Revisiting the Design Issues of Local Models for Japanese Predicate-Argument Structure Analysis

TL;DR

This paper demonstrates that enhancing local models with modern feature embedding and neural network-based feature combination significantly improves Japanese PAS analysis, outperforming existing global models.

Contribution

It shows that sophisticated local models, when combined with recent embedding techniques and neural networks, can surpass global models in Japanese PAS analysis.

Findings

Enhanced local models outperform global models in F1 score

Neural network-based feature combination improves performance

Recent feature embeddings boost local model effectiveness

Abstract

The research trend in Japanese predicate-argument structure (PAS) analysis is shifting from pointwise prediction models with local features to global models designed to search for globally optimal solutions. However, the existing global models tend to employ only relatively simple local features; therefore, the overall performance gains are rather limited. The importance of designing a local model is demonstrated in this study by showing that the performance of a sophisticated local model can be considerably improved with recent feature embedding methods and a feature combination learning based on a neural network, outperforming the state-of-the-art global models in $F_1$ on a common benchmark dataset.