TAIA 2012 papers online

Four TAIA 2012 paper are available online now… we care about time-aware information access

First, “Sustainable questions” by Bart de Goede, Anne Schuth and Maarten de Rijke. Community question answering platforms have large repositories of previously answered questions. Reusing the answers for new questions is tempting. However, not all stored answers will still be relevant. We define a new and challenging problem concerning the sustainability of questions, and present metrics aimed at distinguishing between sustainable and non-sustainable questions. We find that an intuitive approach to sustainability of questions is not sufficient, but that simple properties can already distinguish between sustainable and non-sustainable questions.
PDF

Second,”Time-Aware Exploratory Search: Exploring Word Meaning through Time” by Daan Odijk, Guiseppe Santucci, Maarten de Rijke, Marco Angelini and Guido. Granato. With more longitudinal archives becoming digitized and publicly available, new uses emerge. Collections that span centuries call for a time-aware exploration approach, a coordinated environment supporting understanding the development of word usage and meaning through time, with the means to leverage this for exploration. We present ongoing work on a coordinated time-aware exploratory search approach and present a case study on a prototype system. With this approach, a user is able to gain a deeper understanding of the relevant parts of the collection.
PDF

Third, “OpenGeist: Insight in the Stream of Page Views on Wikipedia” by Hendrike Peetz, Edgar Meij and Maarten de Rijke. We present a RESTful interface that captures insights into the zeitgeist of Wikipedia users. The system is an interface for clustering and comparing concepts based on the time series of the number of views of their Wikipedia page. The functionality is motivated by three use cases, ranging from technical novices to expert users and we also provide two real-life example applications.
PDF

And finally, “Activity prediction” by Wouter Weerkamp and Maarten de Rijke. Social media platforms allow users to share their messages with everyone else. In microblogs, e.g., Twitter, people mostly report on what they did, they talk about current activities, and mention things they plan to do in the near future. In this paper, we propose the task of activity prediction, that is, trying to establish a set of activities that are likely to become popular at a later time. We perform a small-scale initial experiment, in which we try to predict popular activities for the coming evening using Dutch Twitter data. Our experiment shows the feasibility and challenges of the task, with a simple method resulting in human-readable activities. This exploration also identifies several issues (e.g., temporal phrases and activity classification) that need to be addressed in future work.
PDF