User-contributed messages on social media sites such as Twitter have emerged as powerful, real-time means of information sharing on the Web. These short messages tend to reflect a variety of events in real time, making Twitter particularly well suited as a source of real-time event content. In this paper, we explore approaches for analyzing the stream of Twitter messages to distinguish between messages about real-world events and non-event messages. Our approach relies on a rich family of aggregate statistics of topically similar message clusters. Large-scale experiments over millions of Twitter messages show the effectiveness of our approach for surfacing real-world event content on Twitter. 1

Twitter has become exceedingly popular, with hundreds of millions of tweets being posted every day on a wide variety of topics. This has helped make real-time search applications possible with leading search engines routinely displaying relevant tweets in response to user queries. Recent research has shown that a considerable fraction of these tweets are about “events”, and the detection of novel events in the tweet-stream has attracted a lot of research interest. However, very little research has focused on properly displaying this real-time information about events. For instance, the leading search engines simply display all tweets matching the queries in reverse chronological order. In this paper we argue that for some highly structured and recurring events, such as sports, it is better to use more sophisticated techniques to summarize the relevant tweets. We formalize the problem of summarizing event-tweets and give a solution based on learning the underlying hidden state representation of the event via Hidden Markov Models. In addition, through extensive experiments on real-world data we show that our model significantly outperforms some intuitive and competitive baselines.

Twitter, Facebook, and other related systems that we call social awareness streams are rapidly changing the information and communication dynamics of our society. These systems, where hundreds of millions of users share short messages in real time, expose the aggregate interests and attention of global and local communities. In particular, emerging temporal trends in these systems, especially those related to a single geographic area, are a significant and revealing source of information for, and about, a local community. This study makes two essential contributions for interpreting emerging temporal trends in these information systems. First, based on a large dataset of Twitter messages from one geographic area, we develop a taxonomy of the trends present in the data. Second, we identify important dimensions according to which trends can be categorized, as well as the key distinguishing features of trends that can be derived from their associated messages. We quantitatively examine the computed features for different categories of trends, and establish that significant differences can be detected across categories. Our study advances the understanding of trends on Twitter and other social awareness streams, which will enable powerful applications and activities, including user-driven real-time information services for local communities.

Social media sites such as Twitter contain large amounts of user contributed messages for a wide variety of real-world events. While some of these “event messages ” might contain interesting and useful information (e.g., event time, location, participants, opinions), others might provide little value (e.g., using heavy slang, incomprehensible language) to people interested in learning about an event. Techniques for effective selection of quality event content may therefore help improve applications such as event browsing and search. In this paper, we explore approaches for finding representative messages among a set of Twitter messages that correspond to the same event, with the goal of identifying high quality, relevant messages that provide useful event information. We evaluate our approaches using a large-scale dataset of Twitter messages, and show that we can automatically select event messages that are both relevant and useful. 1

In recent years, various Web-based sharing and community services such as Flickr and YouTube have made a vast and rapidly growing amount of multimedia content available online. Uploaded by individual participants, content in these immense pools of content is accompanied by varied types of metadata, such as social network data or descriptive textual information. These collections present, at once, new challenges and exciting opportunities for multimedia research. This article presents an approach for “social multimedia” applications. The approach is based on the experience of building a number of successful applications that are based on mining multimedia content analysis in social multimedia context.

User-contributed Web data contains rich and diverse information about a variety of events in the physical world, such as shows, festivals, conferences and more. This information ranges from known event features (e.g., title, time, location) posted on event aggregation platforms (e.g., Last.fm events, EventBrite, Facebook events) to discussions and reactions related to events shared on different social media sites (e.g., Twitter, YouTube, Flickr). In this paper, we focus on the challenge of automatically identifying user-contributed content for events that are planned and, therefore, known in advance, across different social media sites. We mine event aggregation platforms to extract event features, which are often noisy or missing. We use these features to develop query formulation strategies for retrieving content associated with an event on different social media sites. Further, we explore ways in which event content identified on one social media site can be used to retrieve additional relevant event content on other social media sites. We apply our strategies to a large set of user-contributed events, and analyze their effectiveness in retrieving relevant event content from Twitter, YouTube, and Flickr.

People who are not professional storytellers usually have difficulty composing travel photos and videos from a mundane slideshow into a coherent and engaging story, even when it is about their own experiences. However, consider putting the same person in a conversation with a friend – suddenly the story comes alive. We present Raconteur 2, a system for conversational storytelling that encourages people to make coherent points, by instantiating large-scale story patterns and suggesting illustrative media. It performs natural language processing in real-time on a text chat between a storyteller and a viewer and recommends appropriate media items from a library. Each item is annotated with one or a few sentences in unrestricted English. A large commonsense knowledge base and a novel commonsense inference technique are used to identify story patterns such as problem and resolution or expectation violation. It uses a concept vector representation that goes beyond keyword matching or word co-occurrence based techniques. A small experiment shows that people find Raconteur’s interaction design engaging, and suggestions helpful for real-time storytelling. Author Keywords Storytelling, conversation, chat, life stories, story pattern,

We demonstrate a system for augmenting information about planned events with Twitter messages, using a set of automatic query building strategies. We present two alternative interfaces to our system, namely, a browser plug-in and a customizable Web interface. 1

This paper studies the problem of dynamic relationship and event discovery. A large body of previous work on relation extraction focuses on discovering predefined and static relationships between entities. In contrast, we aim to identify temporally defined (e.g., co-bursting) relationships that are not predefined by an existing schema, and we identify the underlying time constrained events that lead to these relationships. The key challenges in identifying such events include discovering and verifying dynamic connections among entities, and consolidating binary dynamic connections into events consisting of a set of entities that are connected at a given time period. We formalize this problem and introduce an efficient end-to-end pipeline as a solution. In particular, we introduce two formal notions, global temporal constraint cluster and local temporal constraint cluster, for detecting dynamic events. We further design efficient algorithms for discovering such events from a large graph of dynamic relationships. Finally, detailed experiments on real data show the effectiveness of our proposed solution.

We present a technical demonstration of an online city exploration application that helps users identify interesting spots in a city by use of photo clusters corresponding to landmarks and events. Our application, called ClustTour, is based on an efficient landmark and event detection scheme for tagged photo collections. The proposed scheme relies on the combination of a graph-based photo clustering algorithm, making use of both visual and tag information of photos, with a cluster classification and merging module. ClustTour creates a map-based visualization of the identified photo clusters that are classified in prominent categories and are filterable by time and tag. We believe that such an application can greatly facilitate the task of knowing a city through its landmarks and events. So far, the demo has been based on a large photo dataset focused on Barcelona, and it is gradually expanding to contain photo clusters of several major cities of Europe. Furthermore, an Android application is developed that complements the web-based version of ClustTour.