Discovering Semantic Relations by Using Networks Theory

Abstract This thesis work adresses a new multidisciplinary research area, which is about undestanding the relations between semantic web and network theory and applying the theories and computational models of the latter to the former. In particular, two research objectives have been reached: • converting ontologies expressed in semantic web standards into a bipartite graph where the two sets of nodes are respectively: Learn- ing Objects composing an E-learning course, and words represent- ing the topics of this course. • using network theory to improve semantic relations between the two sets of nodes composing this graph. This latter task has been made by implementing a new recommenda- tion algorithm called Network Based Inference method (NBI) proposed by Zhou et al. in PRE 76, 046115 (2007)[1]. In addition to this we have proposed some extensions of the NBI algorithm to improve the algorithm described in the literature. The research work has been done in the context of the BONy project, whose aim is to apply semantic technology to Collaborative Learning. v