Learning to answer questions

TL;DR

This paper introduces an open-domain question-answering system that learns from past interactions using pattern-based answer extraction, improving performance and enabling correction of previous errors.

Contribution

It presents a novel pattern-based learning approach for answer extraction that enhances system performance and adaptability over prior methods.

Findings

The approach improves answer accuracy when combined with traditional answer-extraction strategies.

The system can learn from answered questions to improve future responses.

It can rectify past incorrect or unresolved questions.

Abstract

We present an open-domain Question-Answering system that learns to answer questions based on successful past interactions. We follow a pattern-based approach to Answer-Extraction, where (lexico-syntactic) patterns that relate a question to its answer are automatically learned and used to answer future questions. Results show that our approach contributes to the system's best performance when it is conjugated with typical Answer-Extraction strategies. Moreover, it allows the system to learn with the answered questions and to rectify wrong or unsolved past questions.