Predicting and comparing algorithm performance on graph instances is challenging for multiple reasons. First, there usually no standard set of to benchmark performance. Second, using existing generators results in a restricted spectrum difficulty the resulting graphs are not diverse enough draw sound conclusions. That why recent work proposes new methodology generate by an evolutionary algorith...

The relative difficulty with which spoken corpora can be compiled by the researcher compared with written discourses, coupled with the time needed to fully transcribe spoken data, to say nothing of the additional expenses involved, inevitably has made large spoken corpora a far rarer entity than written corpora. And yet, if we are to further unravel the intricacies of spoken discourse, we need ...

In this paper, we make a qualitative and quantitative analysis of discourse relations within the LUNA conversational spoken dialog corpus. In particular, we describe the adaptation of the Penn Discourse Treebank (PDTB) annotation scheme to the LUNA dialogs. We discuss similarities and differences between our approach and the PDTB paradigm and point out the peculiarities of spontaneous dialogs w...

In this article, we present and evaluate a method for training a statistical partof-speech tagger on data from written language and then adapting it to the requirements of tagging a corpus of transcribed spoken language, in our case spoken Swedish. This is currently a significant problem for many research groups working with spoken language, since the availability of tagged training data from s...

Like the sequence of words in written language, comic book page layouts direct images into a deliberate reading sequence. Conventional wisdom would expect that comic panels follow the order of text: left-to-right and down - a "Z-path" - though several layouts can violate this order, such as Gestalt groupings of panels that deny a Z-path of reading. To examine how layouts pressure readers to cho...