Near-Zero-Shot Suggestion Mining with a Little Help from WordNet

TL;DR

This paper introduces zero-shot suggestion classification methods leveraging entailment and WordNet to identify user suggestions in reviews without training data for specific labels.

Contribution

It proposes novel entailment-based zero-shot approaches for suggestion detection, utilizing WordNet to improve label assignment and prediction accuracy.

Findings

Effective zero-shot suggestion classification achieved

WordNet-enhanced label assignment improves results

Validated through comprehensive experiments

Abstract

In this work, we explore the constructive side of online reviews: advice, tips, requests, and suggestions that users provide about goods, venues, services, and other items of interest. To reduce training costs and annotation efforts needed to build a classifier for a specific label set, we present and evaluate several entailment-based zero-shot approaches to suggestion classification in a label-fully-unseen fashion. In particular, we introduce the strategy of assigning target class labels to sentences in English language with user intentions, which significantly improves prediction quality. The proposed strategies are evaluated with a comprehensive experimental study that validated our results both quantitatively and qualitatively.