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.

[1]  W..  Relative Neighborhood Graphs and Their Relatives , 2004 .

[2]  Godfried T. Toussaint,et al.  Relative neighborhood graphs and their relatives , 1992, Proc. IEEE.

[3]  Daniel P. Lopresti,et al.  A fast technique for comparing graph representations with applications to performance evaluation , 2003, Document Analysis and Recognition.

[4]  Jean-Yves Ramel,et al.  AGORA: the interactive document image analysis tool of the BVH project , 2006, Second International Conference on Document Image Analysis for Libraries (DIAL'06).

[5]  Sébastien Adam,et al.  Clustering document images using a bag of symbols representation , 2005, Eighth International Conference on Document Analysis and Recognition (ICDAR'05).