Zero-Shot Learning for Semantic Utterance Classification

TL;DR

This paper introduces a zero-shot learning approach for semantic utterance classification that leverages a semantic space learned from large query logs, enabling classification of unseen categories with state-of-the-art performance.

Contribution

It presents a novel zero-shot learning framework that learns discriminative semantic features without supervision for SUC tasks.

Findings

Effective zero-shot classification on SUC dataset

Achieved state-of-the-art results by combining semantic and supervised features

Demonstrated the utility of deep neural networks trained on query logs

Abstract

We propose a novel zero-shot learning method for semantic utterance classification (SUC). It learns a classifier $f: X \to Y$ for problems where none of the semantic categories $Y$ are present in the training set. The framework uncovers the link between categories and utterances using a semantic space. We show that this semantic space can be learned by deep neural networks trained on large amounts of search engine query log data. More precisely, we propose a novel method that can learn discriminative semantic features without supervision. It uses the zero-shot learning framework to guide the learning of the semantic features. We demonstrate the effectiveness of the zero-shot semantic learning algorithm on the SUC dataset collected by (Tur, 2012). Furthermore, we achieve state-of-the-art results by combining the semantic features with a supervised method.