Abstract:Extracting semantic relationships between entities from text documents is challenging in information extraction and important for deep information processing and management. This paper proposes to use the convolution kernel over parse trees together with support vector machines to model syntactic structured information for relation extraction. Compared with linear kernels, tree kernels can effectively explore implicitly huge syntactic structured features embedded in a parse tree. Our study reveals that the syntactic structured features embedded in a parse tree are very effective in relation extraction and can be well captured by the convolution tree kernel. Evaluation on the ACE benchmark corpora shows that using the convolution tree kernel only can achieve comparable performance with previous best-reported feature-based methods. It also shows that our method significantly outperforms previous two dependency tree kernels for relation extraction. Moreover, this paper proposes a composite kernel for relation extraction by combining the convolution tree kernel with a simple linear kernel. Our study reveals that the composite kernel can effectively capture both flat and structured features without extensive feature engineering, and easily scale to include more features. Evaluation on the ACE benchmark corpora shows that the composite kernel outperforms previous best-reported methods in relation extraction.

Kernel-based semantic relation detection and classification via enriched parse tree structure

Tree Kernel-Based Relation Extraction with Context-Sensitive Structured Parse Tree Information

Exploring syntactic structured features over parse trees for relation extraction using kernel methods

Exploring syntactic features for relation extraction using a convolution tree kernel

Rescue Implicit and Long-tail Cases: Nearest Neighbor Relation Extraction

Study of Kernel-Based Methods for Chinese Relation Extraction

Discovering relations between named entities from a large raw corpus using tree similarity-based clustering

A Composite Kernel to Extract Relations Between Entities with Both Flat and Structured Features.

A Position Encoding Convolutional Neural Network Based on Dependency Tree for Relation Classification

A Customized Dependency Tree Kernel for Effective Sentiment Classification.

Exploiting Tree Kernels for High Performance Chemical Induced Disease Relation Extraction.

Two novel composite kernels for relation extraction

Tree Kernel-based Negation and Speculation Scope Detection with Structured Syntactic Parse Features.

Using a Hybrid Convolution Tree Kernel for Semantic Role Labeling.

Extracting relation information from text documents by exploring various types of knowledge

Embedding Syntactic Tree Structures into CNN Architecture for Relation Classification.

A Grammar-driven Convolution Tree Kernel for Semantic Role Classification.

Exploring various knowledge in relation extraction

Entity-Centric Fully Connected GCN for Relation Classification

Efficient semantic kernel-based text classification using matching pursuit KFDA

Hierarchical learning strategy in semantic relation extraction