A Vectorial Representation for the Indexation of Structural Informations
This article presents a vectorial representation of structured data to reduce the complexity of dissimilarity computations in an information retrieval context. This representation enables, via a computation of an adapted measure, to approximate the distance between structural representations in both context of distance between graphs and searching occurrences of subgraphs. Preliminary results show that the proposed representation offers comparable performance with those of the literature.
Keywordsgraph signature structured information indexation and retrieval
Unable to display preview. Download preview PDF.
- 2.Jaromczyk, J., Toussaint, G.: Relative neighborhood graphs and their relatives. Proceedings of the IEEE (1992)Google Scholar
- 3.Barbu, E., Heroux, P., Adam, S., Trupin, E.: Clustering document images using a bag of symbols representation, ICDAR, pp. 1216–1220 (2005)Google Scholar
- 4.Ramel, J.Y., Busson, S., Demonet, M.L.: AGORA:the Interactive Document Image Analysis Tool of the BVH Project, DIAL, pp. 145–155 (2006)Google Scholar