A Labeled Graph Kernel for Relationship Extraction
This work addresses relationship extraction for bioinformatics, but it is incremental as it builds on existing kernel methods with specific improvements.
The paper tackles relationship extraction by proposing a labeled graph kernel based on random walk kernels, leveraging words between entities and combining information sources, and reports results comparable to state-of-the-art kernel methods on a protein-protein interaction dataset, with outperformance when combined with other kernels.
In this paper, we propose an approach for Relationship Extraction (RE) based on labeled graph kernels. The kernel we propose is a particularization of a random walk kernel that exploits two properties previously studied in the RE literature: (i) the words between the candidate entities or connecting them in a syntactic representation are particularly likely to carry information regarding the relationship; and (ii) combining information from distinct sources in a kernel may help the RE system make better decisions. We performed experiments on a dataset of protein-protein interactions and the results show that our approach obtains effectiveness values that are comparable with the state-of-the art kernel methods. Moreover, our approach is able to outperform the state-of-the-art kernels when combined with other kernel methods.