Skip to main content
Log in

SSDM2: a Two-Stage Semantic Sequential Dependence Model Framework for Biomedical Question Answering

  • Published:
Cognitive Computation Aims and scope Submit manuscript

Abstract

Biomedical question answering is a hot and challenging topic in artificial intelligence and natural language processing as it helps to analyze multiple, large, and fast-growing biomedical knowledge sources. Most researchers manage to address the problem through constructing a knowledge base but these approaches require much expertise as well as workload. In this paper, we propose a two-stage semantic sequential dependence model (SSDM2 ) framework based on a cognitive-inspired model and sequential dependence model (SDM) to answer biomedical questions with relevant snippets in academic papers. Concretely, we firstly search relevant articles and generate candidate snippets with a SSDM, which is proposed to integrate the semantic and sequential information within questions together. Afterwards, another SSDM is utilized to measure the relevances between the questions and corresponding candidate snippets and rank these snippets. A biomedical question answering system is constructed based on the proposed framework and evaluated on 3-year BioASQ 2013-15 benchmarks. Statistics indicate the proposed framework SSDM2 outperforms several state-of-the-art baselines and BioASQ participants. The proposed SSDM2 is an effective and robust framework for biomedical question answering.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4

Similar content being viewed by others

Notes

  1. http://trec.nist.gov/trec_eval/

  2. For BioASQ 2015, we use map@10 based on the official measure.

References

  1. Balikas G, Kosmopoulos A, Krithara A, Paliouras G, Kakadiaris IA. Results of the BioASQ tasks of the question answering lab at CLEF 2015. Working Notes of CLEF 2015 - Conference and Labs of the Evaluation forum, Toulouse, France, September 8-11, 2015; 2015.

  2. Bauer MA, Berleant D. Usability survey of biomedical question answering systems. Hum Genomics 2012;6(1):17.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Cai F, Chen H. A probabilistic model for information retrieval by mining user behaviors. Cogn Comput 2016; 8(3):494–504.

    Article  Google Scholar 

  4. Cairns BL, Nielsen RD, Masanz JJ, Martin JH, Palmer MS, Ward WH, Savova GK. The MiPACQ clinical question answering system. AMIA Annual Symposium Proceedings, vol 2011, p 171. American Medical Informatics Association; 2011.

  5. Choi S. Snumedinfo at CLEF bioasq 2015. Working notes of CLEF 2015 - Conference and Labs of the Evaluation forum, Toulouse, France, September 8-11, 2015; 2015.

  6. Cohen AM, Yang J, Fisher S, Roark B, Hersh WR. The ohsu biomedical question answering system framework. Text REtrieval Conference; 2007.

  7. Croft B, Metzler D, Strohman T. Search engines—information retrieval in practice. Comput J 2011;54(5):831–832.

    Article  Google Scholar 

  8. Demner-Fushman D, Humphrey SM, Ide NC, Loane RF, Mork JG, Ruch P, Ruiz ME, Smith LH, Wilbur WJ, Aronson AR. Combining resources to find answers to biomedical questions. TREC; 2007.

  9. Figuerola CG, Berrocal JLA, ngel F, Zazo R, Mateos M. 2009. Retrieval of snippets of web pages converted to plain text. more questions than answers. Lecture Notes in Computer Science.

  10. Han H, Athenikos SJ. Biomedical question answering: a survey. Comput Methods Prog Biomed 2010;99(1): 1–24.

    Article  Google Scholar 

  11. Lee M, Cimino J, Zhu HR, Sable C, Shanker V, Ely J, Yu H. Beyond information retrievalłmedical question answering. AMIA annual symposium proceedings, vol. 2006, p. 469. American Medical Informatics Association; 2006.

  12. Li Y, Pan Q, Yang T, Wang S, Tang J, Cambria E. 2017. Learning word representations for sentiment analysis. Cognitive Computation (6), 1–9.

  13. Makar R, Kouta M, Badr A. A service oriented architecture for biomedical question answering system. Services Part II, IEEE Congress on, pp. 73–80; 2008.

  14. Mao Y, Lu Z. NCBI at the 2015 BioASQ challenge task: baseline results from mesh now. Working Notes of CLEF 2015 - Conference and Labs of the Evaluation forum, Toulouse, France, September 8-11, 2015; 2015.

  15. Mao Y, Wei C, Lu Z. NCBI at the 2014 BioASQ challenge task: large-scale biomedical semantic indexing and question answering. Working Notes for CLEF 2014 Conference, Sheffield, UK, September 15-18, 2014., pp. 1319–1327; 2014.

  16. Metzler D, Croft WB. A Markov random field model for term dependencies. SIGIR, pp. 472–479; 2005.

  17. Mittal S, Gupta S, Mittal DA, Bhatia S. 2008. Bioinqa: addressing bottlenecks of biomedical domain through biomedical question answering system. International Conference on Systemics, Cybernetics and Informatics (ICSCI-2008), India.

  18. Papanikolaou Y, Dimitriadis D, Tsoumakas G, Laliotis M, Markantonatos N, Vlahavas IP. Ensemble approaches for large-scale multi-label classification and question answering in biomedicine. Working Notes for CLEF 2014 Conference, Sheffield, UK, September 15-18, 2014., pp 1348–1360; 2014. http://ceur-ws.org/Vol-1180/CLEF2014wn-QA-PapanikolaouEt2014.pdf.

  19. Peng H, Cambria E, Hussain A. A review of sentiment analysis research in chinese language. Cogn Comput 2017;9:1–13.

    Article  Google Scholar 

  20. Sable C, Lee M, Zhu HR, Yu H. Question analysis for biomedical question answering. AMIA Annual Symposium Proceedings, vol. 2005, p. 1102. American Medical Informatics Association; 2005.

  21. Sakai T. Statistical reform in information retrieval SIGIR Forum 2014;48(1):3–12.

    Article  Google Scholar 

  22. Shi Z, Melli G, Wang Y, Liu Y, Gu B, Kashani MM, Sarkar A, Popowich F. Question answering summarization of multiple biomedical documents. Berlin: Springer; 2007.

    Book  Google Scholar 

  23. Sondhi P, Raj P, Kumar VV, Mittal A. Question processing and clustering in indoc: a biomedical question answering system. EURASIP J Bioinforma Syst Biol 2007;2007:1.

    Article  Google Scholar 

  24. Tsatsaronis G, Balikas G, Malakasiotis P, Partalas I, Zschunke M, Alvers MR, Weissenborn D, Krithara A, Petridis S, Polychronopoulos D, Almirantis Y, Pavlopoulos J, Baskiotis N, Gallinari P, Artiėres T, Ngonga A, Heino N, Gaussier Ė, Barrio-Alvers L, Schroeder M, Androutsopoulos I, Paliouras G. An overview of the BIOASQ large-scale biomedical semantic indexing and question answering competition. BMC Bioinf 2015;16:138.

    Article  Google Scholar 

  25. Wang C, Yan J, Zhou A, He X. Transductive non-linear learning for Chinese hypernym prediction. Meeting of the Association for Computational Linguistics, pp. 1394–1404; 2017.

  26. Xiao S, Yan J, Chu SM, Yang X, Zha H. Modeling the intensity function of point process via recurrent neural networks. AAAI; 2017.

  27. Yan J, Yin X, Lin W, Deng C, Zha H, Yang X. A short survey of recent advances in graph matching. ACM on International Conference on Multimedia Retrieval, pp. 167–174; 2016.

  28. Zhang Z, Liu T, Zhang B, Li Y, Zhao CH, Feng S, Yin X, Zhou F. A generic retrieval system for biomedical literatures: USTB at bioasq 2015 question answering task. Working Notes of CLEF 2015 - Conference and Labs of the Evaluation forum, Toulouse, France, September 8-11, 2015; 2015.

  29. Zweigenbaum P. Question answering in biomedicine. Proceedings Workshop on Natural Language Processing for Question Answering, EACL, vol. 2005, pp. 1–4. Citeseer; 2003.

Download references

Funding

The research was partly supported by National Natural Science Foundation of China (61473036).

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Xu-Cheng Yin.

Ethics declarations

Conflict of Interest

The authors declare that they have no conflict of interest.

Ethical Approval

This article does not contain any studies with human participants or animals performed by any of the authors.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zhang, BW., Yin, XC. SSDM2: a Two-Stage Semantic Sequential Dependence Model Framework for Biomedical Question Answering. Cogn Comput 10, 73–83 (2018). https://doi.org/10.1007/s12559-017-9525-x

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s12559-017-9525-x

Keywords

Navigation