Template-based Question Answering using Recursive Neural Networks

Publikation: Beiträge in SammelwerkenAufsätze in KonferenzbändenForschungbegutachtet

Standard

Template-based Question Answering using Recursive Neural Networks. / Athreya, Ram G.; Bansal, Srividya K.; Ngomo, Axel Cyrille Ngonga et al.
Proceedings - 2021 IEEE 15th International Conference on Semantic Computing, ICSC 2021. Institute of Electrical and Electronics Engineers Inc., 2021. S. 195-198 9364639 (Proceedings - 2021 IEEE 15th International Conference on Semantic Computing, ICSC 2021).

Publikation: Beiträge in SammelwerkenAufsätze in KonferenzbändenForschungbegutachtet

Harvard

Athreya, RG, Bansal, SK, Ngomo, ACN & Usbeck, R 2021, Template-based Question Answering using Recursive Neural Networks. in Proceedings - 2021 IEEE 15th International Conference on Semantic Computing, ICSC 2021., 9364639, Proceedings - 2021 IEEE 15th International Conference on Semantic Computing, ICSC 2021, Institute of Electrical and Electronics Engineers Inc., S. 195-198, 15th IEEE International Conference on Semantic Computing - ICSC 2021, Virtual, Laguna Hills, California, USA / Vereinigte Staaten, 27.01.21. https://doi.org/10.48550/arXiv.2004.13843, https://doi.org/10.1109/ICSC50631.2021.00041

APA

Athreya, R. G., Bansal, S. K., Ngomo, A. C. N., & Usbeck, R. (2021). Template-based Question Answering using Recursive Neural Networks. In Proceedings - 2021 IEEE 15th International Conference on Semantic Computing, ICSC 2021 (S. 195-198). Artikel 9364639 (Proceedings - 2021 IEEE 15th International Conference on Semantic Computing, ICSC 2021). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.48550/arXiv.2004.13843, https://doi.org/10.1109/ICSC50631.2021.00041

Vancouver

Athreya RG, Bansal SK, Ngomo ACN, Usbeck R. Template-based Question Answering using Recursive Neural Networks. in Proceedings - 2021 IEEE 15th International Conference on Semantic Computing, ICSC 2021. Institute of Electrical and Electronics Engineers Inc. 2021. S. 195-198. 9364639. (Proceedings - 2021 IEEE 15th International Conference on Semantic Computing, ICSC 2021). doi: 10.48550/arXiv.2004.13843, 10.1109/ICSC50631.2021.00041

Bibtex

@inbook{f10ca7e14624471a87cdd54163274ed6,
title = "Template-based Question Answering using Recursive Neural Networks",
abstract = "Most question answering (QA) systems over Linked Data, i.e. Knowledge Graphs, approach the question answering task as a conversion from a natural language question to its corresponding SPARQL query. A common approach is to use query templates to generate SPARQL queries with slots that need to be filled. Using templates instead of running an extensive NLP pipeline or end-to-end model shifts the QA problem into a classification task, where the system needs to match the input question to the appropriate template. This paper presents an approach to automatically learn and classify natural language questions into corresponding templates using recursive neural networks. Our model was trained on 5000 questions and their respective SPARQL queries from the preexisting LC-QuAD dataset grounded in DBpedia, spanning 5042 entities and 615 predicates. The resulting model was evaluated using the FAIR GERBIL QA framework resulting in 0.419 macro f-measure on LC-QuAD and 0.417 macro f-measure on QALD-7.",
keywords = "Question Answering, Recursive Neural Network, Informatics, Business informatics",
author = "Athreya, {Ram G.} and Bansal, {Srividya K.} and Ngomo, {Axel Cyrille Ngonga} and Ricardo Usbeck",
note = "Funding Information: This paper presents a novel approach for the QA over Linked Data task by converting it into a template classification task followed by a slot filling task. Although earlier template-based approaches have attempted similar solutions, this was the first time (to the best of our knowledge) that recursive neural networks were applied to the template classification task. For completeness, a slot filling approach using an ensemble of the best components for named entity, predicate and class recognition tasks were presented. Our evaluation showed that state-of-the-art neural network techniques such as Long Short Term Memory (LSTM), recursive neural networks, and word embeddings be leveraged for the template classification task. We are aware that our approach has a coverage issue in terms of being bound to the training templates and look forward to mitigating this issue through a finer-grained training process. Acknowledgements. We acknowledge the support of the Federal Ministry for Economic Affairs and Energy (BMWi) project SPEAKER (FKZ 01MK20011A). Publisher Copyright: {\textcopyright} 2021 IEEE.; 15th IEEE International Conference on Semantic Computing - ICSC 2021, ICSC 2021 ; Conference date: 27-01-2021 Through 29-01-2021",
year = "2021",
month = jan,
doi = "10.48550/arXiv.2004.13843",
language = "English",
isbn = "978-1-7281-8900-0",
series = "Proceedings - 2021 IEEE 15th International Conference on Semantic Computing, ICSC 2021",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "195--198",
booktitle = "Proceedings - 2021 IEEE 15th International Conference on Semantic Computing, ICSC 2021",
address = "United States",

}

RIS

TY - CHAP

T1 - Template-based Question Answering using Recursive Neural Networks

AU - Athreya, Ram G.

AU - Bansal, Srividya K.

AU - Ngomo, Axel Cyrille Ngonga

AU - Usbeck, Ricardo

N1 - Conference code: 15

PY - 2021/1

Y1 - 2021/1

N2 - Most question answering (QA) systems over Linked Data, i.e. Knowledge Graphs, approach the question answering task as a conversion from a natural language question to its corresponding SPARQL query. A common approach is to use query templates to generate SPARQL queries with slots that need to be filled. Using templates instead of running an extensive NLP pipeline or end-to-end model shifts the QA problem into a classification task, where the system needs to match the input question to the appropriate template. This paper presents an approach to automatically learn and classify natural language questions into corresponding templates using recursive neural networks. Our model was trained on 5000 questions and their respective SPARQL queries from the preexisting LC-QuAD dataset grounded in DBpedia, spanning 5042 entities and 615 predicates. The resulting model was evaluated using the FAIR GERBIL QA framework resulting in 0.419 macro f-measure on LC-QuAD and 0.417 macro f-measure on QALD-7.

AB - Most question answering (QA) systems over Linked Data, i.e. Knowledge Graphs, approach the question answering task as a conversion from a natural language question to its corresponding SPARQL query. A common approach is to use query templates to generate SPARQL queries with slots that need to be filled. Using templates instead of running an extensive NLP pipeline or end-to-end model shifts the QA problem into a classification task, where the system needs to match the input question to the appropriate template. This paper presents an approach to automatically learn and classify natural language questions into corresponding templates using recursive neural networks. Our model was trained on 5000 questions and their respective SPARQL queries from the preexisting LC-QuAD dataset grounded in DBpedia, spanning 5042 entities and 615 predicates. The resulting model was evaluated using the FAIR GERBIL QA framework resulting in 0.419 macro f-measure on LC-QuAD and 0.417 macro f-measure on QALD-7.

KW - Question Answering

KW - Recursive Neural Network

KW - Informatics

KW - Business informatics

UR - http://www.scopus.com/inward/record.url?scp=85102617112&partnerID=8YFLogxK

UR - https://www.mendeley.com/catalogue/930ff2cd-68eb-3440-9c15-60501e8bba0a/

U2 - 10.48550/arXiv.2004.13843

DO - 10.48550/arXiv.2004.13843

M3 - Article in conference proceedings

AN - SCOPUS:85102617112

SN - 978-1-7281-8900-0

T3 - Proceedings - 2021 IEEE 15th International Conference on Semantic Computing, ICSC 2021

SP - 195

EP - 198

BT - Proceedings - 2021 IEEE 15th International Conference on Semantic Computing, ICSC 2021

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 15th IEEE International Conference on Semantic Computing - ICSC 2021

Y2 - 27 January 2021 through 29 January 2021

ER -

DOI

Zuletzt angesehen

Forschende

  1. Timo Janca

Publikationen

  1. Effects of diversity versus segregation on automatic approach and avoidance behavior towards own and other ethnic groups
  2. Different facets of tree sapling diversity influence browsing intensity by deer dependent on spatial scale
  3. The role of task meaning on output in groups
  4. Dynamic capabilities and routinization
  5. Scaffolding Learner Agency in Technology-Enhanced Language Learning Environments
  6. Comparing Web-Based and Blended Training for Coping With Challenges of Flexible Work Designs
  7. Continental mapping of forest ecosystem functions reveals a high but unrealised potential for forest multifunctionality.
  8. End-to-End Active Speaker Detection
  9. Fermentative utilization of coffee mucilage using Bacillus coagulans and investigation of down-stream processing of fermentation broth for optically pure L(+)-lactic acid production
  10. Planar Multipole Resonance Probe: A kinetic model based on a functional analytic description
  11. Timing matters: Distinct effects of nitrogen and phosphorus fertilizer application timing on root system architecture responses
  12. Dealing with inclusion–teachers’ assessment of internal and external resources
  13. Reconfigurable Control System for Plants with Variable Structure
  14. Possible underestimations of risks for the environment due to unregulated emissions of biocides from households to wastewater
  15. Reconciling conservation and development in protected areas of the Global South
  16. Modelling scenarios to identify a combined sediment-water management strategy for the large reservoirs of the Tuyamuyun hydro-complex
  17. Identifying determinants of teachers' judgment (in)accuracy regarding students' school-related motivations using a Bayesian cross-classified multi-level model
  18. Design rules for environmental biodegradability of phenylalanine alkyl ester linked ionic liquids
  19. Empirical research on mathematical modelling
  20. Assessing the costs and cost-effectiveness of ICare internet-based interventions (protocol)
  21. Higher Wages in Exporting Firms
  22. From GUI to No-UI
  23. Model of mobility demands for future short distance public transport systems
  24. Material utilization of organic residues
  25. Demarcating transdisciplinary research in sustainability science—Five clusters of research modes based on evidence from 59 research projects
  26. Brennball
  27. Energy-aware system design for autonomous wireless sensor nodes
  28. Third International Mathematics and Science Study and Trends in Mathematics and Science Studies (TIMSS)
  29. Development of a magnesium secondary alloy system for mixed magnesium post-consumer scrap
  30. Physicochemical properties and biodegradability of organically functionalized colloidal silica particles in aqueous environment
  31. Planning for Sea Spaces I: Processes, Practices and Future Perspectives
  32. The blue-collar brain
  33. Utilization of organic residues using heterotrophic microalgae and insects