Macquarie University's contribution to the BioASQ challenge (Task 5b Phase B)focused on the use of query-based extractive summarisation techniques for thegeneration of the ideal answers. Four runs were submitted, with approachesranging from a trivial system that selected the first $n$ snippets, to the useof deep learning approaches under a regression framework. Our experiments andthe ROUGE results of the five test batches of BioASQ indicate surprisingly goodresults for the trivial approach. Overall, most of our runs on the first threetest batches achieved the best ROUGE-SU4 results in the challenge.
展开▼
机译：麦格理大学对BioASQ挑战（任务5b B阶段）的贡献集中在使用基于查询的提取摘要技术来生成理想答案。提交了四次运行，方法从选择第一个$ n $片段的琐碎系统到在回归框架下使用深度学习方法。我们的实验和五个BioASQ测试批次的ROUGE结果表明，这种简单方法的结果出奇的好。总体而言，在前三个测试批次中，我们的大多数运行都在挑战中获得了最佳的ROUGE-SU4结果。
展开▼
