Welcome to the Website of

Andy Novobilski, Ph.D.
UTC Chief Research Officer
Assistant Provost for Research / Engagement
Professor and Dept. Head, Computer Science and Engineering

The University of Tennessee at Chattanooga Department of Computer Science & Engineering

Andy (16K)

Curriculum Vitae
Resume
Consulting Projects


Mail:
     Dr. Andy Novobilski
     The University of Tennessee at Chattanooga
     615 McCallie Ave., Dept 2302
     Chattanooga, TN 37403

Office:
     313-D Engineering, Math, and Computer Science Building.

Contact:
     Voice: (423) 425-4202
     Fax: (423) 425-5229
     Email: Andy-Novobilski@utc.edu

Research Focus

One of the fascinating aspects of tool building for datamining is the application of a generalized datamining tool to a specific domain. Often times, this process results in a cross disciplinary analysis of both the datamining technique and the application of the results to the domain itself. This process of cross-disciplinary analysis often leads to improvements of the tool, but more importantly, to a better understanding of the underlying domain model for the domain experts involved.

One area of cross disciplinary research that lends itself to this type of collaboration is found in medical informatics. Over the past 18 months, faculty from the University of Tennessee at Chattanooga and the University of Tennessee College of Medicine, Chattanooga Unit have begun an initiative to research the use of Bayesian Networks in forecasting outcomes related to Acute Coronary Syndrome in an emergency room setting (Fesmire and Novobilski 2003). This is viewed as a key research area as 11,000 patients with acute myocardial infarction (heart attack) and an even greater number of patients with unstable angina (chest pain) are inadvertently discharged from emergency departments nationwide. Adverse outcomes in these patients represent a significant cause of death as well as greater than 25% of malpractice awards.

Having had success with preliminary test data, we have recently begun a project to mine a 17,000 record dataset related to ACS. This project represents a chance to bring together medical informatics, high performance computing, and forecasting under uncertainty using Bayeisan Networks. We look forward to sharing our results (both medical and software) as we explore the new dataset.

Publications

Books

  1. Novobilski, Andrew. PenPoint Programming. Addison-Wesley. 1992.
  2. Cox, Brad, A. Novobilski. Object-Oriented Programming: An Evolutionary Approach, 2nd Edition. Addison-Wesley. 1991.

Journal Articles

  1. Kline, Jeffry, A. Novobilski, et al. "Derivation and Validation of a Bayesian Network to Predict Pretest Probability of Venous Thromboembolism", The Annals of Emergency Medicine, March 2005.
  2. Dumas, Joe, A. Novobilski, D. Ellis, M. Pascal. "VR on a Budget: Developing a Flight Simulator in a Small Institution with Off-The-Shelf Hardware and Open Source Software", The Journal of Computing in Small Colleges, December 2002.
  3. Oman, Paul, A. Novobilski, V. Rajlich, J. Harband, T. McCabe, J.Cross, L. Vanek, L. Davis, K. Gallagher, and N. Wilde. "Maintenance Tools", IEEE Software, pp. 59-65, May 1990.
  4. Novobilski, Andrew. "Pictorial Design Notation", Journal of Object Oriented Programming, pp. 9-14, July/August 1990.

Conference Proceedings

  1. [PDF] Tyler, Tom, A. Novobilski, J. Dumas, A. Warren. "The Utility of Perspecta 3D Volumetric Display for Completion of Tasks." 17th Annual Symposium Electronic Imaging Science and Technology 2005.
  2. [PDF] Novobilski, Andrew, F. Fesmire, D. Sonnemaker. "Mining Bayesian Networks to Forecast Adverse Outcomes Related to Acute Coronary Syndrome." ." The 17th International FLAIRS Conference 2004.
  3. [PDF] Novobilski, Andrew, J. Kline, F. Fesmire. "Using a Genetic Algorithm to Identify Predictive Bayesian Models in Medical Informatics." The International Conference on Information Technology (ITCC) 2004.
  4. Fesmire FM, Novobilski A. "First step in the Erlanger Artificial Intelligence Initiative: development of a Bayesian network utilizing initial triage history to risk stratify chest pain patients for thirty-day adverse outcome." [Abstract]. Ann Emerg Med 2003;42 (in press).
  5. [PDF] Novobilski, Andrew. "The Random Selection and Manipulation of Legally Encoded Bayesian Networks in Genetic Algorithms", The 2003 International Conference on Artificial Intelligence (ICAI) 2003.
  6. [PDF] Novobilski, Andrew, F. Kamangar. "Bayesian Learning with Selective Subsets of Populations in Genetic Programming", The Conference on Smart Engineering System Design: Neural Networks, Fuzzy Logic, Evolutionary Programming, Complex Systems and Data Mining (ANNIE) 2002.
  7. [PDF] Novobilski, Andrew. "Pervasive/Invasive Computing: Two Sides of the Location Enabled Coin", The 2002 International Conference on Parallel and Distributed Processing Techniques and Applications (PDPTA) 2002.
  8. [PDF] Novobilski, Andrew, F. Kamangar. "A Genetic Algorithm Based Approach for Discovering Temporal Trends Using Bayesian Networks", The 6th World Conference on Systemics, Cybernetics, and Informatics. 2002.
  9. [PDF] Novobilski, Andrew, F. Kamangar. "Average Percent Error Based Fitness Functions for Evolving Forecast Models." The 14th International FLAIRS Conference, 2001.
  10. Novobilski, Andrew, F. Kamangar. "Inferencing Bayesian Networks from Time Series Data Using Natural Selection", The 13th International FLAIRS Conference, 2000.
  11. [PDF] Novobilski, Andrew, F. Kamangar. "A Two-Tiered Cognitive Model for Forecasting Time Series Data", Second International ICSC Symposium on Neural Computation, 2000.
  12. Novobilski, Andrew, F. Kamangar. "A Genetic Algorithm Based Approach for Discovering Temporal Trends Using Bayesian Networks." The 19th International Symposium on Forecasting. 1999.