Question answering over linked data: What is difficult to answer? What affects the F scores?

Research output: Contributions to collected editions/worksArticle in conference proceedingsResearchpeer-review

Authors

  • Muhammad Saleem
  • Samaneh Nazari Dastjerdi
  • Ricardo Usbeck
  • Axel Cyrille Ngonga Ngomo

We present a fine-grained analysis of the Question Answering over Linked Data (QALD-6) challenge. We divide the QALD-6 questions into 8 main categories and compare state-of-the-art questions answering (QA) systems over Linked Data against the individual categories. We show the difficulty (in terms of overall F scores of the QA systems) of each category. We show the effect of various natural language and SPARQL features such as the number of triple patterns, number of keywords, the answer size, the type of answers, the effect of aggregate functions, and the SPARQL query forms on the overall F scores of the QA systems.

Original languageEnglish
Title of host publicationJoint Proceedings of BLINK2017: 2nd International Workshop on Benchmarking Linked Data and NLIWoD3 : Natural Language Interfaces for the Web of Data
EditorsRcardo Usbeck, Axel-Cyrille Ngonga Ngomo, Jin-Dong Kim, Key-Sun Choi, Philipp Cimiano, Irini Fundulaki, Anastasia Krithara
Volume1932
PublisherSun Site Central Europe (RWTH Aachen University)
Publication date2017
Publication statusPublished - 2017
Externally publishedYes
EventJoint BLINK 2017: 2nd International Workshop on Benchmarking Linked Data and NLIWoD3: Natural Language Interfaces for the Web of Data, BLINK 2017-NLIWoD3 2017 - Vienna, Austria
Duration: 21.10.201722.10.2017
https://dblp.org/db/conf/semweb/blink2017.html

Recently viewed

Publications

  1. Integrating methods for ecosystem service assessment
  2. Integrating Common Ground and Informativeness in Pragmatic Word Learning
  3. Analytics and Intuition in the Process of Selecting Talent
  4. Efficacy of a web-based intervention with and without guidance for employees with risky drinking
  5. Study of Single Filament Dielectric Barrier Discharge in Argon
  6. Influence of three different unstable shoe constructions on EMG-activity during treadmill walking
  7. Using rating scales for the assessment of physical self-concept
  8. Discourse, practice, policy and organizing
  9. Action Errors, Error Management, and Learning in Organizations
  10. FaQuAD
  11. Recurring patterns and blueprints of industrial symbioses as structural units for an it tool
  12. Principled Interpolation in Normalizing Flows
  13. Facing Up to Third Party Liability for Space Activities
  14. Learning with summaries
  15. Optimization of a gaseous multitube detector for soft X-ray detection
  16. Simon Denny
  17. Comparison of Backpropagation and Kalman Filter-based Training for Neural Networks
  18. Extraterritorial Asylum Processing. The Libya-Niger Emergency Transit Mechanism
  19. Assessing empirical research on value-based management
  20. Towards a Comprehensive Framework for Environmental Management Accounting
  21. Early-Career Researchers’ Perceptions of the Prevalence of Questionable Research Practices, Potential Causes, and Open Science
  22. Is Code Law? Kritik in Zeiten algorithmischer Gouvernementalität
  23. ℓp-norm multiple kernel learning
  24. How much can we learn about voluntary climate action from behavior in public goods games?
  25. Disentangling trade-offs and synergies around ecosystem services with the influence network framework
  26. Using Daily Stretching to Counteract Performance Decreases as a Result of Reduced Physical Activity—A Controlled Trial
  27. Halb voll oder halb leer?
  28. When Testing Becomes Learning—Underscoring the Relevance of Habituation to Improve Internal Validity of Common Neurocognitive Tests