Towards Unsupervised Spoken Language Understanding: Exploiting Query Click Logs for Slot Filling

Gokhan Tur, Dilek Hakkani-Tur, Dustin Hillard, and Asli Celikyilmaz

Abstract

In this paper, we present a novel approach to exploit user queries mined from search engine query click logs to bootstrap or improve slot filling models for spoken language understanding. We propose extending the earlier gazetteer population techniques to mine unannotated training data for semantic parsing. The automatically annotated mined data can then be used to train slot specific parsing models. We show that this method can be used to bootstrap slot filling models and can be combined with any available annotated data to improve performance. Furthermore, this approach may eliminate the need for populating and maintaining in-domain gazetteers, in addition to providing complementary information if they are already available.

Details

Publication typeInproceedings
PublisherAnnual Conference of the International Speech Communication Association (Interspeech)
> Publications > Towards Unsupervised Spoken Language Understanding: Exploiting Query Click Logs for Slot Filling