I am a Phd student, I work in the field of shape matching and recognition.
I would like implement a method based on graph clustering. My method operate on 3D models, where vertexes represent graph nodes, and distances between vertexes are interpreted on edges.
So, I have been build the graph corresponding to a model (C++), and I'll tray to calculate the graph distance matrix to partition vertexes into clusters using the Grph library (I hope!).
Best,
Bil.
I want to use Graph to represent ontologies, and develop similarity measures based on concepts hierarchy.
Would you guide me to get started ?
--
You received this message because you are subscribed to the Google Groups "Grph: High Performance Graph Library for Java" group.
To unsubscribe from this group and stop receiving emails from it, send an email to grph-high-performance-gr...@googlegroups.com.
To post to this group, send an email to grph-high-performa...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
Luc Hogie
Phone: 06 80 91 40 71
Skype: luchogie
"Je suis heureux parce que c'est bon pour la santé" Voltaire
Hi Luc,
Thanks so much for this. I'm using it to scale up our prototype GPS-SNR mapping (really SLAM) algorithm that is based on Loopy Belief Propagation. I've been using Neo4j as the persistent graph store and Grph as the in memory buffer. It's quite a nice combination! I should say that my first try with JGraphT did not perform very well at all, but Grph made all the difference. Thanks again!
Andrew
PhD Candidate, UC Santa Barbara
--
You received this message because you are subscribed to the Google Groups "Grph: High Performance Graph Library for Java" group.
To unsubscribe from this group and stop receiving emails from it, send an email to grph-high-performance-gr...@googlegroups.com.
To post to this group, send an email to grph-high-performa...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.