A Lexicalized Tree Kernel for Open Information Extraction

Ying Xu,Yusuke Miyao,Grzegorz Kondrak,Randy Goebel,Christoph Ringlstetter,Mi-Young Kim

doi:10.3115/v1/p15-2046

Abstract

In contrast with traditional relation extraction, which only considers a fixed set of relations, Open Information Extraction (Open IE) aims at extracting all types of relations from text. Because of data sparseness, Open IE systems typically ignore lexical information, and instead employ parse trees and Part-of-Speech (POS) tags. However, the same syntactic structure may correspond to different relations. In this paper, we propose to use a lexicalized tree kernel based on the word embeddings created by a neural network model. We show that the lexicalized tree kernel model surpasses the unlexicalized model. Experiments on three datasets indicate that our Open IE system performs better on the task of relation extraction than the stateof-the-art Open IE systems of Xu et al. (2013) and Mesquita et al. (2013).

Highlights

Relation Extraction (RE) is the task of recognizing relationships between entities mentioned in text
Open Information Extraction (Open IE) models that extract N-ary relations have been proposed, here we concentrate on binary relations
Previous work suggests that Open IE would benefit from lexical information because the same syntactic structure may correspond to different relations

Summary

Introduction

Relation Extraction (RE) is the task of recognizing relationships between entities mentioned in text. Open IE models that extract N-ary relations have been proposed, here we concentrate on binary relations. Most Open IE systems employ syntactic information such as parse trees and part of speech (POS) tags, but ignore lexical information. Previous work suggests that Open IE would benefit from lexical information because the same syntactic structure may correspond to different relations. We propose a lexicalized tree kernel model that combines both syntactic and lexical information. We are the first to apply word embeddings and to use lexicalized tree kernel models for Open IE. We examine alternative approaches for including lexical information, and find that excluding named entities from the lexical information results in an improved F-score

System Architecture

Relation Candidates

Lexicalized Tree Kernel

Tree Structure

Tree Kernels

Experiments

Findings

Conclusion