Lahar is a warehousing system for Markovian streams—a common class of uncertain data streams produced via inference on probabilistic models. Example Markovian streams include text inferred from speech, location streams inferred from GPS or RFID readings, and human activity streams inferred from sensor data. Lahar supports OLAP-style queries on Markovian stream archives by leveraging novel appro...

We describe an experiment investigating how tagging affects human memory. We compare a low-cost tagging by mouse-clicking interface (Click2Tag) to traditional tagging by typing and to a baseline, no tagging condition. Our results suggest that tagging is beneficial for memory. However, the two tagging methods impact distinct aspects of memory: tagging by typing promotes elaboration of read conte...

This paper discusses the creation of typographic systems as artworks in three dimensional online builder's worlds, also known as the metaverse. An installation based upon text will be presented as an exemplar that will attempt to delineate the playful approach that has been taken whilst constructing a typographic art ecology that can be traversed and experienced through the agency of avatars. T...

Music can be used both to direct and enhance the impact a story can have on its listeners. This work makes use of two creative systems to provide emotionally-targeted musical accompaniment for stories. One system assigns emotional labels to text, and the other generates original musical compositions with targeted emotional content. We use these two programs to generate music to accompany audio ...

The biologists searching for environmental understanding in geographical scale are gradually shifting to the use of remote sensing and geoprocessing techniques to better study the Earth. However, the rapid obsolescence of remote sensing teaching material, requires new methods and new languages to better achieve the coming students. Although theoretical material is absolutely necessary, there mu...

Human readings of structured documents exhibit a much richer intonation than that observed in read isolated sentences. It is a challenge to capture this richness in an automatic way using datadriven techniques. In this paper, we extend our previous research on intonation modelling for isolated sentences in different respects: (i) the RNN (Recurrent Neural Network) intonation model is now traine...

Data-centered approaches to multimedia presentation design and implementation can be developed by extending methodologies and technologies common in text-based applications. A multimedia report is a multimedia presentation built on a set of data returned by one or more queries to multimedia repositories, integrated according to a template with appropriate spatial layout and temporal synchroniza...

This working paper presents preliminary ideas of five possible text-based notations for representing safety cases, which may be easier for non-graphically inclined people to use and understand than the currently popular graphics-based representations.