Mining Dynamic Document Spaces with Massively Parallel Embedded Processors

Jacobs, Jan W.M. and Dai, Rui and Smit, Gerard J.M. (2006) Mining Dynamic Document Spaces with Massively Parallel Embedded Processors. In: 6th International Workshop on Embedded Computer Systems: Architectures, Modeling, and Simulation, SAMOS 2006, 17-20 July 2006, Samos, Greece.

Preview

PDF
209Kb

Abstract:	Currently Océ investigates future document management services. One of these services is accessing dynamic document spaces, i.e. improving the access to document spaces which are frequently updated (like newsgroups). This process is rather computational intensive. This paper describes the research conducted on software development for massively parallel processors. A prototype has been built which processes streams of information from specified newsgroups and transforms them into personal information maps. Although this technology does speed up the training part compared to a general purpose processor implementation, however, its real benefits emerges with larger problem dimensions because of the scalable approach. It is recommended to improve on quality of the map as well as on visualisation and to better profile the performance of the other parts of the pipeline, i.e. feature extraction and visualisation.
Item Type:	Conference or Workshop Item
Copyright:	© 2006 Springer
Faculty:	Electrical Engineering, Mathematics and Computer Science (EEMCS)
Research Group:	Computer Architecture for Embedded Systems (CAES)
Link to this item:	http://purl.utwente.nl/publications/66033
Official URL:	http://dx.doi.org/10.1007/11796435_9
Export this item as:	BibTeX EndNote HTML Citation Reference Manager

Show download statistics for this publication

Repository Staff Only: item control page

Metis ID: 238059