Intent Features for Rich Natural Language Understanding

TL;DR

This paper introduces intent features and a new neural network architecture to improve natural language understanding in dialog systems by leveraging shared, domain-agnostic intent properties learned from syntactic cues.

Contribution

The paper proposes intent features and the Global-Local neural network model, enabling better sharing and understanding of intents across different domains and contexts.

Findings

Significant improvement over baselines in intent feature identification

Effective in multi-intent natural language understanding modules

Applicable to classification tasks using full context

Abstract

Complex natural language understanding modules in dialog systems have a richer understanding of user utterances, and thus are critical in providing a better user experience. However, these models are often created from scratch, for specific clients and use cases, and require the annotation of large datasets. This encourages the sharing of annotated data across multiple clients. To facilitate this we introduce the idea of intent features: domain and topic agnostic properties of intents that can be learned from the syntactic cues only, and hence can be shared. We introduce a new neural network architecture, the Global-Local model, that shows significant improvement over strong baselines for identifying these features in a deployed, multi-intent natural language understanding module, and, more generally, in a classification setting where a part of an utterance has to be classified utilizing the whole context.