ExGED

ExGED

Graph Data Repository For Extended Graph Edit Distance

RFAI - LIFAT - Tours / France

Repository description

This dataset is proposed to evaluate the accuracy of Extended or Classical Graph Edit Distance (GED) methods,
This data set is origined from SESYD data set, link for SESYD: http://mathieu.delalandre.free.fr/projects/sesyd/
From the models in SESYD, the Vectograph is used to generate the distorted symbols.
The added noises can rotate, rescale or deform the original symbol.
The graph representation for each symbol can be referred to [1].

[1] Qureshi, R.J., Ramel, J.Y., Cardot, H., Mukherji, P.: Combination of symbolic andstatistical features for symbols recognition. In: 2007 International Conference onSignal Processing, Communications and Networking. pp. 477–482. IEEE (2007).

Download

Download ZIP FILE

Contacts

Technical questions about the database, the format, or problems obtaining the data should be directed to the database editors:

Diem Kieu HO, "kieu.ho at univ-tours.fr"
Nicolas Monmarché nicolas.monmarche at univ-tours.fr
Jean-Yves Ramel, "ramel at univ-tours.fr"

Lab. LIFAT - PolytechTours
64, Av. Jean Portalis
37200 TOURS - FRANCE
Tel: +33 2.47.36.14.26