All-paths graph kernel for protein-protein interaction extraction with evaluation of cross-corpus learning
Open Access
- 19 November 2008
- journal article
- Published by Springer Nature in BMC Bioinformatics
- Vol. 9 (S11) , S2
- https://doi.org/10.1186/1471-2105-9-s11-s2
Abstract
Automated extraction of protein-protein interactions (PPI) is an important and widely studied task in biomedical text mining. We propose a graph kernel based approach for this task. In contrast to earlier approaches to PPI extraction, the introduced all-paths graph kernel has the capability to make use of full, general dependency graphs representing the sentence structure.Keywords
This publication has 22 references indexed in Scilit:
- Comparative analysis of five protein-protein interaction corporaBMC Bioinformatics, 2008
- OpenDMAP: An open source, ontology-driven concept analysis engine, with applications to capturing knowledge regarding protein transport, protein interactions and cell-type-specific gene expressionBMC Bioinformatics, 2008
- Corpus annotation for mining biomedical events from literatureBMC Bioinformatics, 2008
- Frontiers of biomedical text mining: current progressBriefings in Bioinformatics, 2007
- RelEx—Relation extraction using dependency parse treesBioinformatics, 2006
- Extracting Protein-Protein Interaction Information from Biomedical Text with SVMIEICE Transactions on Information and Systems, 2006
- Comparative experiments on learning information extractors for proteins and their interactionsArtificial Intelligence in Medicine, 2005
- Parsing Biomedical LiteraturePublished by Springer Nature ,2005
- On Graph Kernels: Hardness Results and Efficient AlternativesPublished by Springer Nature ,2003
- The use of the area under the ROC curve in the evaluation of machine learning algorithmsPattern Recognition, 1997