Antonella Ianni

Detecting and Characterizing Eating-Disorder Communities on Social Media

Eating disorders are complex mental disorders and responsible for the highest mortality rate among mental illnesses. Recent studies reveal that user-generated content on social media provides useful information in understanding these disorders. Most previous studies focus on studying communities of people who discuss eating disorders on social media, while few studies have explored community structures and interactions among individuals who suffer from this disease over social media. In this paper, we first develop a snowball sampling method to automatically gather individuals who self-identify as eating disordered in their profile descriptions, as well as their social network connections with one another on Twitter. Then, we verify the effectiveness of our sampling method by: 1. quantifying differences between the sampled eating disordered users and two sets of reference data collected for non-disordered users in social status, behavioral patterns and psychometric properties; 2. building predictive models to classify eating disordered and non-disordered users. Finally, leveraging the data of social connections between eating disordered individuals on Twitter, we present the first homophily study among eating-disorder communities on social media. Our findings shed new light on how an eating-disorder community develops on social media.

Learning correlated equilibria in population games

The paper develops a framework for the analysis of finite n-player games, recurrently played by randomly drawn n-tuples of players, from a finite population. We first relate the set of equilibria of this game to the set of correlated equilibria of the underlying game, and then focus on learning processes modelled as Markovian adaptive dynamics. For the class of population games for which the underlying game has identical interests, we show that, independently of the matching technology, any myopic-best reply dynamics converges (in probability) to a correlated equilibrium. We also analyze noisy best reply dynamics, where players’ behaviour is perturbed by payoff-dependent mistakes, and explicitly characterize the limit distribution of the perturbed game in terms of the correlated equilibrium payoff of the underlying game.

Bayesian Social Learning with Local Interactions

We study social learning in a large population of agents who only observe the actions taken by their neighbours. Agents have to choose one, out of two, reversible actions, each optimal in one, out of two, unknown states of the world. Each agent chooses rationally, on the basis of private information and of the observation of his neighbours’ actions. Agents can repeatedly update their choices at revision opportunities that they receive in a random sequential order. We show that if agents receive equally informative signals and observe both neighbours, then actions converge exponentially fast to a configuration where some agents are permanently wrong. In contrast, if agents are unequally informed (in that some agents receive a perfectly informative signal and others are uninformed) and observe one neighbour only, then everyone will eventually choose the correct action. Convergence, however, obtains very slowly, at rate √t.

Competition And Cascades In Markets: An Agent‐Based Model Of Endogenous Mergers

We present an agent-based model of endogenous merger formation in a market with turnover of market participants. We describe the dynamics of the model and identify the conditions under which market competition is sufficiently disrupted to prompt extended periods during which mergers are desirable. We also demonstrate how merger waves can be triggered by industry shocks and firm overconfidence.

A simple locally interactive model of ergodic and nonergodic growth

In this paper we provide a simple locally interactive dynamic model of technology choice and output production. We assume a Cobb-Douglas type production function for two available technologies. The returns to technology 0 are not affected by local spillovers. Technology 1 is more costly, as there is an overhead cost, but it has a higher marginal productivity with respect to net capital. The superiority of technology 1 positively and monotonically depends on the fraction of neighbours using it. We study the aggregate process of technology choices in a model with countably many firms and repeated choices. The model explains: (i) persistent aggregate fluctuations in the presence of only idiosyncratic shocks, (ii) cross sectional heterogeneity along the dynamics and (iii) the possibility of multiple equilibria. The main contribution of the paper over the existing literature is that the model explains the endogeneous formation of large areas, homogeneous in terms of technology choice and output level, that look stationary along the dynamics.

Social interactions in online eating disorder communities: A network perspective

Online health communities facilitate communication among people with health problems. Most prior studies focus on examining characteristics of these communities in sharing content, while limited work has explored social interactions between communities with different stances on a health problem. Here, we analyse a large communication network of individuals affected by eating disorders on Twitter and explore how communities of individuals with different stances on the disease interact online. Based on a large set of tweets posted by individuals who self-identify with eating disorders online, we establish the existence of two communities: a large community reinforcing disordered eating behaviours and a second, smaller community supporting efforts to recover from the disease. We find that individuals tend to mainly interact with others within the same community, with limited interactions across communities and inter-community interactions characterized by more negative emotions than intra-community interactions. Moreover, by studying the associations between individuals’ behavioural characteristics and interpersonal connections in the communication network, we present the first large-scale investigation of social norms in online health communities, particularly on how a community approves of individuals’ behaviours. Our findings shed new light on how people form online health communities and can have broad clinical implications on disease prevention and online intervention.

Estimating determinants of attrition in online eating disordered communities: an instrumental variable approach

Background: The use of social media as key health-information source has increased steadily among people affected by eating disorders. Intensive research has examined characteristics of individuals engaging in online communities, while little is known about discontinuation of engagement and the phenomenon of participants dropping out of these communities. Objective: This study aims to investigate characteristics of dropout behaviors among eating disordered individuals on Twitter and to estimate the causal effects of personal emotions and social networks on dropout behaviors. Methods: Using a snowball sampling method, we collected a set of individuals who self-identified with eating disorders in their Twitter profile descriptions, as well as their tweets and social networks, leading to 241,243,043 tweets from 208,063 users. Individuals’ emotions are measured from their language use in tweets using an automatic sentiment analysis tool, and network centralities are measured from users’ following networks. Dropout statuses of users are observed in a follow-up period 1.5 years later (from Feb. 11, 2016 to Aug. 17, 2017). Linear and survival regression instrumental variables models are used to estimate the effects of emotions and network centrality on dropout behaviors. An individual’s attributes are instrumented with the attributes of the individual’s followees (i.e., people who are followed by the individual). Results: Eating disordered users have relatively short periods of activity on Twitter, with one half of our sample dropping out at 6 months after account creation. Active users show more negative emotions and higher network centralities than dropped-out users. Active users tend to connect to other active users, while dropped-out users tend to cluster together. Estimation results suggest that users’ emotions and network centralities have causal effects on their dropout behaviors on Twitter. More specifically, users with positive emotions are more likely to drop out and have shorter-lasting periods of activity online than users with negative emotions, while central users in a social network have longer-lasting participation than peripheral users. Findings on users’ tweeting interests further show that users who attempt to recover from eating disorders are more likely to drop out than those who promote eating disorders as a lifestyle choice. Conclusions: Presence in online communities is strongly determined by individual’s emotions and social networks, suggesting that studies analyzing and trying to draw condition and population characteristics through online health communities are likely to be biased. Future research needs to examine in more detail the links between individual characteristics and participation patterns if better understanding of the entire population is to be achieved. At the same time, such attrition dynamics need to be acknowledged and controlled for when designing online interventions so as to accurately capture their intended populations. (JMIR Preprints 06/05/2018:10942) DOI: https://doi.org/10.2196/preprints.10942 https://preprints.jmir.org/preprint/10942 [unpublished, non-peer-reviewed preprint]

The Origin of Money: An Agent-Based Model

The benefits of money as a medium of exchange are obvious, but the historical origin of money is less clear. An existing economic model of monetary search is reproduced as an agentbased simulation and an evolutionary algorithm is used to model social learning. This approach captures the way in which different equilibria can arise, including solutions in which one or two goods come to be used as money. In the case where monetary goods have identical properties, multiple equilibria can be reached with a dependence on the starting beliefs of agents. In our analysis we also consider the evolutionary dynamics that allow for a small chance of mutations in strategies. In some cases our findings show evolutionary paths by which use of particular monetary goods can collapse.

Competition and cascades in the financial markets

We present an agent-based model of endogenous merger formation in a market with turnover of market participants. We describe the dynamics of the model and identify the conditions under which market competition is sufficiently disrupted to prompt extended periods during which mergers are desirable. We also demonstrate how merger waves can be triggered by industry shocks and firm overconfidence.

Competition and cascades in the financial markets: An agent-based model of endogeneous mergers

We present an agent-based model of endogenous merger formation in a market with turnover of market participants. We describe the dynamics of the model and identify the conditions under which market competition is sufficiently disrupted to prompt extended periods during which mergers are desirable. We also demonstrate how merger waves can be triggered by industry shocks and firm overconfidence.