propose a new finer-grained perspective: studying the impartiality of individual messages. While the notion of message impartiality is quite intuitive, the lack of an objective definition and of a way to measure it directly has largely obstructed scientific examination. In this work we operationalize message impartiality in terms of how discernible the affiliation of its author is, and introduce a methodology for quantifying it automatically. Unlike a supervised machine learning approach, our method can be used in the context of emerging events where impartiality labels are not immediately available. Our framework enables us to study the effects of (im)partiality on social media discussions at scale. We show that this phenomenon is highly consequential, with partial messages being twice more likely to spread than impartial ones, even after controlling for author and topic. By taking this fine-grained approach to polarization, we also provide new insights into the temporal evolution of online discussions centered around major political and sporting events.