A Scalable Neural Shortlisting-Reranking Approach for Large-Scale Domain Classification in Natural Language Understanding
This addresses the scalability challenge in domain classification for real-world applications like IPDAs, but it is incremental as it builds on existing shortlisting-reranking methods.
The paper tackles the problem of efficiently selecting the best domain from thousands of overlapping ones for natural language understanding in intelligent personal digital assistants, achieving effective results through a neural shortlisting-reranking approach tested on 1,500 domains.
Intelligent personal digital assistants (IPDAs), a popular real-life application with spoken language understanding capabilities, can cover potentially thousands of overlapping domains for natural language understanding, and the task of finding the best domain to handle an utterance becomes a challenging problem on a large scale. In this paper, we propose a set of efficient and scalable neural shortlisting-reranking models for large-scale domain classification in IPDAs. The shortlisting stage focuses on efficiently trimming all domains down to a list of k-best candidate domains, and the reranking stage performs a list-wise reranking of the initial k-best domains with additional contextual information. We show the effectiveness of our approach with extensive experiments on 1,500 IPDA domains.