# Macquarie University at BioASQ 5b -- Query-based Summarisation Techniques for Selecting the Ideal Answers

## TL;DR

This paper describes Macquarie University's participation in BioASQ 5b, where they applied various query-based extractive summarisation techniques, including deep learning, to generate ideal answers, with surprisingly strong results even from trivial methods.

## Contribution

The paper presents multiple extractive summarisation approaches, including a simple baseline and deep learning, demonstrating effective results in the BioASQ challenge.

## Key findings

- Trivial snippet selection achieved high ROUGE scores.
- Most runs outperformed competitors on initial test batches.
- Deep learning approaches showed promising results.

## Abstract

Macquarie University's contribution to the BioASQ challenge (Task 5b Phase B) focused on the use of query-based extractive summarisation techniques for the generation of the ideal answers. Four runs were submitted, with approaches ranging from a trivial system that selected the first $n$ snippets, to the use of deep learning approaches under a regression framework. Our experiments and the ROUGE results of the five test batches of BioASQ indicate surprisingly good results for the trivial approach. Overall, most of our runs on the first three test batches achieved the best ROUGE-SU4 results in the challenge.