Macquarie University at BioASQ 5b -- Query-based Summarisation Techniques for Selecting the Ideal Answers
This work addresses the problem of biomedical question-answering summarization for participants in the BioASQ challenge, but it is incremental as it compares existing methods without introducing major innovations.
The researchers tackled the problem of generating ideal answers for the BioASQ challenge using query-based extractive summarization techniques, and found that a trivial approach selecting the first n snippets achieved surprisingly good results, with most runs on the first three test batches achieving the best ROUGE-SU4 scores in the challenge.
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.