Abductive Matching in Question Answering

TL;DR

This paper presents a hybrid semantic parsing approach for question answering over semi-structured data, combining machine learning for non-syntactic matching with rule-based parsing, achieving state-of-the-art accuracy on Wikipedia tables.

Contribution

It introduces a novel abductive matching method that integrates machine learning with manual rules for improved question answering accuracy.

Findings

Achieved 40.42% accuracy on Wikipedia table dataset.

Demonstrated improved interpretability and debuggability of the QA system.

Validated effectiveness of abductive matching in semantic parsing.

Abstract

We study question-answering over semi-structured data. We introduce a new way to apply the technique of semantic parsing by applying machine learning only to provide annotations that the system infers to be missing; all the other parsing logic is in the form of manually authored rules. In effect, the machine learning is used to provide non-syntactic matches, a step that is ill-suited to manual rules. The advantage of this approach is in its debuggability and in its transparency to the end-user. We demonstrate the effectiveness of the approach by achieving state-of-the-art performance of 40.42% accuracy on a standard benchmark dataset over tables from Wikipedia.