Enhancing Complex Question Answering over Knowledge Graphs through Evidence Pattern Retrieval

TL;DR

This paper introduces Evidence Pattern Retrieval (EPR), a novel method that explicitly models structural dependencies among evidence facts to improve subgraph extraction in knowledge graph question answering, leading to significant performance gains.

Contribution

The paper proposes EPR, a new approach that indexes atomic adjacency patterns to better capture structural dependencies during subgraph extraction in KGQA.

Findings

EPR improves F1 scores by over 10 points on ComplexWebQuestions.

EPR achieves competitive results on WebQuestionsSP.

Explicit modeling of structural dependencies enhances subgraph extraction.

Abstract

Information retrieval (IR) methods for KGQA consist of two stages: subgraph extraction and answer reasoning. We argue current subgraph extraction methods underestimate the importance of structural dependencies among evidence facts. We propose Evidence Pattern Retrieval (EPR) to explicitly model the structural dependencies during subgraph extraction. We implement EPR by indexing the atomic adjacency pattern of resource pairs. Given a question, we perform dense retrieval to obtain atomic patterns formed by resource pairs. We then enumerate their combinations to construct candidate evidence patterns. These evidence patterns are scored using a neural model, and the best one is selected to extract a subgraph for downstream answer reasoning. Experimental results demonstrate that the EPR-based approach has significantly improved the F1 scores of IR-KGQA methods by over 10 points on ComplexWebQuestions and achieves competitive performance on WebQuestionsSP.