Project 3: Geriatric applications of Bayesian networks and survival analysis Abstract The overall aim of this project is to combine the approach of Bayesian belief networks with that of stwvival analysis. The analysis can take advantage of the benefits of continuous data and censoring from survival analysis but still have the ability to represent causal relationships by using Bayesian belief networks. We wish to apply the methodology to data from the medical domain, in particular data for lengths and outcomes of spells in geriatric departments. This application of the model to include dynamic aspects of geriatric data is likely to lead to a greater understanding of how diseases are caused and what factors affect duration of an illness or length of spell in hospital. Such factors are clearly of interest to hospital administrators and health care managers. Keywords: Bayesian belief networks, survival analysis, geriatric data & data mining. Organization: University of Ulster at Jordanstown. Northern Ireland, Faculty of Informafics Primary Contact: Adele Marshall Address: School of Information & Software Engineering University of Ulster at Jordanstown Newtownabbey Co. Antrim N.Ireland BT37 0QB Email: AH.Marshall@ulst.ac.uk URL: http://www.infc.ulst.ac.uk/cgibin/infdb/resproiview?projid= 196 Duration: 1 year Number of People: 3 Tools Purchased: HUGIN, BIFROST, CoCo Academic Disciplines: Data Mining, Statistics & Computing Funding Sources: Department of Education for Northern Ireland Project 4. Discovering knowledge in DNA andprotein data Abstract This research investigates a method for discovering knowledge in structural data. We have implemented' the SUBDUE substructure discovery system which discovers interesting and repetitive subgraphs in a labeled graph representation using the minimum description length principle. Experiments have shown SUBDUE's applicability in a variety of domains. We are currently applying SUBDUE to both DNA and protein data from the Brookhaven PDB, where SUBDUE was able to find patterns in secondary structure that are both characteristic and unique to categories of proteins, such as hemoglobin and myoglobin. Ultimately, we plan to use SUBDUE to find structural patterns in functional groups of proteins and the boundaries of genes in DNA. Organization: University of Texas at Arlington, Department of Computer Science and Engineering Contact: Lawrence Holder Address: University of Texas at Arlington Department of Computer Science and Engineering Box 19015 Arlington, TX 76019-0015 Email: holder@cse.uta.edu URL: http://cygnus.uta.edu/subdue Duration: 6 years Number of People: 6 Tools Developed: SUBDUE Academic Disciplines: computer science, molecular biology Funding Sources: National Science Foundation, State of Texas Keywords: molecular biology, proteins, DNA, pattern discovery Project Related Publications: G. Galal, D. J. Cook and L. B. Holder. "Improving Scalability in a Knowledge Discovery System by Exploiting Parallelism", In the Proceedings of the Thirdlnternational Conference on Knowledge Discovery and Data Mining, 171-174, 1997. D. J. Cook, L. B. Holder, and S. Djoko. "Scalable Discovery of Informative Structural Concepts Using Domain Knowledge", 1EEEExpert, 11, 5, 59-68, 1996.

/lp/association-for-computing-machinery/discovering-knowledge-in-dna-and-protein-data-ubC9vUGaPS