The OKBQA 2017 has successfully finished. We would like to thank everyone who attended.
Introduction
The huge and rapidly increasing amount of structured and unstructured data available on the Web makes it both possible and necessary to support users in finding relevant information. The trend moves more and more towards smart knowledge services that are able to find information, aggregate them, draw inferences, and present succinct answers without requiring the user to wade through a large number of documents. The novel avenues made possible by knowledge services are numerous and diverse, including ubiquitous information access (from smartphones, tablets, smart watches, etc.), barrier-free access to data (especially for the blind and disabled) and knowledge discovery.
Over the last years, several challenges and calls for research projects have pointed out the dire need for pushing natural language interfaces. In this context, the importance of Semantic Web data as a premier knowledge source is rapidly increasing. But we are still far from having accurate natural language interfaces that allow handling complex information needs in a user-centric and highly performant manner. The development of such interfaces requires the collaboration of a range of different fields, including natural language processing, information extraction, knowledge base construction and population, reasoning, and question answering.
The main goal of this workshop is to join forces in the collaborative development of open frameworks for knowledge extraction and question answering, to share standards, and to foster the creation of an ecosystem of tools and benchmarks. The workshop will therefore not only comprise short and long paper presentations but also a hands-on session on already existing frameworks, standards, and benchmarking campaigns, as well as a social meet-up.
Invited Speakers
Hideyuki Shibuki, Yokohama National University, JapanTitle: Challenges in QA for University Entrance Exam at NTCIR QA-Lab: From Multiple-Choice to Essay Questions
Panel Discussion
Joint Panel Discussion with KG4IR (in the KG4IR workshop room)
Keynote: Bast / Uni-Freiburgfeaturing:
- Hannah Bast, Universitaet Freiburg, Germany
- Noriko Kando, National Institute of Informatics, Japan
- Jaap Kamps, University of Amsterdam, The Netherlands
- Edgar Meij, Bloomberg L.P., U.K.
- Bogdan Arsintescu, LinkedIn, U.S.A
- David Carmel, Yahoo!, Israel
Proceedings
The proceedings of OKBQA 2017.(PDF, 7.1MB)Program
Camera-ready papers are available from the links in the table.
Friday, August 11, 2017 |
|
9:00 | Opening |
9:00–10:30 | Session I | 9:00– 9:30 | Controlling Expressiveness of Question Interpretation with a Constrained Tree Transducer Induction Pascual Martínez-Gómez and Yusuke Miyao |
9:30–10:00 | TaxoPhrase: Exploring Knowledge Base via Joint Learning of Taxonomy and Topical Phrases Weijing Huang, Wei Chen, Tengjiao Wang and Shibo Tao |
10:00–10:30 | Multilingualization of Question Answering Using Universal Dependencies Hiroshi Kanayama and Koichi Takeda |
10:30–10:50 | Coffee Break |
10:50–12:20 | Session II |
10:50– 11:30 | Invited Talk: Challenges in QA for University Entrance Exam at NTCIR QA-Lab: From Multiple-Choice to Essay Questions Hideyuki Shibuki, Yokohama National University, Japan |
11:30–12:00 | Wikipedia Based Essay Question Answering System for University Entrance Examination Takaaki Matsumoto, Francesco Ciannella, Fadi Botros, Evan Chan, Cheng-Ta Chung, Keyang Xu, Tian Tian and Teruko Mitamura |
12:00–12:20 | Demo Introduction (3 presentations. 6 min. each) |
12:20–14:00 | Lunch Break
A lunch during the workshop day at SIGIR 2017 will be provided. The venue will be most probably "Hana" at 4F (Workshops are 42/43F) of Keio Plaza Hotel. |
14:00–15:30 | Session III |
14:00–14:30 | Applying Linked Open Data to Machine Translation for Cross-lingual Question Answering Takaaki Matsumoto and Teruko Mitamura |
14:30–15:00 | Chronological and Geographical Measures for Evaluation of World History Essay QA in University Entrance Exams Hideyuki Shibuki, Kotaro Sakamoto, Madoka Ishioroshi, Akira Fujita, Yoshinobu Kano, Teruko Mitamura, Tatsunori Mori and Noriko Kando |
15:00–15:30 | Demonstrations (3 systems) |
|
|
15:30–15:50 | Coffee Break |
15:50–17:20 | Joint Panel Discussion with KG4IR (in the KG4IR workshop room) |
Keynote: Bast / Uni-Freiburg Joint Discussion Panel with KG4IR workshop, featuring:
|
Paper Submissions
OKBQA 2017 solicits the submission of original research papers in two types:
- Full paper submissions (up to 8 pages, plus two extra pages for references) must describe substantial and original work.
- Short paper submissions (up to 4 pages, plus two extra pages for references) must describe an original work which may present a small, focused contribution, a work in progress, or an interesting application case.
Please use the ACM proceedings style file http://www.acm.org/publications/proceedings-template
It has LaTeX files, Microsoft Word templeate file, and a sample PDF file.
Before submitting your paper please ensure you have read the Instructions for Authors and that your paper uses the prescribed style files. To submit your work, please use the submission page at the following Easychair address https://easychair.org/conferences/?conf=okbqa2017.
At least one author of each paper is expected to register for the workshop and attend to present the paper.
Demo Submissions
The OKBQA2017 workshop, to be held in Tokyo, 11th August, invites proposals for system demonstrations. Submissions may range from early prototypes to mature production-ready systems. Areas of interest include all topics related to KB construction and QA. Please refer to the workshop page for a full lists topics of interest.
Submission Deadline: 5, July 2017
A proposal may be an extended abstract upto two pages including references, and is expected to include:
- summary of the system with an emphasis on innovative features,
- reference to published papers, if any,
- URL of a working system, if any,
- URL of screencast, if any,
- URL of project page, e.g., GitHub project, if any
Note that a proposal should deliver sufficient details about the system to allow evaluation of its quality, relevance to the workshop, and so on.
Please use the ACM proceedings style file http://www.acm.org/publications/proceedings-template
It has LaTeX files, Microsoft Word templeate file, and a sample PDF file.
Before submitting your paper please ensure you have read the Instructions for Authors and that your paper uses the prescribed style files. To submit your work, please use the submission page at the following Easychair address https://easychair.org/conferences/?conf=okbqa2017.
Accepted submissions will be included in the OKBQA2017 workshop proceedings (as abstracts), and will be presented during the workshop.
Topics
Specific topics include but are not limited to:
-
Natural Language Interface for Web of Data
- Browsing Linked Data
- Question Answering over Linked Data
- Benchmarking Natural Language Interfaces
- Term Matching and Entity Disambiguation
- SPARQL Query Pattern Generation
- Schema-agnostic Query Generation
- Discovery of Linked Data Sources
- Endpoint Profiling
- Dealing with Data and Schema Heterogeneity
- Providing Justifications of Answers and Conveying Trust
- Knowledge Base Design for Question Answering
- Language Resources and NLP Software for Question Answering
- Reasoning for Question Answering
- Natural Language Querying of RDF exposed as Linked Data
- Natural Language Querying of Web Services
- User Feedback and Interaction
- Dialogue Systems
- Personal Assistants
-
Knowledge Base Construction for Question Answering
- Entity-centered Knowledge Bases
- Event-centered Knowledge Bases
- Human Intervention of Knowledge with QA
- NLP Annotations for Knowledge Extraction and Machine Reading
- Gold Standard Data Sets and Quality Assessment
- Indexing and Mappings from Existing Sources (Structured, Semi-structured or Unstructured)
-
NLP Annotation Framework for Knowledge Base and Question Answering
-
NLP annotation framework
- to represent the layers and their structure of language understanding in graph formats such as XML and RDF, including metadata and underlying ontologies
-
NLP annotation knowledge base
- to access existing annotated NLP resources efficiently and effectively for further processing for knowledge base engineering and question answering
-
NLP annotation framework
Important Dates
- Submission deadline (paper): 27, June 2017 (EXTENDED) (Anywhere on Earth)
- Submission deadline (demo): 5, July 2017(Anywhere on Earth)
- Notification of acceptance: 17 July, 2017 (EXTENDED)
- Camera-ready versions of accepted papers: 24 July, 2017 (EXTENDED)
- Workshop date: 11 Augst, 2017
Organizers
- Key-Sun Choi, KAIST, Korea
- Teruko Mitamura, Carnegie Mellon University, USA
- Piek Vossen, Vrije Universiteit Amsterdam, Netherlands
- Jin-Dong Kim, Database Center for Life Science (DBCLS), Japan
- Axel-Cyrille Ngonga Ngomo, Universität Leipzig, Germany
Programme Committee
- André Freitas (University of Passau, Germany)
- Axel-Cyrille Ngonga Ngomo (University of Leipzig, Germany)
- Christina Unger (Universität Bielefeld, Germany)
- Eun-Kyung Kim (KAIST, Korea)
- Jin-Dong Kim (Database Center for Life Science (DBCLS), Japan)
- Key-Sun Choi (KAIST, Korea)
- Piek Vossen (Vrije Universiteit Amsterdam, Netherlands)
- Pum-mo Ryu (Busan University of Foreign Studies, Korea)
- Ricardo Usbeck (University of Leipzig, Germany)
- Jun Araki (Carnegie Mellon University, USA)
- Peter Clark (Allen Institute for AI, USA)
- Eduard Hovy (Carnegie Mellon University, USA)
- Madoka Ishioroshi (National Institute of Informatics, Japan)
- Hiroshi Kanayama (IBM Research - Tokyo, IBM Japan)
- Bernardo Magnini (Fondazione Bruno Kessler, Italy)
- Tatsunori Mori (Yokohama National University, Japan)
- Yuta Nakashima (Osaka University, Japan)
- Eric Nyberg (Carnegie Mellon University, USA)
- Anselmo Peñas (UNED, Spain)
- John M. Prager (IBM T.J. Watson Research Center, USA)
- Kotaro Sakamoto (Yokohama National University, Japan)
- Hideyuki Shibuki (Yokohama National University, Japan)
- Hideki Shima (Duolingo, Inc., USA)
- Koichi Takeda (IBM Research - Tokyo, IBM Japan)
- Chuan-Jie Lin (National Taiwan Ocean University, Taiwan)
- Di Wang (Carnegie Mellon University, USA)
- Luke Zettlemoyer (University of Washington, USA)