![]() * - OpenBabel: The Open Babel Package, version 2.1.1. links to) the following projects, also licensed under GPL: A Quickstart in Frequent Structure Mining Can Make a Difference. includes code from) the following project, licensed under GPL: * LibFminer is licensed under the terms of the GNU General Public License (GPL, see LICENSE). Click here for a flash-animated version, indicating occurrences. Data: CPDB salmonella mutagenicity Euclidean embedding: Schulz et. (De)activating features are (red) green, (In)active instances (salmon) blue. * Co-occurrence-based 2D embedding of molecules and backbone refinement class features showing close to perfect separation of target classes along top left to bottom right. * feasible which previously might have been intractable. * confirms that the novel descriptors render large training sets * such as the dataset coverage and the class size * yields insight into important properties of the proposed descriptors, Further analysis using large-scale datasets * lower running times compared to ordinary (static) upper ![]() * adjustment), which is also confirmed in the experiments in * to an improved statistical constraint (dynamic upper bound * mining, a large part of the search space can be pruned due * to the complete set of trees but significantly better than * runs shows that their classification accuracy is similar * by >90 % and >30 % compared to complete tree mining and * experiments, the proposed method reduces feature set sizes * characteristic for open or closed fragment mining. * inter-feature entropy as opposed to occurrences, which is The method is able to optimize structural * The classes are defined in terms of fragments sharing a common * of fragments to reduce feature set size and running times. * frequency and significance constraints, using classes * efficiently mines tree-shaped subgraph descriptors under minimum * based on so-called backbone refinement classes. ![]() * We present a new approach to large-scale graph mining * : KDD conference proceedings are online. * : The paper has been selected for oral presentation at MLG 2009. * : The Backbone Refinement Class paper (co-authored by Christoph Helma and Stefan Kramer) has been accepted for the KDD 2009 conference on Data Mining and Knowledge Discovery (Jun 28 - in Paris) for a presentation at the conference and inclusion in the conference proceedings. * Contact details are located at the end of this page. The paper is entitled "Large Scale Graph Mining using Backbone Refinement Classes". * You may download the scientific documentation from. * The Fminer frontend application is available from. * This is the Fminer library, available at.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |