We present a new algorithm for inferring the home locations of Twitter users at different granularities, such as city, state, or time zone, using the content of their tweets and their tweeting behavior. Unlike existing approaches, our algorithm uses an ensemble of statistical and heuristic classifiers to predict locations. We find that a hierarchical classification approach can improve prediction accuracy. Experimental evidence suggests that our algorithm works well in practice and outperforms the best existing algorithms for predicting the location of Twitter users.

When experiencing or describing a new place people will often compare it against other places that they already know. However, this human attention to the simultaneous similarities and differences between places is not reflected in the design of user interfaces of current place search technologies. In this demo, we present Frankenplace, an application for doing similarity-based place search that allows users to interactively find new places based on mixtures of features drawn from different places. The features of places are derived from a combination of authoritative data sources and unstructured observation data from social media, and organized into an extensible set of layers. We demonstrate the Frankenplace interface, which lets a user build a profile of a target place by selecting the most relevant of the properties shared by known places.

News articles are extremely time sensitive by nature. There is also intense competition among news items to propagate as widely as possible. Hence, the task of predicting the popularity of news items on the social web is both interesting and challenging. Prior research has dealt with predicting eventual online popularity based on early popularity. It is most desirable, however, to predict the popularity of items prior to their release, fostering the possibility of appropriate decision making to modify an article and the manner of its publication. In this paper, we construct a multi-dimensional feature space derived from properties of an article and evaluate the efficacy of these features to serve as predictors of online popularity. We examine both regression and classification algorithms and demonstrate that despite randomness in human behavior, it is possible to predict ranges of popularity on twitter with an overall 84% accuracy. Our study also serves to illustrate the differences between traditionally prominent sources and those immensely popular on the social web.

We present a computational framework for understanding the social aspects of emotions in Twitter conversations. Using unannotated data and semisupervised machine learning, we look at emotional transitions, emotional influences among the conversation partners, and patterns in the overall emotional exchanges. We find that conversational partners usually express the same emotion, which we name Emotion accommodation, but when they do not, one of the conversational partners tends to respond with a positive emotion. We also show that tweets containing sympathy, apology, and complaint are significant emotion influencers. We verify the emotion classification part of our framework by a human-annotated corpus.

Although people receive trusted, personalized recommendations and auxiliary social benefits when they ask questions of their friends, using a search engine is often a more effective way to find an answer. Attempts to integrate social and algorithmic search have thus far focused on bringing social content into algorithmic search results. However, more of the benefits of social search can be preserved by reversing this approach and bringing algorithmic content into natural question-based conversations. To do this successfully, it is necessary to adapt search engine interaction to a social context. In this paper, we present SearchBuddies, a system that responds to Facebook status message questions with algorithmic search results. Via a three-month deployment of the system to 122 social network users, we explore how people responded to search content in a highly social environment. Our experience deploying SearchBuddies shows that a socially embedded search engine can successfully provide users with unique and highly relevant information in a social context and can be integrated into conversations around an information need. The deployment also illuminates specific challenges of embedding a search engine in a social environment and provides guidance toward solutions.

People regularly face tasks that can be understood as navigation in information networks, where the goal is to find a path between two given nodes. In many such situations, the navigator only gets local access to the node currently under inspection and its immediate neighbors. This lack of global information about the network notwithstanding, humans tend to be good at finding short paths, despite the fact that real-world networks are typically very large. One potential reason for this could be that humans possess vast amounts of background knowledge about the world, which they leverage to make good guesses about possible solutions. In this paper we ask the question: Are human-like high-level reasoning skills really necessary for finding short paths? To answer this question, we design a number of navigation agents without such skills, which use only simple numerical features. We evaluate the agents on the task of navigating Wikipedia, a domain for which we also possess large-scale human navigation data. We observe that the agents find shorter paths than humans on average and therefore conclude that, perhaps surprisingly, no sophisticated background knowledge or high-level reasoning is required for navigating the complex Wikipedia network.

Micro-blog forums, such as Twitter, constitute a powerful medium today that people use to express their thoughts and intentions on a daily, and in many cases, hourly, basis. Extracting ‘Real-Time Intention’ (RTI) of a user from such short text updates is a huge opportunity towards web personalization and social net- working around dynamic user context. In this paper, we explore the novel problem of detecting and classifying RTIs from micro-blogs. We find that employing a heuristic based ensemble approach on a reduced dimension of the feature space, based on a wide spectrum of linguistic and statistical features of RTI expressions, achieves significant improvement in detect- ing RTIs compared to word-level features used in many social media classification tasks today. Our solution approach takes into account various salient characteristics of micro-blogs towards such classification – high dimensionality, sparseness of data, limited context, grammatical in-correctness, etc.

Social groups often exhibit a high degree of dynamism. Some groups thrive, while many others die over time. Modeling destructive dynamics and understanding whether/why/when a person will depart from a group can be important in a number of social domains. In this paper, we take the World of Warcraft game as an exemplar platform for studying destructive group dynamics. We build models to predict if and when an individual is going to quit his/her guild, and whether this quitting event will inflict substantial damage on the guild. Our predictors start from in-game census data and extract features from multiple perspectives such as individual-level, guild-level, game activity, and social interaction features. Our study shows that destructive group dynamics can often be predicted with modest to high accuracy, and feature diversity is critical to prediction performance.

Social media plays an important role in the process of political engagement, especially in societies where significant constraints over traditional media and participation still exist. Little is known about how social media use is related to these constraints. This study examines how citizens’ perceptions of government control predict social media use and how this use is related to offline participation in the context of a city-state, Singapore. Based on a national survey of 2000 respondents, we found that perceptions of control over traditional media and political activity increase content production on social media and that perceived control of the mass media motivates citizens to consume political content on social media. Interestingly, perceptions of government control over the Internet reduced rather than increased social media production. More importantly, we find that social media use is related to a greater likelihood of offline citizen participation, namely attendance of political rallies. The findings suggest that social media alters the balance of power in the dependency relationships that exist between the government, media organizations and citizens, creating new venues for online political discourse which in turn help promote real-world political participation.

In this paper we present several methods for collecting Web textual contents and filtering noisy data. We show that knowing which user publishes which contents can contribute to detecting noise. We begin by collecting data from two forums and from Twitter. For the forums, we extract the meaningful information from each discussion (texts of question and answers, IDs of users, date). For the Twitter dataset, we first detect tweets with very similar texts, which helps avoiding redundancy in further analysis. Also, this leads us to clusters of tweets that can be used in the same way as the forum discussions: they can be modeled by bipartite graphs. The analysis of nodes of the resulting graphs shows that network structure and content type (noisy or relevant) are not independent, so network studying can help in filtering noise.