Concept recognition as a machine translation problem.
Concept recognition as a machine translation problem.
复制标题
DOI：
10.1186/s12859-021-04141-4
复制
发表时间：
2021-12-17
影响因子：
3
通讯作者：
Hunter LE
中科院分区：
文献类型：
--
作者：
Boguslav MR;Hailu ND;Bada M;Baumgartner WA Jr;Hunter LE
在生物医学自然语言处理中，特定本体概念的自动分配是一项关键任务，也是许多开放共享任务的主题。虽然目前的技术水平涉及使用神经网络语言模型作为后处理步骤，但要识别的大量本体类和有限的黄金标准训练数据阻碍了完全基于机器学习的端到端系统的创建。最近，Hailu等人将概念识别问题重新定义为一种机器翻译，并证明序列到序列机器学习模型有可能优于多类分类方法。通过对替代方法和超参数选择的广泛研究，我们系统地描述了影响几种序列到序列机器学习方法的准确性和效率的因素。我们不仅在各种各样的本体中识别出性能最好的系统和参数，而且还提供了对各种不同的资源需求和替代方法的超参数鲁棒性的见解。这种系统的优点和缺点的分析表明，有前途的途径，为未来的改进，以及设计选择，可以提高计算效率与性能的小成本。用于跨度检测的生物医学文本挖掘（BioBERT）转换器的双向编码器表示沿着用于概念规范化的神经机器翻译（OpenNMT）开源工具包，为CRAFT语料库中注释的大多数本体实现了最先进的性能。这种方法比几种替代方法使用更少的计算资源，包括硬件、内存和时间。机器翻译是完全基于机器学习的概念识别的一种有前途的途径，它在CRAFT语料库上实现了最先进的结果，通过与2019年CRAFT共享任务的先前结果进行直接比较进行评估。阐明序列到序列方法针对本体标识符的令人惊讶的良好性能的原因的实验表明，通过映射到替代目标概念表示，可能会取得进一步的进展。所有代码和模型可以在https://github.com/UCDenver-ccp/Concept-Recognition-as-Translation上找到。
Automated assignment of specific ontology concepts to mentions in text is a critical task in biomedical natural language processing, and the subject of many open shared tasks. Although the current state of the art involves the use of neural network language models as a post-processing step, the very large number of ontology classes to be recognized and the limited amount of gold-standard training data has impeded the creation of end-to-end systems based entirely on machine learning. Recently, Hailu et al. recast the concept recognition problem as a type of machine translation and demonstrated that sequence-to-sequence machine learning models have the potential to outperform multi-class classification approaches. We systematically characterize the factors that contribute to the accuracy and efficiency of several approaches to sequence-to-sequence machine learning through extensive studies of alternative methods and hyperparameter selections. We not only identify the best-performing systems and parameters across a wide variety of ontologies but also provide insights into the widely varying resource requirements and hyperparameter robustness of alternative approaches. Analysis of the strengths and weaknesses of such systems suggest promising avenues for future improvements as well as design choices that can increase computational efficiency with small costs in performance. Bidirectional encoder representations from transformers for biomedical text mining (BioBERT) for span detection along with the open-source toolkit for neural machine translation (OpenNMT) for concept normalization achieve state-of-the-art performance for most ontologies annotated in the CRAFT Corpus. This approach uses substantially fewer computational resources, including hardware, memory, and time than several alternative approaches. Machine translation is a promising avenue for fully machine-learning-based concept recognition that achieves state-of-the-art results on the CRAFT Corpus, evaluated via a direct comparison to previous results from the 2019 CRAFT shared task. Experiments illuminating the reasons for the surprisingly good performance of sequence-to-sequence methods targeting ontology identifiers suggest that further progress may be possible by mapping to alternative target concept representations. All code and models can be found at: https://github.com/UCDenver-ccp/Concept-Recognition-as-Translation.
登录
查看更多内容
影响因子：
3
作者：
Bada M;Eckert M;Evans D;Garcia K;Shipley K;Sitnikov D;Baumgartner WA Jr;Cohen KB;Verspoor K;Blake JA;Hunter LE
通讯作者：
Hunter LE
影响因子：
3
作者：
Funk C;Baumgartner W Jr;Garcia B;Roeder C;Bada M;Cohen KB;Hunter LE;Verspoor K
通讯作者：
Verspoor K
影响因子：
8.6
作者：
Corbett, Peter;Boyle, John
通讯作者：
Boyle, John
影响因子：
1.9
作者：
Basaldella M;Furrer L;Tasso C;Rinaldi F
通讯作者：
Rinaldi F
影响因子：
8.6
作者：
Furrer, Lenz;Jancso, Anna;Rinaldi, Fabio
通讯作者：
Rinaldi, Fabio