AnswerFinder [News] [Members] [Join AnswerFinder] [Related Links] [Publications] [Presentation Slides] [Media Clips]

This work is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 License.

Have you ever been frustrated by the poor search results returned by your favourite search engine? Have you ever wished that your search engine returned not documents, but the exact piece of text that matches your query? Have you ever dreamt of a Web tool that consults the Web to find the specific answer to your question? If you have answered "yes" to any of these questions, then you know why we want to develop AnswerFinder.

AnswerFinder is a new project that is being developed by the Centre for Language Technology at Macquarie University. The goal is to produce a system that is able to find the answer to arbitrary user questions by consulting the information available in text documents.

The main problem that AnswerFinder faces is the representation and fast processing of the semantic contents of the text documents. AnswerFinder uses a variation of the Flat Logical Forms (FLFs) used in ExtrAns, an answer extraction system developed at the University of Zurich.

To improve efficiency, AnswerFinder will combine ExtrAns' technology with methodologies developed for running question-answering systems such as those competing in the QA track of the TREC conference.

News

• [2007/05/04] Pre-release of Afner (0.1), the named entity recogniser of AnswerFinder. If you would like the program send an email to Diego. Include in your email a one-paragraph description of your background and a one-paragraph of what you would like to do with Afner.

Members

• Diego Mollá (project director and chief investigator)
• Robert Dale (chief investigator)
• Menno van Zaanen (research associate)
• Luiz Augusto Pizzato (PhD student)
• Elena Akhmatova (PhD student)
• Daniel Smith (research programmer)
• Stephen Wan (research programmer)

Join AnswerFinder

There are several ways you can Join AnswerFinder:

• As a PhD student: If you have an interesting topic of research that has to do with question answering, send us an email and we will get in touch with you.
• As a Masters/Honours student: If you are a Masters or an Honours student in Macquarie University and you are searching for a project, consult the list of Honours projects. Many of these projects can be adapted to Masters projects.
• As an undergraduate student: If you are a student enrolled in Macquarie Ubiversity you can also work for AnswerFinder and get paid for it. Consult the list of summer projects.

Related Links

• AnswerFinder project programming guidelines
• Daniel Smith's page on Named Entity Recognition
• Luiz Pizzato's page on WWW Question Answering
• ExtrAns' home page at the University of Zurich
  • Demo at the University of Zurich
• TREC web site
• Related Honours Projects
• Related PhD Scholarships
• Other projects at the Centre for Language Technology

Publications

See the AnswerFinder publications stored in Bibsonomy

1. D. Mollá, M. van Zaanen and S. Cassidy. Named Entity Recognition in Question Answering of Speech Data (2007). Proceedings ALTA 2007, 57-65, Melbourne. [poster]
2. L.A. Pizzato and D. Mollá. Question Prediction Language Model (2007). Proceedings ALTA 2007, 92-99, Melbourne.
3. M. van Zaanen and D. Mollá. A Named Entity Recogniser for Question Answering (2007). Proceedings PACLING 2007, 8 pages, Melbourne.
4. M. van Zaanen and D. Mollá. AnswerFinder at QA @ CLEF 2007 (2007). Working Notes for the CLEF 2007 Workshop, 9 pages, Budapest.
5. D. Mollá, S. Cassidy and M. van Zaanen. AnswerFinder at QAst 2007: Named Entity Recognition for QA on Speech Transcripts (2007). Working Notes for the CLEF 2007 Workshop, 9 pages, Budapest.
6. D. Mollá and J. L. Vicedo. Question Answering in Restricted Domains: An Overview (2007). Computational Linguistics, 33(1):41-61.
7. D. Mollá, M. van Zaanen, and D. Smith. Named Entity Recognition for Question Answering (2006). Proceedings ALTW 2006, Sydney.
8. L.A. Pizzato, D. Mollá, and C. Paris. Pseudo Relevance Feedback Using Named Entities for Question Answering (2006). Proceedings ALTW 2006, 8 pages, Sydney.
9. E. Akhmatova and D. Mollá. Recognizing Textual Entailment via Atomic Propositions (2006). In Machine Learning Challenges, LNCS 3944/2006, Springer, pp385-403.
10. D. Mollá and S. Wan. Macquarie University at DUC 2006: Question Answering for Summarisation (2006). Proceedings DUC 2006, 62-69. [poster]
11. D. Mollá. Learning of Graph-based Question Answering Rules (2006). Proceedings HLT/NAACL 2006 Workshop on Graph Algorithms for Natural Language Processing, 37-44.
12. D. Mollá and M. van Zaanen. AnswerFinder at TREC 2005 (2006). The Fourteenth Text REtrieval Conference (TREC 2005). [poster]
13. L.A. Pizzato and D. Mollá. Extracting Exact Answers using a Meta Question Answering System (2005). Proc. ALTW05, Sydney, December 2005 [poster].
14. D. Mollá and M. van Zaanen. Learning of Graph Rules for Question Answering (2005). Proc. ALTW05, Sydney, December 2005.
15. M. van Zaanen, L.A. Pizzato and D. Mollá. Classifying Sentences Using Induced Structure (2005). Proc. Twelfth Edition of the Symposium on String Processing and Information Retrieval (SPIRE2005), Buenos Aires, November 2005.
16. M. van Zaanen, L.A. Pizzato and D. Mollá. Question Classification by Structure Induction (2005). In Leslie Pack Kaelbling (ed.), Proc. Nineteenth International Joint Conference on Artificial Intelligence (IJCAI05), poster presentation, Edinburgh, July 2005. ISBN 0-938075-93-4. [poster]
17. E. Akhmatova. Textual Entailment Resolution via Atomic Propositions (2005). Proc. PASCAL Recognising Textual Entailment Challenge Workshop.
18. D. Mollá and M. Gardiner. AnswerFinder at TREC 2004 (2005). The Thirteenth Text REtrieval Conference (TREC 2004) . [poster]
19. L.A. Pizzato. Using a Trie-based Structure for Question Analysis (2004). Proc. ALTW04, Sydney, December 2004.
20. D. Mollá and M. Gardiner. AnswerFinder - Question Answering by Combining Lexical, Syntactic and Semantic Information (2004). Proc. ALTW04, Sydney, December 2004. [slides]
21. D. Mollá. Hacia el Uso de la Informaci�n Sint�ctica y Sem�ntica en los Sistemas de B�squeda de Respuestas (2004). Procesamiento del Lenguaje Natural, 33:17-24.
22. D. Mollá. AnswerFinder in TREC 2003 (2004). NIST Special Publication 500-255: The Twelfth Text REtrieval Conference (TREC 2003).
23. D. Mollá. Towards Semantic-Based Overlap Measures for Question Answering (2003). Proc. ALTW03, Melbourne, December 2003. [slides]
24. D. Moll� and B. Hutchinson. Intrinsic versus Extrinsic Evaluations of Parsing Systems (2003). Proceedings European Association for Computational Linguistics (EACL), workshop on Evaluation Initiatives in Natural Language Processing. Budapest, 14 April 2003. [slides]
25. D. Mollá and B. Hutchinson. Dependency-based Semantic Interpretation for Answer Extraction (2002). Proc. 2002 Australasian NLP Workshop (ANLP'02), Canberra. [slides]

Presentation Slides

• Graph-based Question Answering
  Presented at the CLT seminar of 30 Aug. 2004.
• Answer me Exactly
  Presented at the CLT seminar of 28 Jun. 2004.
• AnswerFinder in TREC-QA 2003: How did it Go?
  Presented at the CLT seminar of 13 Oct. 2003.
• A Proposal for CLT's Participation in TREC-QA 2003
  Presented at the CLT seminar of 5 May 2003.
• Question Answering: What is the Best Scoring Mechanism?
  Presented at the CLT seminar of 17 Feb. 2003.

Media Clips

• Answering questions Google can't (Australian IT, 14 Dec 2005)