Towards Handling Unconstrained User Preferences in Dialogue

TL;DR

This paper introduces a dialogue system that allows users to specify unconstrained preferences by retrieving relevant entities from unstructured knowledge, enhancing flexibility over traditional schema-driven methods.

Contribution

It proposes a novel approach using unstructured knowledge snippets and transformer-based classifiers to handle unconstrained user preferences in dialogue systems.

Findings

Transformer-based classifiers achieve high relevance accuracy.

Unstructured knowledge improves user preference handling.

The approach outperforms traditional schema-driven systems.

Abstract

A user input to a schema-driven dialogue information navigation system, such as venue search, is typically constrained by the underlying database which restricts the user to specify a predefined set of preferences, or slots, corresponding to the database fields. We envision a more natural information navigation dialogue interface where a user has flexibility to specify unconstrained preferences that may not match a predefined schema. We propose to use information retrieval from unstructured knowledge to identify entities relevant to a user request. We update the Cambridge restaurants database with unstructured knowledge snippets (reviews and information from the web) for each of the restaurants and annotate a set of query-snippet pairs with a relevance label. We use the annotated dataset to train and evaluate snippet relevance classifiers, as a proxy to evaluating recommendation accuracy. We show that with a pretrained transformer model as an encoder, an unsupervised/supervised classifier achieves a weighted F1 of .661/.856.