Our website is currently in the process of a much needed update.
Check back soon for updated content.
The Social Intelligence Lab
at Bar-Ilan University
Social Information Technologies for a Better World
Research
SocMed - The Social Medicine Initiative
Our approach to the use of social networks for emergency response is based on a new conceptualization of ‘social medicine’ (SM). We have all heard, or asked, some form of the question: “I have a headache – do you have an aspirin?” We’ve asked colleagues, flight attendants, teachers, cab drivers – and have offered the same when asked. The archetypal question being asked and answered in our conceptualization of social medicine is “Do you have an aspirin?” where “aspirin” is replaced with a critical life-saving treatment and “you” is replaced with millions of people, any one of whom may be carrying such a treatment nearby at any given time. As the prevalence of chronic conditions continues to rise, by creating Emergency Response Communities (ERC's) around chronic conditions such as acute allergies, asthma, diabetes, or heart disease, we have the ability to reshape society’s response to individuals in medical distress – by reshaping the response of individuals in our society. Our most recent extension of the ERC concept is now being studied as an opioid overdose intervention for the emergency provision of naloxone.
This international initiative brings together researchers in information and communications technologies, social networking, and crowd behavior, with leading medical specialists from the fields of Emergency Medicine, Emergency Medical Response, and Public Health, alongside medical domain experts from a range of chronic diseases including Allergy and Asthma, Addiction, Diabetes, and COPD.
Social CRM
Over the past two decades, social media have become an integral part of our society. Social media in general, and social networks in particular, enables global communication, knowledge sharing, and message exchange, regardless of location and time. The immediacy of social media presents a new reality to companies that face consumers who expect immediate interaction, change their opinions rapidly, and make prompt decisions. To keep pace in this era of social networking, many companies adopt social media as a leading channel to communicate with their customers, and communicate information to their customers. In fact, research shows that social media, and in particular Twitter, is often used by companies as a part of their Customer Relations Management (CRM) efforts, denoted Social CRM, or SCRM. This correlates well with consumers preferences and expectations: reports show that 30% of social media users prefer customer service via social media over traditional customer service call centers, people under 35 spend nearly four hours a day on social media, and most of that time is being spent engaging with brands. Moreover, about 37% of the people have used Twitter for customer service.
Our work in this regards focuses on companies' dynamics on Twitter: the growth and decay rates of followers, the desicion used when folloing/unfollowing companies, and the resulted degree distribution of existing, new and departing followers (spoiler: different mixtures of power law and exponential distributions). We further develop rate equations model to capture and simulate companies' dynamics. Qualitatively, we analysed over 70 S&P companies, and identified a 2-stage growth model of the use of Twitter as SCRM: branding, were the focus is on customer acquisition up to a certain volume, and social care, were the focus is on retaining customers and increase engagement. Evident from the data, we argue that companies have very little control over their growth stage, and in fact cannot moderate the number of followers, aiming to increase engagement.
SocSec - The Social Security Project
Information leakage is a complex problem. It occurs when a member of a social network posts seemingly inconspicuous information which when correlated or cross-referenced with other such it information leads to the exposure of information that was intended to remain secret or unexposed. This presents a two-fold challenge. First, since the specific content released does not contain confidential information it is difficult for content-based detection techniques to identify; Second, since the individual releasing the information would not inherently be viewed as a threat, the topology of their social network sub-map may not behave in ways that mimic classic information diffusion patterns of a leak.
UIL Behavior: We study the interplay between online news articles, reader comments, and social networks, to detect and characterize a new form of unintentional information leakage (UIL) - the accidental disclosure of confidential information not intended for public release. Organizations including the military and the courts use censorship to enhance security, with non-identification of individuals seen as necessary protection for military personnel, witnesses, minors, victims or suspects requiring anonymity We are investigating both qualitative methods for recognition and characterization of UIL as well as by a quantitative methods that automatically detects UIL comments. Our work extends across multiple social media and networking platforms including Facebook and Twitter.
Network Topology Anomalies: We examine the topological impact of leaked information on the social networks. That is, we examine whether such information is likely to increase the number of new edges in the network (followers), and/ or increase use of already existing edges (data sharing). Current state of the art focuses on content analysis for this research problem. We propose a two phase approach to tackle this question. First, we model the baseline topology dynamics in terms of followers (increase, decrease), and information diffusion. We further develop a data-driven simulation that generates multiple data scenarios. We use these scenarios as alternative realizations of the data that follows the expected baseline network dynamics. In the second step, we analyze anomalous behavior in the data. In other words, we pose the questions of (1) what is anomalous topology (theoretical definition), and (2) does leaked information cause anomalous behavior in the network (data-driven question).
This research is funded by Israel Ministry of Science and Technology research grant 3-9770 “Data Leakage in Social Networks: Detection and Prevention”.
Social interaction in online auctions: user models to predict actor behaviors
Wholesale and retail transactions generally involve the setting of a price by a seller and a purchase decision by a buyer, or at most some interaction in the form of a negotiation process to set price. In an auction, however, each bidder exists within a social environment that includes both the seller (and possibly multiple sellers) and every other potential bidder or buyer. Therefore the behavior of a bidder must take into account not only their own price-value models but of necessity the expected behavior of the other bidders (and sellers), as well as the number of other bidders and possible relationships between them.
In web contexts, there is a need for methods that analyze user behaviors, build user models and attempt to predict decisions and actions. Indeed, such methods are known in the art, yet they exhibit weaknesses which confine their practical application. Specifically, existing methods trade off behavior prediction accuracy and computation complexity. To address this problem, this research aims to devise an accurate yet efficient user model. By identifying user's significant characteristics, which will in turn support a model that provides for high-quality prediction of users' preferences and needs at a low computational complexity, this research will introduce a significant divergence from contemporary methods. Unlike existing user models, our models are implicitly inferred from users' behavior and not from information surveyed or provided directly, and encompass characteristics that influence users' decision processes.