Diploma Thesis, TU-Vienna, 2006

The semantic web is an initiative caring about standards and technologies for the establishment of machine understandable information contents. It is about enabling "intelligent" infrastructure that would overcome most of todays communication barriers, especially within machine-machine communication patterns. Although various standards for meta information annotation are already available, implementors of semantic web middleware face various challenges that arise when dealing with heterogenous data or data at a web-wide scale.

 

The idea that I pursue within my thesis is to overcome barriers coming with the semantic web itself by utilization of bionic approaches for semantic web speci.c tasks. I aim to propose a space based and channel theoretic environment that handles and hides complexity arising from data distribution and heterogenity and thus provides simplified environmental conditions for bionic method operation. Bionic approaches like swarm intelligence or cellular automatas perfectly suite distributed scenarios in general as they rely on local information only and solve scaleability issues as they meet computational complexity with smart heuristics and parallelizeability. The phenomenons of emergence and self-organization play a central role as they enable bottum-up handling of vast web-wide data amounts, without need for central or global control instances.

 

This thesis introduces selected semantic web technologies, a framework for deriving nature inspired algorithms and applications of such algorithms for semantic web specific tasks. It outlines how semantic web environments, swarm intelligence and space-based computing complement each other.