Combining Feature-based Kernel with Tree Kernel for Extracting Relations
Author
Term
10. term
Publication year
2012
Submitted on
2012-06-06
Pages
40
Abstract
This report proposes a composite kernel method of semantic rela- tion extraction between named entities within natural language docu- ments. Using two kernel methods, the benets from both are combined to gain an increase in performance compared to earlier approaches. 1) A linear kernel processing linguistic features, such as word-span, order- of-entities and word-type. 2) A tree kernel computing the similarity of a relation type with the shortest path-enclosed tree between a pair of candidate entities. Experiments are done using previous implemented methods, such as context sensitiveness and latent annotations to mea- sure their impact on the performance. Evaluating on a dataset for the relation extraction task at the Conference on Computational Natural Language Learning from 2004, the results obtained are on par with previous state-of-the-art approaches on the same dataset.
Documents