Hierarchical Multitask Learning with Dependency Parsing for Japanese Semantic Role Labeling Improves Performance of Argument Identification

TL;DR

This paper introduces a hierarchical multitask learning approach with dependency parsing to improve Japanese semantic role labeling, achieving state-of-the-art results especially in argument identification.

Contribution

The study presents a novel hierarchical multitask learning model incorporating dependency parsing for Japanese SRL, focusing on deep case analysis and joint argument identification and classification.

Findings

Multitasking with dependency parsing enhances argument identification accuracy.

The proposed model achieves state-of-the-art results in Japanese SRL.

Joint argument identification and classification improve overall performance.

Abstract

With the advent of FrameNet and PropBank, many semantic role labeling (SRL) systems have been proposed in English. Although research on Japanese predicate argument structure analysis (PASA) has been conducted, most studies focused on surface cases. There are only few previous works on Japanese SRL for deep cases, and their models' accuracies are low. Therefore, we propose a hierarchical multitask learning method with dependency parsing (DP) and show that our model achieves state-of-the-art results in Japanese SRL. Also, we conduct experiments with a joint model that performs both argument identification and argument classification simultaneously. The result suggests that multitasking with DP is mainly effective for argument identification.