Joint Slot Filling and Intent Detection via Capsule Neural Networks

TL;DR

This paper introduces a capsule neural network model for joint slot filling and intent detection in natural language understanding, explicitly modeling hierarchical relationships to improve performance.

Contribution

The paper proposes a novel capsule-based neural network with dynamic routing and re-routing schemas for joint slot filling and intent detection.

Findings

Outperforms existing models on real-world datasets

Effectively captures hierarchical semantic relationships

Improves slot filling accuracy through intent-aware re-routing

Abstract

Being able to recognize words as slots and detect the intent of an utterance has been a keen issue in natural language understanding. The existing works either treat slot filling and intent detection separately in a pipeline manner, or adopt joint models which sequentially label slots while summarizing the utterance-level intent without explicitly preserving the hierarchical relationship among words, slots, and intents. To exploit the semantic hierarchy for effective modeling, we propose a capsule-based neural network model which accomplishes slot filling and intent detection via a dynamic routing-by-agreement schema. A re-routing schema is proposed to further synergize the slot filling performance using the inferred intent representation. Experiments on two real-world datasets show the effectiveness of our model when compared with other alternative model architectures, as well as existing natural language understanding services.