Entries in microblogging sites such as Twitter are very short: a “tweet ”can contain at most 140 characters. Given a user query, retrieving relevant tweets is particularly challenging since their extreme brevity exacerbates the well-known vocabulary mismatch problem. In this preliminary study, we explore standard query expansion approaches as a way to address this problem. Since the tweets are ...

Urban legends are a genre of modern folklore, consisting of stories about rare and exceptional events, just plausible enough to be believed, which tend to propagate inexorably across communities. In our view, while urban legends represent a form of “sticky” deceptive text, they are marked by a tension between the credible and incredible. They should be credible like a news article and incredibl...

We examined the content of 2 talk-show TV programs in order to better understand the challenges posed by this program genre to automatic transcription. Six talk-show episodes were first segmented, transcribed and annotated by experts. Most of the speech content was found in conversational style with a significant portion of overlapped speech, about 18%. Then, automatic speech recognition experi...

Science reporting in the media often involves contested issues, such as, e.g., biotechnology, climate change, and more recently, geoengineering. The reporter’s framing of the issue is likely to influence readers’ perception of it. The notion of framing is related to how individuals and groups perceive and communicate about the world. Framing is typically studied by means of content analysis, fo...

We describe the results of large scale perception experiments showing improvements in synthesising two distinct kinds of prominence: standard pitch-accent and strong emphatic accents. Previously prominence assignment has been mainly evaluated by computing accuracy on a prominence-labelled test set. By contrast we integrated an automatic pitch-accent classifier into the unit selection target cos...

In this work we propose certain novel measures to understand non-projectivity in various syntactic phenomena in Hindi. This is an attempt to go beyond the analysis of non-projectivity in terms of certain graphical measures such as edge degree, planarity etc. Our measures are motivated by the findings in the processing literature that have investigated the interaction between working-memory cons...

We focus on automatic coreference resolution for blogs and news articles with user comments as part of a project on opinion mining. We aim to study the effect of the genre shift from edited structured newspaper text to unedited, unstructured blog data. We compare our coreference resolution system on three data sets: newspaper articles, mixed newspaper articles and reader comments, and blog data...

When machine translation researchers participate in evaluation tasks, they typically design their primary submissions using ideas that are not genre-specific. In fact, their systems look much the same from one evaluation campaign to another. In this paper, we analyze two popular genres: spoken language and written news, using publicly available corpora which stem from the popular WMT and IWSLT ...