Dr Sandra WilliamsResearch Fellow in Natural Language Generation
- Generating Maths Word Problems and Solutions I am investigating automatic generation of mathematical word problems with hints for solution methods in easy-to-follow steps, similar to those produced by basic-skills tutors, see Math Stories. I am keen to collaborate on this research with psychologists, maths teachers, educational researchers, and intelligent tutoring researchers.
- Rendering the semantic web accessible to people who want to create or modify semantic content but do not know ontology languages and logics. My role in the SWAT project was to provide natural language support for ontology verbalisation and authoring. I developed grammars and discourse structuring algorithms for the SWAT verbaliser and a tool to extract ontology identifier names from text.
- Computer-generation of numerical quantity expressions. Numerical quantities are often presented poorly (especially in the media) which frustrates numerate people who are not given enough information while at the same time it perplexes innumerate people who do not understand basic mathematical concepts. I believe that it is possible to vary descriptions of numerical quantities to suit different audiences, situations and writing styles. I recently completed an ESRC Small Project, NumGen, to investigate feasibility and potential areas for future research. I am currently seeking opportunities for follow-up project proposals.
Other interests are discourse structure and microplanning, generation for people with limited literacy, discourse analysis, speech act analysis, intonation analysis, generation of prosodically-annotated text, dialogue management in spoken language systems and automatic text summarisation.
I am a member of the Open University Natural Language Generation (NLG) research group.
- Ph.D. Natural Language Generation (NLG) of discourse relations for different reading levels. University of Aberdeen, 2004
- M.Phil. Computer Speech and Language Processing. University of Cambridge, 1991
- B.A. (Hons.) Artificial Intelligence and Computing. University of Sussex, 1985
I am currently funded by an Open University Pro-Vice Chancellor Fellowship for generating mathematical word problems and step-by-step hints for solving them, see Maths Stories.
- Sharon Moyo, Ph.D. student (co-supervisor)
- Tu Anh Nguyen, Ph.D. student (co-supervisor)
- Stephen Taylor, M.Sc. student (specialist advisor)
- Susana Bautista Blasco, Summer Intern. 2010 (supervisor)
Grants and Awards
- 2008 ESRC Small Grant, £70,000, Generating Intelligent Descriptions of Numerical Quantities for People with Different Levels of Numeracy (NumGen). ESRC Ref. RES-000-22-2760
- 2000 EPSRC PhD Studentship, Aberdeen University.
- 1999 DTI SMART Award, £40,000, Speech and Language Technology for Basic Skills Training Applications (jointly with C. Webb, CTAD Ltd.)
- 1998 Small Research Grant, AU$7,000, Macquarie University, Prosodic Annotation of a Corpus of Route Descriptions.
- 1990 British Telecommunications Training Award, for MPhil at the University of Cambridge
Membership of Programme Committees and Review Panels
- RANLP 2013 Recent Advances in Natural Language Processing, Hissar, Bulgaria.
- ENLG 2013 14th European Workshop on Natural Language Generation, Sofia, Bulgaria.
- DO 2013 International Workshop on Definitions in Ontologies in conjunction with 4th International Conference on Biomedical Ontology (ICBO 2013), Montreal, Canada.
- Patient Education and Counseling Journal, 2012.
- COLING 2012, Natural Language Generation area, December, Mumbai India.
- QG2012 Fifth International Workshop on Question Generation, was to be held as the 11th International Conference on Intelligent Tutoring Systems (ITS2012) but was cancelled.
- NAACL-SRW 2012 Student Research Workshop and Doctoral Consortium
- ENLG 2011 13th European Workshop on Natural Language Generation, 28-30 September, Nancy, France.
- RANLP 2011 Recent Advances in Natural Language Processing, 12-14 September, Hissar, Bulgaria.
- Dialogue and Discourse (open access journal), Special Issue on Question Generation 2011
- ACL-HLT 2011 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies Portland, Oregon, USA.
- ICL 2010 International Conference on Interactive Computer Aided Learning, Hasselt, Belgium.
- INLG 2010 6th International Natural Language Generation Conference, Dublin.
- MOG 2010 3rd Workshop on Multimodal Output Generation, Dublin.
- EACL 2009 12th European Chapter of the Association for Computational Linguistics, Athens.
- Transactions on Information Systems The official journal of the ACM.
- INLG 2008 5th International Natural Language Generation Conference, Ohio.
- ICL 2008 Interactive Computer Aided Learning, Special Track on Computer-based Knowledge & Skill Assessment and Feedback, Villach.
- RANLP 2007 International Conference on Recent Advances in NLP, Borovets.
- JoLLi Journal of Logic, Language and Information, Special Issue on Coherence in Dialogue and Generation, 16:4, 2007.
- ESSLLI 2006 18th European Summer School in Logic, Language and Information, Malaga. COLING 2002 19th International Conference on Computational Linguistics, Taipei.
- ACL/EACL 1997 Association for Computational Linguistics, Madrid.
- 2012 Seminar at University of Aberdeen, Computing Department.
- 2009 Seminar at Macquarie University, Computing Department.
- 2008 Seminar at Aberdeen University, Computing Department.
- 2007 Seminar at National Informatics Institute, Tokyo, Japan.
- 2005 Seminar at Aberdeen University, Centre for Linguistics Research.
- 2002 Seminar at Monash University, Computer Science Department.
- 2002 Seminar at Stirling University, English Studies Department.
Williams, Sandra and Richard Power (2013) Hedging and Rounding in Numerical Expressions. Accepted for publication in the Journal of Pragmatics and Cognition, John Benjamins Publishing Co.
Susana Bautista, Raquel Hervás, Pablo Gervás, Richard Powerand Sandra Williams (2013) A System for the Simplification of Numerical Expressions at Different Levels of Understandability. Proceedings of the workshop on Natural Language Processing for Improving Textual Accessibility (NLP4ITA), Atlanta, USA, pp.10-19. pdf
Nguyen, Tu Anh T.; Power, Richard; Piwek, Paul and Williams, Sandra (2013). Predicting the understandability of OWL inferences. In: Extended Semantic Web Conference 2013 (ESWC 2013) - Research Track, 26th May to 30th May, 2013, Montpellier, France. pdf
Sandra Williams (2013) An Information Extraction System for English Ontology Identifier Names. Technical Report TR2013/03, Department of Computing, The Open University. pdf (360KB)
Sandra Williams (2013) An Analysis of POS Tag Patterns in Ontology Identifiers and Labels. Technical Report TR2013/02, Department of Computing, The Open University. pdf (385KB)
Nguyen, Tu Ahn; Power, Richard; Piwek, Paul and Williams, Sandra (2012). Measuring the understandability of deduction rules for OWL. In: First International Workshop on Debugging Ontologies and Ontology Mappings (WoDOOM12), 8th October 2012, Galway, Ireland. pdf (313KB)
Nguyen, Tu Anh T.; Power, Richard; Piwek, Paul and Williams, Sandra (2012). Planning accessible explanations for entailments in OWL ontologies. In: 7th International Natural Language Generation Conference (INLG 2012), 30 May-1 June 2012, Utica, IL, USA. pdf (273KB)
Robert Stevens, James Malone, Sandra Williams, Richard Power, Alan Third (2011) Automating generation of textual class definitions from OWL to English. Journal of Biomedical Semantics, 2011 May 17, Vol. 2 Suppl 2:S5. pdf (1,278KB)
Sandra Williams, Allan Third and Richard Power (2011) Levels of organisation in ontology verbalisation. Proceedings of the 13th European Workshop on Natural Language Generation, September 2011, Nancy, France, pp. 158-163. Paper: pdf (113KB). Poster: pdf (1628KB).
Sandra Williams (2011) Generating Mathematical Word Problems. Proceedings of the AAAI Fall Symposium on Question Generation, November 2011, Arlington, VA, USA, pp. 61-64. Paper: pdf (446KB). Slides: pdf (387KB).
Susana Bautista, Raquel Hervás, Pablo Gervás, Richard Power, Sandra Williams (2011) Experimental Identification of the Use of Hedges in the Simplification of Numerical Expressions. The Second Workshop on Speech and Language Processing for Assistive Technologies (SLPAT), 30 July 2011, Edinburgh, Scotland. pdf (170KB)
Susana Bautista, Raquel Hervás, Pablo Gervás, Richard Power, Sandra Williams (2011) How to Make Numerical Information Accessible: Experimental Identification of Simplification Strategies. In Part 1 Proceedings of 13th IFIP TC 13 International Conference on Human-Computer Interaction (INTERACT 2011), P. Campos, N. Graham, J. Jorge, N. Nunes, P. Palanque and M. Winckler (Eds.), Lecture Notes in Computer Science (LNCS) Volume 6946, pp. 57-64, September 5-9, DOI: 10.1007/978-3-642-23774-4. pdf (163KB)
Tu Anh T. Nguyen, Richard Power, Paul Piwek and Sandra Williams (2011) Justification Patterns for OWL DL Ontologies. Technical Report TR2011-05, Department of Computing, The Open University. ISSN 1744-1986. pdf (288KB)
Allan Third, Sandra Williams , Richard Power, (2011) OWL to English: a tool for generating organised easily-navigated hypertexts from ontologies. In: 10th International Semantic Web Conference (ISWC 2011), 23 - 27 Oct 2011, Bonn, Germany. Paper: pdf (208KB). Poster: pdf (2271KBG)
Ehud Reiter and Sandra Williams  Generating Texts in Different Styles. In S Argamon, K Burns, S Dubnov (eds.) The Structure of Style: Algorithmic Approaches to Manner and Meaning. Springer-Verlag. Preprint, pdf (219KB)
Sandra Williams and Richard Power  Grouping axioms for more coherent ontology descriptions. Proceedings of the 6th International Natural Language Generation Conference (INLG 2010), pp. 197-202 Dublin. pdf (116KB)
Robert Stevens, James Malone, Sandra Williams and Richard Power  Automating Class Definitions From OWL to English. Bio-Ontologies 2010: Semantic Applications in Life Sciences SIG at 18th Annual International conference on Intelligent Systems for Molecular Biology (ISMB 2010), Boston. pdf(187KB)
Sandra Williams and Richard Power  A fact-aligned corpus of numerical expressions. Proceedings of the Seventh conference on International Language Resources and Evaluation (LREC 2010), Malta, pp. 2396-2400. pdf (347KB)
Sandra Williams and Richard Power [in preparation] Hedging and rounding in numerical expressions.
Sandra Williams and Richard Power  Precision and mathematical form in first and subsequent mentions of numerical facts and their relation to document structure. Proceedings of 12th European Workshop on Natural Language Generation, Athens, March 30th - 31st. pdf (39KB)
Sandra Williams and Ehud Reiter  SkillSum: basic skills screening with personalised, computer-generated feedback. Interactive Computer Aided Learning (ICL 2008), Special Track on Computer-based Knowledge & Skill Assessment and Feedback in Learning Settings (CAF 2008), September 2008, pages 1-8. pdf (366KB)
Sandra Williams and Richard Power  Deriving rhetorical complexity data from the RST-DT Corpus. Proceedings of the 6th Language Resources and Evaluation Conference (LREC 2008), Marrakech, Morocco, 28-30 May, 2008. pdf (520KB). Additional data in fig. 3 of our poster pdf (101KB)
Sandra Williams, Richard Power and Paul Piwek  Simulating emotional reactions in medical dramas. Proceedings of the Symposium on Affective Language in Human and Machine, Volume 2, The Society for the Study of Artificial Intelligence and the Simulation of Behaviour (AISB 2008) Convention: Communication, Interaction and Social Intelligence, Aberdeen, April 2008, pp. 25-32. pdf (215KB)
Ehud Reiter and Sandra Williams  Three Approaches to Generating Texts in Different Styles. Proceedings of the Symposium on Style in text: creative generation and identification of authorship, Volume 7, The Society for the Study of Artificial Intelligence and the Simulation of Behaviour (AISB 2008) Convention: Communication, Interaction and Social Intelligence, Aberdeen, April 2008, pp. 26-33. pdf (190KB)
Sandra Williams, Paul Piwek and Richard Power  Generating monologue and dialogue to present personalised medical information to patients. Proceedings of the 11th European Workshop on Natural Language Generation (ENLG'07), pp. 167-170. pdf (401KB)
Paul Piwek, Richard Power and Sandra Williams . Generating scripts for personalised medical dialogues for patients. Technical Report 2006/06. Computing Department, The Open University. ISSN 1744-1986. pdf (87KB)
Åhlfeldt, H., Borin, L., Daumke, P., Grabar, N., Hallett, C., Hardcastle, D., Kokkinakis, D., Mancini, C., Markó, K., Merkel, M., Pietsch, C., Power, R., Scott, D., Silvervarg, A., Toporowska Gronostaj, M., Williams, S., Willis, A. . Literature review on patient-friendly documentation systems. Technical Report no. 2006/04. Department of Computing, Faculty of Mathematics and Computing, The Open University. ISSN 1744-1986. pdf (706KB)
Ehud Reiter, Sandra Williams and Lesley Crichton  Generating Feedback Reports for Adults Taking Basic Skills Tests. Proceeding of the The Twenty-fifth SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence, Cambridge, UK. In A Macintosh, R Ellis, and T Allen (ed) Applications and Innovations in Intelligent Systems XIII (Proceedings of ES-05), pages 50-63. pdf (112KB)
Sandra Williams and Ehud Reiter  Generating readable texts for readers with low basic skills. Proceeding of the 10th European Workshop on Natural Language Generation, Aberdeen, pages 140-147. pdf (118KB)
Sandra Williams and Ehud Reiter  Appropriate Microplanning Choices for Low-Skilled Readers. Proceedings of the Nineteenth International Joint Conference on Artificial Intelligence, Edinburgh, pages 1704-1708. pdf (93KB)
Sandra Williams and Ehud Reiter  Deriving content selection rules from a corpus of non-naturally occurring documents for a novel NLG application. Proceedings of the Workshop on Using Corpora for Natural Language Generation, Information Technology Research Institute (ITRI) Technical Report, ITRI-05-03, University of Brighton, pages 41-48. pdf (157KB)
Sandra Williams and Ehud Reiter  Reading errors made by skilled and unskilled readers: evaluating a system that generates reports for people with poor literacy. Poster at the Fourteenth Annual Meeting of the Society for Text and Discourse, Chicago. poster pdf (327KB), paper pdf (42KB)
Sandra Williams and Ehud Reiter  Reading errors made by skilled and unskilled readers: evaluating a system that generates reports for people with poor literacy. University of Aberdeen Department of Computing Science Technical Report AUCS/TR0407, pages 1-6. pdf (123KB)
Sandra Williams  Natural Language Generation (NLG) of discourse relations for different reading levels. PhD Thesis, University of Aberdeen.
Sandra Williams  Language choice models for microplanning and readability. Proceedings of the Student Workshop of the Human Language Technology and North American Chapter of the Association for Computational Linguistics Conference (HLT-NAACL03 Student Workshop), Edmonton, pp. 13-18, May 2003. pdf (62KB)
Sandra Williams, Ehud Reiter and Liesl Osman  Experiments with discourse-level choices and readability. Proceedings of the 9th European Workshop on Natural Language Generation, Budapest, pp. 127-134, April 2003. pdf (144KB)
Ehud Reiter, Somayajulu Sripada and Sandra Williams  Acquiring and Using Limited User Models in NLG. Proceedings of the 9th European Workshop on Natural Language Generation, Budapest, pp. 87-94, April 2003. pdf (105KB)
Sandra Williams and Ehud Reiter  A corpus analysis of discourse relations for Natural Language Generation. Proceedings of Corpus Linguistics 2003, pp. 899-908, Lancaster University, March 2003. pdf (277KB)
Sandra Williams  Natural language generation of discourse connectives for different reading levels. The UK special interest Group for computational linguistics, 5th Annual CLUK. Research Colloquium, Leeds. pdf (248KB)
Sandra Williams and Catherine I. Watson  A Profile of the Discourse and Intonational Structures of Route Descriptions. Proceedings of the 6th European Conference on Speech Communication and Technology, Eurospeech'99, September 5-9, 1999, Budapest, Hungary, Volume 4, pp. 1659-1662. pdf (43KB)
Sandra Williams  Generating Pitch Accents in a Concept-To-Speech System Using a Knowledge Base. Proceedings of the 5th International Conference on Spoken Language Processing, ICSLP'98), Volume 4, pp. 1159-1162, Sydney, Australia, 30th November - 4th December 1998. pdf (126KB)
Robert Dale, Stephen Green, Maria Milosavljevic, Cécile Paris, Cornelia Verspoor and Sandra Williams  Dynamic Document Delivery: Generating Natural Language Texts on Demand. 9th International Conference and Workshop on Database and Expert Systems Applications. August 24-28, Vienna, Austria. pdf (325KB)
Robert Dale, Stephen Green, Maria Milosavljevic, Cécile Paris, Cornelia Verspoor and Sandra Williams  Using Natural Language Generation Techniques to Produce Virtual Documents. Proceedings of the Third Australian Document Computing Symposium, August 21st, Sydney, Australia. pdf (477KB)
Robert Dale, Stephen Green, Maria Milosavljevic, Cécile Paris, Cornelia Verspoor and Sandra Williams  The Realities of Generating Natural Language from Databases. 11th Australian Joint Conference on Artificial Intelligence, 12-17 July, Brisbane, Australia. pdf (444KB)
Cornelia Verspoor, Robert Dale, Stephen Green, Maria Milosavljevic, Cécile Paris, and Sandra Williams  Intelligent Agents for Information Presentation: Dynamic Description of Knowledge Base Objects. In Proceedings of the International Workshop on Intelligent Agent on the Internet and Web, Mexico City, Mexico, 16-20 March 1998, pp. 75-86. pdf (404KB)
Sandra Williams, Mark Harvey and Keith Preston  Rule-based reference resolution for unrestricted text using part-of-speech tagging and noun phrase parsing. Discourse Anaphora and Anaphor Resolution Colloquium (DAARC), Lancaster, U.K., July 1996 pdf (93KB)
Sandra Williams  Anaphoric reference and ellipsis resolution in a telephone-based spoken language system for accessing email. Discourse Anaphora and Anaphor Resolution Colloquium (DAARC), Lancaster, U.K., July 1996. Also in Simon Botley and Anthony McEnery (eds.) Corpus-based and Computational Approaches to Discourse Anaphora, John Benjamins Publishing Company, ISBN 902722272X, 2000. pdf (73KB)
Sandra Williams  Dialogue management in a mixed-initiative, cooperative, spoken language system. 11th Twente Workshop on Language Technology (TWLT11) Dialogue Management in Natural Language Systems, Enschade, Netherlands, June 1996 pdf (101KB)
Peter Wyard, Alison Simons, Steve Appleby, Edward Kaneen, Sandra Williams and Keith Preston  Spoken Language Systems. BT Technology Journal, January 1996. pdf (444KB)
Peter Wyard, Steven Appleby, Edward Kaneen, Sandra Williams and Keith Preston  A Combined Speech and Visual Interface to the BT Business Catalogue. ESCA Workshop on Spoken Dialogue Systems, 30th May - 2nd June 1995
Keith Preston and Sandra Williams  Managing the Information Overload. Physics in Business, Institute of Physics, June 1994 pdf (20KB)
Here is one way to Calculate a 20% Discount, an animation that I created with GoAnimate.com.
Poor Bridget wants a cup of coffee but she only has $1...
To calculate how much you pay with a 20% discount, divide the full price by 10 and multiply by 8. Why? With a 20% discount, you pay 80% of the full price. If the full price is $1.20, divide by 10 to get 10% (or one tenth). That gives us $0.12. Then multiply by 8 to get 80%. So you pay $0.96
Maths Word Problem, an animation from GoAnimate.com that I created.
A computer-generated medical drama
The script for this drama was generated by a computer using natural language generation technology. It is acted by embodied agents. Watch it here
- Page updated: May 2013, by Sandra Williams.
- CSS Copyright © 2007 Simpletex. All Rights Reserved | Designed by Free CSS Templates