Domain Knowledge Empowered Structured Neural Net for End-to-End Event Temporal Relation Extraction

TL;DR

This paper introduces a novel framework that integrates probabilistic domain knowledge as distributional constraints into neural networks, significantly improving end-to-end event temporal relation extraction performance across multiple datasets.

Contribution

It presents a new method combining domain knowledge with neural models using Lagrangian Relaxation for constrained inference, addressing limitations of previous approaches.

Findings

Significant performance improvements on news and clinical datasets.

Effective incorporation of domain knowledge enhances neural network accuracy.

Framework outperforms baseline models with statistical significance.

Abstract

Extracting event temporal relations is a critical task for information extraction and plays an important role in natural language understanding. Prior systems leverage deep learning and pre-trained language models to improve the performance of the task. However, these systems often suffer from two short-comings: 1) when performing maximum a posteriori (MAP) inference based on neural models, previous systems only used structured knowledge that are assumed to be absolutely correct, i.e., hard constraints; 2) biased predictions on dominant temporal relations when training with a limited amount of data. To address these issues, we propose a framework that enhances deep neural network with distributional constraints constructed by probabilistic domain knowledge. We solve the constrained inference problem via Lagrangian Relaxation and apply it on end-to-end event temporal relation extraction tasks. Experimental results show our framework is able to improve the baseline neural network models with strong statistical significance on two widely used datasets in news and clinical domains.