A Study of Query Performance Prediction for Answer Quality Determination

Abstract

We study a constrained retrieval setting in which either a single qualitative answer is provided as a response to a user-query or none. Given a user-query and the "best" answer that was retrieved from the underlying search engine, we wish to determine whether or not to accept it. To address this challenge, we propose an answer quality determination approach which leverages a novel set of answer-level query performance prediction (QPP) features, derived from a couple of recent discriminative QPP frameworks. Using various search benchmarks with both ad-hoc retrieval and non-factoid question answering (QA) tasks, we demonstrate the effectiveness of our approach.