Explainable Conversational Question Answering over Heterogeneous Sources via Iterative Graph Neural Networks

TL;DR

This paper introduces EXPLAIGNN, a novel method for conversational question answering that integrates multiple heterogeneous sources with explanations, improving answer accuracy and interpretability.

Contribution

The paper presents a new iterative graph neural network approach that combines diverse data sources and provides user-understandable explanations for answers.

Findings

EXPLAIGNN outperforms existing baselines in accuracy.

User study confirms answers are understandable.

Method effectively integrates multiple sources for comprehensive answers.

Abstract

In conversational question answering, users express their information needs through a series of utterances with incomplete context. Typical ConvQA methods rely on a single source (a knowledge base (KB), or a text corpus, or a set of tables), thus being unable to benefit from increased answer coverage and redundancy of multiple sources. Our method EXPLAIGNN overcomes these limitations by integrating information from a mixture of sources with user-comprehensible explanations for answers. It constructs a heterogeneous graph from entities and evidence snippets retrieved from a KB, a text corpus, web tables, and infoboxes. This large graph is then iteratively reduced via graph neural networks that incorporate question-level attention, until the best answers and their explanations are distilled. Experiments show that EXPLAIGNN improves performance over state-of-the-art baselines. A user study demonstrates that derived answers are understandable by end users.