Self-Organising Maps are utilised in many data mining and knowledge management applications. Although various visualisations have been proposed for SOM, these techniques lack in distinguishing between the items mapped to the same unit. Here we present a novel technique for the visualisation of Self-Organising Maps that displays inputs not in the centre of the map units, but shifts them towards the closest neighbours, the degree of the movement depending on the similarity to the neighbours. The night-sky visualisation facilitates better understanding of the underlying data. We report results from applying our method on two synthetic and a real-life data set.
Information Realisation is the process of presenting data as Textual, Graphical or Audial information to a human user. In this paper, we discuss the importance of this concept with respect to the accessibility of Semantic Web data to a diverse target audience. We provide an ontological point of view, defining the expressive characteristics and application domain of representation formats, thus presenting a system which produces representations customised to the user environment and the nature of the source data. Our approach considers the semantics of the data, not just the structure, and aims to present the information in the most semantically appropriate manner for the given target environment. We provide examples of a simple data set being realised as popular target representation formats: textual (XHTML, RSS); graphical (SVG, X3D); and audial (SoundML, VoiceXML).
This paper presents an applocation called Visual Browser that is able to visualize RDF data. It explains the advantage of the RDF/RDFS as well as the way RDF triples are displayed in the visualisation process. One of the most importand features of the programm comprises a two level visualization – the data and the so-called perspective of view. The scope of this architecture is shown on examples. Two very different domains are visualized: WordNet and a Universal information Robot’s knowldege base.
The successful and efficient querying of information in electronic information pools is becoming increasingly important in today’s information society. At the same time the quantity of existing information is continually growing. Querying scientific literature and selecting relevant hits are typical examples for this. [Börner and Chen 02] present possibilities to display structured information pools visually in order to simplify querying and selection processes. However, current interaction possibilities are limited to the manipulation of hit images (cutting out, zooming, rotating). The structure features of the underlying information pool remain unconsidered. Our JADE interface uses this structure information additionally to support the refinement task as well as the navigation within the space of query hits. It is based on mathematical procedures known as formal concept analysis. We carried out an evaluation study in order to determine the efficiency of the interactive visualisation techniques provided by JADE. Psychology students were made to perform various query tasks with a literature database. One group worked with a common web interface. The other group worked with JADE. The query results articulate a clear advantage for utilising interactive visualisation techniques in regard to the common parameters of query tasks, precision and recall.
This paper aims at analyzing by bibliometric methods the economic literature dealing with Innovation in Services, giving at the same time an overview of the main topics of discussion. To achieve this, the software toll BibTechMon was useed. Our analysis shows that in the last 15 years attention has been primarily focused on technological change in telecommunications, media and the software industry and its consequences for the market structure in these sectors. These results are supported by an additional cluster analysis.