Smoking Termination with cOmputerised Personalisation (STOP)

** To try out the online version of STOP, click here **

** To download the STOP non-tailored letter, click here **

STOP is a project at the University of Aberdeen involving:

• The Department of General Practice and Primary Care
• The Department of Medicine and Therapeutics
• The Department of Computing Science

Project:-

• Aims
• Results
• Funding
• Examples
• People
• Papers

Aims:

• To develop a computer system for generating tailored letters to help people stop smoking
• To research knowledge acquisition (KA) techniques to acquire text-planning and sentence-planning rules from domain experts
• To evaluate the clinical effectiveness of the computer generated letters in a general practice setting
• To evaluate the cost effectiveness of this brief smoking cessation intervention

Results:

The results of our clinical trial suggested that while sending smokers a letter could help a small but useful number of people quit, the tailored letters were no more effective in this regard than the non-tailored letters. The tailored letters may have been slightly more effective with heavy smokers and others who found it especially difficult to quit, but the evidence for this is not conclusive. For more information on the outcome of the clinical trial, see the following papers:

• (for medical audiences) AS Lennox, LM Osman, E Reiter, R Robertson, J Friend, I McCann, D Skatun, and P Donnan (2001). The Cost-Effectiveness of Computer-Tailored and Non-Tailored Smoking Cessation Letters in General Practice: A Randomised Controlled Trial. British Medical Journal 322:1396-1400. (eBMJ archive)
• (for computing audiences) E Reiter, R Robertson, and LM Osman (2003). Lessons from a Failure: Generating Tailored Smoking Cessation Letters. Artificial Intelligence 144:41-58. (PDF)

• E Reiter, S Sripada, and R Robertson (2003). Acquiring Correct Knowledge for Natural Language Generation. Journal of Artificial Intelligence Research 18:491-516. (PDF).

Funding:

Examples:

• Sample Letter
• Prototype online version of STOP

People:

• Dr. Scott Lennox (General Practice and Primary Care)
• Prof. James Friend (Medicine and Therapeutics)
• Dr. Liesl Osman (Medicine and Therapeutics)
• Dr. Ehud Reiter (Computing Science)

• Ian McCann(General Practice and Primary Care)
• Roma Robertson   (Computing Science)

• Annette Hermse   (General Practice and Primary Care)

• Duncan MacIver
• Steven Porter
• Yvonne McKay (Roff) worked on the initial Pilot Study

Papers:

Journal papers

E Reiter, R Robertson, and LM Osman (2003). Lessons from a Failure: Generating Tailored Smoking Cessation Letters. Artificial Intelligence 144:41-58. (PDF)

• Summarises the entire STOP project from a computer science perspective, and discusses some of the methodological issues in reporting negative results. The is the best single paper on STOP for computer scientists.

E Reiter, S Sripada, and R Robertson (2003). Acquiring Correct Knowledge for Natural Language Generation. Journal of Artificial Intelligence Research 18:491-516. (PDF).

• Summarises our findings about knowledge acquisition for NLG in STOP and also in our later SumTime project.

AS Lennox, LM Osman, E Reiter, R Robertson, J Friend, I McCann, D Skatun, and P Donnan (2001). The Cost-Effectiveness of Computer-Tailored and Non-Tailored Smoking Cessation Letters in General Practice: A Randomised Controlled Trial. British Medical Journal 322:1396-1400. (eBMJ archive)

• Summarises the entire STOP project from a medical perspective. This is the best STOP paper for medical audiences.

E. Reiter (2000). Pipelines and Size Constraints. Computational Linguistics 26:251-259. (PDF)

• Discusses the problem of generating letters which must fit in a fixed number of pages, and experimentally evaluates three alternative architectures for meeting this constraint. This is probably the most interesting aspect of STOP from a technical NLG perspective.

Conference and workshop papers

E Reiter, R Robertson, S Lennox, and L Osman (2001). Using a Randomised Controlled Clinical Trial to Evaluate an NLG System. In Proceedings of ACL-2001, pages 434-441. Proceedings available from the Association of Computational Linguistics, 75 Paterson St, New Brunswick, NJ, USA. (PDF)

• Describes the STOP clinical trial from a computing perspective. Includes some qualitative observations and comments which are not in the BMJ paper.

E. Reiter, R. Robertson, and L. Osman (2000). Knowledge Acquisition for Natural Language Generation. In Proceedings of the First International Conference on Natural Language Generation (INLG-2000), pages 217-224. Proceedings available from the Association of Computational Linguistics, 75 Paterson St, New Brunswick, NJ, USA. (PDF version)

• Describes and evaluates the Knowledge Acquisition (KA) techniques used to build the STOP software.

E. Reiter (1999). Shallow vs. Deep Techniques for Handling Linguistic Constraints and Optimisations. In Proceedings of the KI-99 Workshop on May I Speak Freely: Between Templates and Free Choice in Natural Language Generation. (PDF).

• An early workshop paper describing some of the design problems and choices in the STOP software,

E. Reiter, R. Robertson, and L. Osman (1999). Types of Knowledge Required to Personalise Smoking Cessation Letters. In W.Horn et al (Eds.):Proceedings of the Joint European Conference on Artificial Intelligence in Medicine and Medical Decision Making (AIMDM'99), pages 389-399. Springer-Verlag. (PDF).

• An early paper which discusses the types of knowledge that can be used to personalise smoking-cessation letters.

E Reiter, A. Cawsey, L Osman, and Y Roff (1997). Knowledge Acquisition for Content Selection. In Proceedings of the 1997 European Workshop on Natural Language Generation, pages 117-126. Duisberg, Germany (PDF).

• A very early paper on knowledge acquisition in STOP. People interested in KA issues should probably read our INLG-2000 paper instead.

E. Reiter and L. Osman (1997). Tailored patient information: some issues and questions. In Proceedings of the ACL-1997 Workshop on From Research to Commercial Applications: Making NLP Technology Work in Practice, pages 29-34 (PDF).

• A very early workshop discussion paper about computer systems that produce personalised patient information material.

For further information, contact Dr. Ehud Reiter

Top of Page