HERALD: An Annotation Efficient Method to Detect User Disengagement in Social Conversations

TL;DR

HERALD introduces an annotation-efficient framework that automatically labels and denoises training data to effectively detect user disengagement in social conversations, reducing manual effort and maintaining high accuracy.

Contribution

HERALD presents a novel denoising-based annotation framework that enhances user disengagement detection in dialog systems with minimal manual labeling.

Findings

Achieves 86% detection accuracy on two dialog datasets.

Significantly improves annotation efficiency over manual labeling.

Effective in real-time user engagement monitoring.

Abstract

Open-domain dialog systems have a user-centric goal: to provide humans with an engaging conversation experience. User engagement is one of the most important metrics for evaluating open-domain dialog systems, and could also be used as real-time feedback to benefit dialog policy learning. Existing work on detecting user disengagement typically requires hand-labeling many dialog samples. We propose HERALD, an efficient annotation framework that reframes the training data annotation process as a denoising problem. Specifically, instead of manually labeling training samples, we first use a set of labeling heuristics to label training samples automatically. We then denoise the weakly labeled data using the Shapley algorithm. Finally, we use the denoised data to train a user engagement detector. Our experiments show that HERALD improves annotation efficiency significantly and achieves 86% user disengagement detection accuracy in two dialog corpora.