Abstract: Recent years have shown an increased interest in bringing the field of Graph Theory into Natural Language Processing, Vision, Speech Recognition and other areas of semantic computing. Recent research has shown that graph-based representations give rise to novel and efficient solutions in a variety of tasks, ranging from part of speech tagging and information extraction in NLP to image classification and face recognition in Vision to pitch accent classification in Speech Recognition. Graph structures have been studied extensively and using a graph theoretical perspective numerous developed theories and algorithms can be used. We intend to bring together researchers from diverse areas to provide a forum for discussing graph-based algorithms from different perspectives, discover commonalities and differences in approaches and exchange of ideas.
We intend to invite researchers from different areas mentioned above to have a panel discussion about NLP and how graph theoretical results are relevant to applications in semantic computing. This panel will provide a theoretical discussion of empirical results which is important for further development of graph-based approaches.
We invite submissions of papers on graph-based methods and algorithms applied to semantic computing problems.
Topics include, but are not limited to:
* Graph algorithms for text mining, data mining and image retrieval
* Random walk graph methods
* Spectral graph clustering
* Encoding semantic distances in graphs
* Ranking algorithms based on graphs
* Small world graphs
* Semi-supervised graph-based methods
* Statistical network analysis
* Dynamic graph representations
Submissions will consist of regular full papers of max. 8 pages in double-column IEEE format following the submission guidelines available on the ICSC2008 Web page. Workshop papers will be included in the IEEE conference proceedings. Please submit your contributions via EDAS.
Ralucca Gera, Naval Postgraduate School (rgera _AT_nps_DOT_edu)