Two-stage Voice Application Recommender System for Unhandled Utterances in Intelligent Personal Assistant

TL;DR

This paper introduces a two-stage recommender system for unhandled voice requests in intelligent personal assistants, improving skill matching accuracy and user satisfaction through novel retrieval and bias mitigation techniques.

Contribution

It proposes a new shortlister-reranker framework with methods to handle incomplete data and exposure bias, enhancing unhandled utterance resolution.

Findings

Significant improvement in offline skill matching accuracy.

Online A/B tests show increased user satisfaction.

Effective handling of incomplete ground truth and exposure bias.

Abstract

Intelligent personal assistants (IPA) enable voice applications that facilitate people's daily tasks. However, due to the complexity and ambiguity of voice requests, some requests may not be handled properly by the standard natural language understanding (NLU) component. In such cases, a simple reply like "Sorry, I don't know" hurts the user's experience and limits the functionality of IPA. In this paper, we propose a two-stage shortlister-reranker recommender system to match third-party voice applications (skills) to unhandled utterances. In this approach, a skill shortlister is proposed to retrieve candidate skills from the skill catalog by calculating both lexical and semantic similarity between skills and user requests. We also illustrate how to build a new system by using observed data collected from a baseline rule-based system, and how the exposure biases can generate discrepancy between offline and human metrics. Lastly, we present two relabeling methods that can handle the incomplete ground truth, and mitigate exposure bias. We demonstrate the effectiveness of our proposed system through extensive offline experiments. Furthermore, we present online A/B testing results that show a significant boost on user experience satisfaction.