Thumbnails:
List:
Year:
Category:
Session:
Poster:
Getting poster data...
Marjan Trutschl (Louisiana State University, Shreveport, United States of America)
Large amounts of high-dimensional biomedical data not only create the need for the analysis of the data and interpretation of results, but also the need for the development of tools and methods that can handle such data. Many techniques are graphical in nature with ability to represent a small number of variables at a time. Our novel neural network enhanced information visualization techniques enhance knowledge extraction and are targeted towards complex data and provide for a very small, if any, loss of information. The algorithms we have developed are based on a self-organizing map algorithm where the (dis)placement of records in a neural-network augmented plot is directed by all (or a subset of) dimensional values. This significantly improves the treatment of dimensional values. Users no longer have to choose which dimensions of their rich data sets to discard in order to visualize the data set in a classic low-dimensional visualization. We present techniques that combine well-understood classic visualizations and neural network algorithms, creating meaningful visual representations of high-dimensional biomedical data. Computational complexity of algorithms is addressed using mu