Biomedical Question Answering via Weighted Neural Network Passage Retrieval

Galkó, Ferenc; Eickhoff, Carsten

doi:10.1007/978-3-319-76941-7_39

Cited by 10 publications

(6 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We select the biomedical domain given its complexity and because it is not usually the focus of QA systems. Although some biomedical QA systems have been proposed [22], these are limited in comparison to general ones. Biomedicalspecific approaches are essential due to the relevance of this topic to the society.…”

Section: Discussionmentioning

confidence: 99%

Generating Biomedical Question Answering Corpora From Q&A Forums

2020

View full text Add to dashboard Cite

Question Answering (QA) is a natural language processing task that aims at obtaining relevant answers to user questions. While some progress has been made in this area, biomedical questions are still a challenge to most QA approaches, due to the complexity of the domain and limited availability of training sets. We present a method to automatically extract question-article pairs from Q&A web forums, which can be used for document retrieval, a crucial step of most QA systems. The proposed framework extracts from selected forums the questions and the respective answers that contain citations. This way, QA systems based on document retrieval can be developed and evaluated using the question-article pairs annotated by users of these forums. We generated the BiQA corpus by applying our framework to three forums, obtaining 7,453 questions and 14,239 question-article pairs. We evaluated how the number of articles associated with each question and the number of votes on each answer affects the performance of baseline document retrieval approaches. Also, we demonstrated that the articles given as answers are significantly similar to the questions and trained a state-of-the-art deep learning model that obtained similar performance to using a dataset manually annotated by experts. The proposed framework can be used to update the BiQA corpus from the same forums as new posts are made, and from other forums that support their answers with documents. The BiQA corpus and the framework used to generate it are available at https://github.com/lasigeBioTM/BiQA.

show abstract

Section: Discussionmentioning

confidence: 99%

Generating Biomedical Question Answering Corpora From Q&A Forums

2020

View full text Add to dashboard Cite

show abstract

“…how relevant is this part of the context to the question of interest?). This general capability (with appropriate reformulation) has the potential to be broadly useful, both for determining similarity and relevance on other datasets, and for question answering in specialized domains [61].…”

Section: Question Answeringmentioning

confidence: 99%

A Survey of Deep Learning for Scientific Discovery

Raghu¹,

Schmidt²

2020

Preprint

View full text Add to dashboard Cite

Over the past few years, we have seen fundamental breakthroughs in core problems in machine learning, largely driven by advances in deep neural networks. At the same time, the amount of data collected in a wide array of scientific domains is dramatically increasing in both size and complexity. Taken together, this suggests many exciting opportunities for deep learning applications in scientific settings. But a significant challenge to this is simply knowing where to start. The sheer breadth and diversity of different deep learning techniques makes it difficult to determine what scientific problems might be most amenable to these methods, or which specific combination of methods might offer the most promising first approach. In this survey, we focus on addressing this central issue, providing an overview of many widely used deep learning models, spanning visual, sequential and graph structured data, associated tasks and different training methods, along with techniques to use deep learning with less data and better interpret these complex models -two central considerations for many scientific use cases. We also include overviews of the full design process, implementation tips, and links to a plethora of tutorials, research summaries and open-sourced deep learning pipelines and pretrained models, developed by the community. We hope that this survey will help accelerate the use of deep learning across different scientific domains.

show abstract

“…Numerous passage retrieval approaches have been proposed at BioASQ. For example, (Galkó and Eickhoff, 2018) proposed to apply a word embedding representation for question-passage sequences and then to compute their semantic relationship employing a weighted cosine distance. Another relevant approach, which obtained the best results in the 2018 BioASQ edition, was presented by the auen-nlp team (Brokos et al, 2018).…”

Section: Related Workmentioning

confidence: 99%

A Deep Metric Learning Method for Biomedical Passage Retrieval

Rosso-Mateus

González

Montes-y-Gómez

2020

Proceedings of the 28th International Conference on Computational Linguistics

View full text Add to dashboard Cite

Passage retrieval is the task of identifying text snippets that are valid answers for a natural language posed question. One way to address this problem is to look at it as a metric learning problem, where we want to induce a metric between questions and passages that assign smaller distances to more relevant passages. In this work, we present a novel method for passage retrieval that learns a metric for questions and passages based on their internal semantic interactions. The method uses a similar approach to that of triplet networks, where the training samples are composed of one anchor (the question) and two positive and negative samples (passages). However, and in contrast with triplet networks, the proposed method uses a novel deep architecture that better exploits the particularities of text and takes into consideration complementary relatedness measures. Besides, the paper presents a sampling strategy that selects both easy and hard negative samples which improves the accuracy of the trained model. The method is particularly well suited for domain-specific passage retrieval where it is very important to take into account different sources of information. The proposed approach was evaluated in a biomedical passage retrieval task, the BioASQ challenge, outperforming standard triplet loss substantially by 10%, and state-of-the-art performance by 26%.

show abstract

Biomedical Question Answering via Weighted Neural Network Passage Retrieval

Cited by 10 publications

References 12 publications

Generating Biomedical Question Answering Corpora From Q&A Forums

Generating Biomedical Question Answering Corpora From Q&A Forums

A Survey of Deep Learning for Scientific Discovery

A Deep Metric Learning Method for Biomedical Passage Retrieval

Contact Info

Product

Resources

About