The use of data mining in healthcare has become a necessity as providers are under increased pressure to make their processes more efficient and as ever more patient data are being collected via EHR systems. This workshop will provide an introduction to the two main classes of data mining algorithms, predictive / supervised and descriptive / unsupervised, including data preprocessing considerations, guidelines for appropriate use, and review of published case studies. Topics covered include dimensionality reduction, classification (e.g., decision trees, support vector machines, neural networks), cluster analysis, association rules mining, and/or outlier detection. Several exercises using freely-available data analysis software (R) will allow attendees to gain experience in applying common predictive modeling and pattern recognition methods to healthcare data sets.