Prompt-based Generative Approach towards Multi-Hierarchical Medical Dialogue State Tracking

TL;DR

This paper introduces a prompt-based generative method for multi-hierarchical medical dialogue state tracking, addressing challenges of complex entity representation and data scarcity, and demonstrates superior performance on a newly published Chinese medical dialogue dataset.

Contribution

It defines a novel multi-hierarchical state structure, creates and releases a Chinese medical dialogue dataset, and proposes a prompt-based generative approach that outperforms existing methods especially with limited data.

Findings

Outperforms other DST methods in experiments.

Effective in scenarios with limited data.

Introduces a new Chinese medical dialogue dataset.

Abstract

The medical dialogue system is a promising application that can provide great convenience for patients. The dialogue state tracking (DST) module in the medical dialogue system which interprets utterances into the machine-readable structure for downstream tasks is particularly challenging. Firstly, the states need to be able to represent compound entities such as symptoms with their body part or diseases with degrees of severity to provide enough information for decision support. Secondly, these named entities in the utterance might be discontinuous and scattered across sentences and speakers. These also make it difficult to annotate a large corpus which is essential for most methods. Therefore, we first define a multi-hierarchical state structure. We annotate and publish a medical dialogue dataset in Chinese. To the best of our knowledge, there are no publicly available ones before. Then we propose a Prompt-based Generative Approach which can generate slot values with multi-hierarchies incrementally using a top-down approach. A dialogue style prompt is also supplemented to utilize the large unlabeled dialogue corpus to alleviate the data scarcity problem. The experiments show that our approach outperforms other DST methods and is rather effective in the scenario with little data.