Knowledge Acquisition for Case-Based Design

CBR systems solve new problems by retrieving and reusing solutions for previously solved problems that are stored as case knowledge in a database. Case knowledge is at the heart of a CBR system but additional knowledge may improve problem-solving. Automated tools have been developed to acquire retrieval and adaptation knowledge. These tools have been applied to a challenging tablet formulation domain where CBR is used to design recipes for tablets. Additional ingredients are required to balance the physical and chemical properties of the drug to enable the manufacture of the tablet, its handling by the patient, and disintegration after swallowing. Knowledge-light learning uses only the cases in the case-base to learn explicit knowledge to improve retrieval and achieve adaptation. Effective retrieval knowledge was learned for an initial tablet formulation CBR system and refined when the formulation task changed. Learned adaptation knowledge improved the retrieved formulations, and ensemble learning had a dramatic effect on the prediction of the binder component in a tablet - more than doubling the accuracy of binder after adaptation. [More] Partners AstraZeneca, ISoft (France) Funding EPSRC GR/L98015 RGU RDI Research Team Susan Craw (PI) Nirmalie Wiratunga Jacek Jarmulak Related Projects CBR for Remote Patient Health-Care Monitoring (KTP) CBR for Anomaly Report Processing Complexity Modelling for CBR Systems MediCACE — Medical Decision Support with CBR Knowledge Acquisition for Case-Based Design Posters Knowledge Acquisition for Case-Based Design

Publications

Craw, S. (2007). The case base - the sequel, Proceedings of the 12th UK CBR Workshop, Cambridge, UK. Invited Talk.

Craw, S., Wiratunga, N., & Rowe, R. C. (2006). Learning Adaptation Knowledge to Improve Case-Based Reasoning. In Artificial Intelligence 170(16-17): 1175-1192, Elsevier. doi: 10.1016/j.artint.2006.09.001.

Massie, S., Craw, S., & Wiratunga, N. (2004). A Visualisation Tool to Explain Case-Base Reasoning Solutions for Tablet Formulation. 24th SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence , Cambridge, UK

Craw, S. (2003). Introspective learning to build case-based reasoning (CBR) knowledge containers. 3rd International Conference on Machine Learning and Data Mining in Pattern Recognition (MLDM 03), Leipzig, Germany, on July 5-7, 2003.

Craw, S., & Rowe, R. (2002). Case-based reasoning for tablet formulation: a novel technology to re-use past formulations. In Drug Delivery Companies Report, Pharmaventures , page(s) 14-17

Wiratunga, N., Craw, S., & Rowe, R. (2002). Learning to adapt for case-based design. 6th European Conference on Case-Based Reasoning (ECCBR 2002), Aberdeen, UK

Craw, S. (2001). Case-based reasoning for tablet formulation. British Pharmaceutical Science Conference , Glasgow, Scotland

Craw, S., Jarmulak, J., & Rowe, R. (2001). Learning and applying case-based adaptation knowledge. 4th International Conference on Case-Based Reasoning (ICCBR 01), Vancouver, Canada

Craw, S., Jarmulak, J., & Rowe, R. (2001). Maintaining retrieval knowledge in a case-based reasoning system. In Computational Intelligence, Vol 17, page(s) 346-363

Jarmulak, J., Craw, S., & Rowe, R. (2001). Using case-base data to learn adaptation knowledge for design. 17th International Joint Conference on Artificial Intelligence (IJCAI 01), Seattle, WA

Jarmulak, J., Craw, S., & Rowe, R. (2000). Genetic algorithms to optimise CBR retrieval. 5th European Workshop on Case-Based Reasoning (EWCBR 2K), Trento, Italy

Jarmulak, J., Craw, S., & Rowe, R. (2000). Self-optimising CBR retrieval. 12th International Conference on Tools with AI (ICTAI-2000), Vancouver, Canada