Filters








33 Hits in 2.5 sec

Neural Question Answering at BioASQ 5B

Georg Wiese, Dirk Weissenborn, Mariana Neves
2017 BioNLP 2017  
At the core of our system, we use FastQA, a state-ofthe-art neural QA system.  ...  This paper describes our submission to the 2017 BioASQ challenge. We participated in Task B, Phase B which is concerned with biomedical question answering (QA).  ...  Conclusion In this paper, we summarized the system design of our BioASQ 5B submission for factoid and list questions. We use a neural architecture which is trained end-to-end on the QA task.  ... 
doi:10.18653/v1/w17-2309 dblp:conf/bionlp/WieseWN17 fatcat:k2qbakbyfva5ddsjnkj75v4vt4

Neural Question Answering at BioASQ 5B [article]

Georg Wiese, Dirk Weissenborn, Mariana Neves
2017 arXiv   pre-print
At the core of our system, we use FastQA, a state-of-the-art neural QA system.  ...  This paper describes our submission to the 2017 BioASQ challenge. We participated in Task B, Phase B which is concerned with biomedical question answering (QA).  ...  Conclusion In this paper, we summarized the system design of our BioASQ 5B submission for factoid and list questions. We use a neural architecture which is trained end-to-end on the QA task.  ... 
arXiv:1706.08568v1 fatcat:unzporpn6zg7nn46ig3e2t572e

Macquarie University at BioASQ 5b – Query-based Summarisation Techniques for Selecting the Ideal Answers

Diego Molla
2017 BioNLP 2017  
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.  ...  Our experiments and the ROUGE results of the five test batches of BioASQ indicate surprisingly good results for the trivial approach.  ...  Conclusions At the time of writing, only the ROUGE scores of BioASQ 5b were available.  ... 
doi:10.18653/v1/w17-2308 dblp:conf/bionlp/Aliod17 fatcat:teia6uzribcshmhfuioszcigca

BioAMA: Towards an End to End BioMedical Question Answering System

Vasu Sharma, Nitish Kulkarni, Srividya Pranavi, Gabriel Bayomi, Eric Nyberg, Teruko Mitamura
2018 Proceedings of the BioNLP 2018 workshop  
In this paper, we present a novel Biomedical Question Answering system, BioAMA: "Biomedical Ask Me Anything" on task 5b of the annual BioASQ challenge (Balikas et al., 2015) .  ...  0.72 and a ROUGE-SU4 score of 0.71 on ideal answer questions (7% improvement over the previous best model).  ...  We achieve state of the art results in automatic evaluation measures for the ideal answer questions in Task 5b of the BioASQ dataset, yielding a 7% improvement over the previous state of the art system  ... 
doi:10.18653/v1/w18-2312 dblp:conf/bionlp/SharmaKPBNM18 fatcat:dyqv3lgsyvexzcfikxt4emsf2i

Macquarie University at BioASQ 5b -- Query-based Summarisation Techniques for Selecting the Ideal Answers [article]

Diego Molla-Aliod
2017 arXiv   pre-print
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.  ...  Our experiments and the ROUGE results of the five test batches of BioASQ indicate surprisingly good results for the trivial approach.  ...  -Phase B: Question answering. • BioASQ 5c: Funding information extraction from biomedical literature. The questions used in BioASQ 5b were of three types: yes/no, factoid, list, and summary.  ... 
arXiv:1706.02095v2 fatcat:orw6s35izndexgjms2gfvndnie

Measuring Domain Portability and ErrorPropagation in Biomedical QA [article]

Stefan Hosein, Daniel Andor, Ryan McDonald
2019 arXiv   pre-print
In this work we present Google's submission to the BioASQ 7 biomedical question answering (QA) task (specifically Task 7b, Phase B).  ...  However, the biggest quality bottleneck is at the retrieval stage, where we see large drops in metrics – over 10pts absolute – when using non gold inputs to the QA model.  ...  Table 1 . 1 Table showing the differences between the input at test time for each model. Table 3 . 3 Performance on BioASQ Task 4 and 5b, Phase B averaged over all batches.  ... 
arXiv:1909.09704v2 fatcat:36n2tglfevevjhnbgwtvia2oli

How to Pre-Train Your Model? Comparison of Different Pre-Training Models for Biomedical Question Answering [article]

Sanjay Kamath, Brigitte Grau, Yue Ma
2019 arXiv   pre-print
We compare two pre-training models based on reading comprehension and open domain question answering models and determine the performance when fine-tuned and tested over BIOASQ question answering dataset  ...  Similarly for question answering, pre-training and fine-tuning can be done in several ways.  ...  In Question Answering (QA) specifically on open domain, several neural network models have been introduced, such as Convolutional Neural Networks(CNN), Recurrent Neural Networks(RNN) using GRUs or LSTMs  ... 
arXiv:1911.00712v1 fatcat:vsf5ur22ovbvvovvog55hq4lpq

Results of the fifth edition of the BioASQ Challenge

Anastasios Nentidis, Konstantinos Bougiatiotis, Anastasia Krithara, Georgios Paliouras, Ioannis Kakadiaris
2017 BioNLP 2017  
The fifth challenge consisted of three tasks: semantic indexing, question answering and a new task on information extraction.  ...  The goal of the BioASQ challenge is to engage researchers into creating cuttingedge biomedical information systems.  ...  BioASQ is grateful to NLM for providing baselines for tasks 5a and 5c and the CMU team for providing the baselines for task 5b. Finally, we would also like to thank all teams for their participation.  ... 
doi:10.18653/v1/w17-2306 dblp:conf/bionlp/NentidisBKPK17 fatcat:zj42lfjjdvgvzcxarhbhx3la4y

Assessing the performance of Olelo, a real-time biomedical question answering application

Mariana Neves, Fabian Eckert, Hendrik Folkerts, Matthias Uflacker
2017 BioNLP 2017  
Question answering (QA) can support physicians and biomedical researchers to find answers to their questions in the scientific literature.  ...  In addition to the BioASQ evaluation, we compared our system to other on-line biomedical QA systems in terms of the response time and the quality of the answers. 1 http://hpi.de/plattner/olelo sults for  ...  task 5b phases B (exact answers).  ... 
doi:10.18653/v1/w17-2344 dblp:conf/bionlp/NevesEFU17 fatcat:lerk6gynazallo6vamr4coqubq

Bio-AnswerFinder: a system to find answers to questions from biomedical texts

2019 Database: The Journal of Biological Databases and Curation  
The answer sentences are further ranked by a fine-tuned bidirectional encoder representation from transformers (BERT) classifier trained using 100 answer candidate sentences per question for 492 BioASQ  ...  Our unsupervised baseline system achieves a mean reciprocal rank score of 0.46 and Precision@1 of 0.32 on 936 questions from BioASQ.  ...  (8) introduced a neural system to detect answer spans in the gold standard snippets provided by BioASQ for question answering task (Task B) competition.  ... 
doi:10.1093/database/baz137 pmid:31925435 pmcid:PMC7053013 fatcat:scxluxhtd5burcx2xkebte6die

Simple and Effective Semi-Supervised Question Answering

Bhuwan Dhingra, Danish Danish, Dheeraj Rajagopal
2018 Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 2 (Short Papers)  
Recent success of deep learning models for the task of extractive Question Answering (QA) is hinged on the availability of large annotated corpora.  ...  Our system exploits the document structure to create cloze-style questions from these base documents; pre-trains a powerful neural network on the cloze style questions; and further finetunes the model  ...  We also test on the BioASQ 5b dataset, which consists of question-answer pairs from PubMed abstracts.  ... 
doi:10.18653/v1/n18-2092 dblp:conf/naacl/DhingraPR18 fatcat:ba7chq5x3jc37pypo53nzbezka

Simple and Effective Semi-Supervised Question Answering [article]

Bhuwan Dhingra, Danish Pruthi, Dheeraj Rajagopal
2018 arXiv   pre-print
Recent success of deep learning models for the task of extractive Question Answering (QA) is hinged on the availability of large annotated corpora.  ...  Our system exploits the document structure to create cloze-style questions from these base documents; pre-trains a powerful neural network on the cloze style questions; and further fine-tunes the model  ...  We also test on the BioASQ 5b dataset, which consists of question-answer pairs from PubMed abstracts.  ... 
arXiv:1804.00720v1 fatcat:byoni2t5zbcklmrepj7iwsz3ru

Biomedical Question Answering: A Survey of Approaches and Challenges [article]

Qiao Jin, Zheng Yuan, Guangzhi Xiong, Qianlan Yu, Huaiyuan Ying, Chuanqi Tan, Mosha Chen, Songfang Huang, Xiaozhong Liu, Sheng Yu
2021 arXiv   pre-print
Automatic Question Answering (QA) has been successfully applied in various domains such as search engines and chatbots.  ...  There have been tremendous developments of BQA in the past two decades, which we classify into 5 distinctive approaches: classic, information retrieval, machine reading comprehension, knowledge base and question  ...  At BioASQ 5B, Mollá [117] observes that a trivial baseline that returns the top retrieved snippets as the ideal answer is hard to beat.  ... 
arXiv:2102.05281v2 fatcat:gngt6kffszbzbc3donik55dd64

Results of the Seventh Edition of the BioASQ Challenge [chapter]

Anastasios Nentidis, Konstantinos Bougiatiotis, Anastasia Krithara, Georgios Paliouras
2020 Communications in Computer and Information Science  
The aim of the BioASQ challenge is the promotion of systems and methodologies through the organization of a challenge on the tasks of large-scale biomedical semantic indexing and question answering.  ...  The results of the seventh edition of the BioASQ challenge are presented in this paper.  ...  Acknowledgments Google was a proud sponsor of the BioASQ Challenge in 2018. The seventh edition of BioASQ is also sponsored by the Atypon Systems inc.  ... 
doi:10.1007/978-3-030-43887-6_51 fatcat:76ind5ijizdphjmcnzgk7fjmii

SciFive: a text-to-text transformer model for biomedical literature [article]

Long N. Phan, James T. Anibal, Hieu Tran, Shaurya Chanana, Erol Bahadroglu, Alec Peltekian, Grégoire Altan-Bonnet
2021 arXiv   pre-print
BERT, BioBERT, Base T5) on tasks in named entity relation, relation extraction, natural language inference, and question-answering.  ...  We test SciFive on the factoid questions from the BioASQ 4b, 5b, and 6b challenges Tsatsaronis et al. (2015) . To preprocess the BioASQ data, we use the same approach as Lee et al. (2019) .  ...  For the BioASQ tasks Table 3 . 3 Expert assessment result on Question Answering tasks (Lenient Accuracy) Task BioBERT T5 SciFive (PubMed+PMC) SciFive (PMC) SciFive (Pubmed) BioAsq 4b  ... 
arXiv:2106.03598v1 fatcat:vce5ki2sone3nab4jz2gp3moci
« Previous Showing results 1 — 15 out of 33 results