Biomedical Question Answering via Weighted Neural Network Passage Retrieval

@article{Galk2018BiomedicalQA,
  title={Biomedical Question Answering via Weighted Neural Network Passage Retrieval},
  author={Ferenc Galk{\'o} and Carsten Eickhoff},
  journal={ArXiv},
  year={2018},
  volume={abs/1801.02832}
}
The amount of publicly available biomedical literature has been growing rapidly in recent years, yet question answering systems still struggle to exploit the full potential of this source of data. In a preliminary processing step, many question answering systems rely on retrieval models for identifying relevant documents and passages. This paper proposes a weighted cosine distance retrieval scheme based on neural network word embeddings. Our experiments are based on publicly available data and… 
Generating Biomedical Question Answering Corpora From Q&A Forums
TLDR
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.
A Deep Metric Learning Method for Biomedical Passage Retrieval
TLDR
This work presents a novel method for passage retrieval that learns a metric for questions and passages based on their internal semantic interactions, and uses a novel deep architecture that better exploits the particularities of text and takes into consideration complementary relatedness measures.
Semantic Sequential Query Expansion for Biomedical Article Search
TLDR
The semantic sequential dependence model is proposed, which provides an innovative combination of semantic information and the conventional SDM and the experimental results show that the query expansion approach outperforms the baseline and other participants in the BioASQ competitions.
Embedding Electronic Health Records for Clinical Information Retrieval
TLDR
This study presents an end-to-end neural clinical decision support system that recommends relevant literature for individual patients via distant supervision on the well-known MIMIC-III collection (abundant resource) and shows significant improvements in retrieval effectiveness over traditional statistical as well as purely locally supervised retrieval models.
Investigation of Passage Based Ranking Models to Improve Document Retrieval
TLDR
Experimental results indicate that for the passage level technique, the worst-performing queries are damaged slightly and the those that perform well are boosted for the WebAp collection, however, the rank-based similarity function boosted the performance of the difficult queries in the Ohsumed collection.
Using Deep Learning Based Natural Language Processing Techniques for Clinical Decision-Making with EHRs
TLDR
It is found that the distance to revolutionize the existing healthcare sector using deep learning methods still remains long, but the recent progress made by these proposed methods have already made a promising good start.
Online Disease Diagnosis with Inductive Heterogeneous Graph Convolutional Networks
TLDR
A disease self-diagnosis system with a corresponding EHR Graph-based Symptom Retrieval System (GraphRet) that can search and provide a list of relevant alternative symptoms by tracing the predefined meta-paths and validate the superiority of the model on a large-scale EHR dataset.
Vocabulary Mismatch Avoidance Techniques
TLDR
This survey paper deals about various methods created by different researchers using methods such as query expansion, stemming and full-text indexing to solve problems of IR systems such as ambiguity, vocabulary mismatch due to polysemy and synonymy.
A Survey of Deep Learning for Scientific Discovery
TLDR
This survey provides 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.
DC3 -- A Diagnostic Case Challenge Collection for Clinical Decision Support
TLDR
DC3 is presented, a collection of 31 extremely difficult diagnostic case challenges, manually compiled and solved by clinical experts, and a number of temporally ordered physician-generated observations alongside the eventually confirmed true diagnosis are presented.
...
...

References

SHOWING 1-10 OF 17 REFERENCES
HPI Question Answering System in BioASQ 2016
TLDR
This work describes the participation in the task 4b of the BioASQ challenge using two QA systems that were developed for biomedicine and preliminary results show that their systems achieved first and second positions in the snippet retrieval sub-task and for the generation of ideal answers.
Using Centroids of Word Embeddings and Word Mover’s Distance for Biomedical Document Retrieval in Question Answering
TLDR
A document retrieval method for question answering is proposed that represents documents and questions as weighted centroids of word embeddings and reranks the retrieved documents with a relaxation of Word Mover's Distance and is competitive with PUBMED.
Large-Scale Semantic Indexing and Question Answering in Biomedicine
TLDR
This paper presents the methods and approaches employed in terms of participation in the 2016 version of the BioASQ challenge, and extended the successful ensemble approach of last year with additional models for the semantic indexing task and for the question answering task.
An overview of the BIOASQ large-scale biomedical semantic indexing and question answering competition
TLDR
Overall, BioASQ helped obtain a unified view of how techniques from text classification, semantic indexing, document and passage retrieval, question answering, and text summarization can be combined to allow biomedical experts to obtain concise, user-understandable answers to questions reflecting their real information needs.
SQuAD: 100,000+ Questions for Machine Comprehension of Text
TLDR
A strong logistic regression model is built, which achieves an F1 score of 51.0%, a significant improvement over a simple baseline (20%).
Learning to Answer Biomedical Questions: OAQA at BioASQ 4B
TLDR
The system extends the Yang et al. (2015) system and integrates additional biomedical and generalpurpose NLP annotators, machine learning modules for search result scoring, collective answer reranking, and yes/no answer prediction.
KSAnswer: Question-answering System of Kangwon National University and Sogang University in the 2016 BioASQ Challenge
TLDR
A question– answering system that returns relevant documents and snippets (with particular emphasis on snippets) from a large medical document collection that retrieves candidate answer sentences using a cluster–based language model and re–ranks the retrieved top-n sentences using five independent similarity models based on shallow semantic analysis.
A Deep Relevance Matching Model for Ad-hoc Retrieval
TLDR
A novel deep relevance matching model (DRMM) for ad-hoc retrieval that employs a joint deep architecture at the query term level for relevance matching and can significantly outperform some well-known retrieval models as well as state-of-the-art deep matching models.
End-To-End Memory Networks
TLDR
A neural network with a recurrent attention model over a possibly large external memory that is trained end-to-end, and hence requires significantly less supervision during training, making it more generally applicable in realistic settings.
Question answering in TREC
TLDR
A brief summary of the findings of the TREC question answering track to date is provided and the future directions of the track are discussed.
...
...