Introduction to statistical pattern recognition. The main topics include Bayes decision theory, discriminant functions, maximum likelihood estimation, PCA, LDA, semi-supervised kernel learning, and graph embedding. This course will discuss some applications of data mining in different application fields, such as business, marketing, medical imaging, biology. Prerequisite: MATH 3330.

Covers computational methods used to implement modern statistical and scientific analyses in data-intensive research. Topics include numerical linear algebra and matrix computations for statistical models; numerical optimization for estimation and learning; Monte Carlo and stochastic simulation methods; and strategies for working with large or high-dimensional datasets. Emphasizes programming patterns for vectorization, modular algorithm design, and use of high-performance or parallel computing resources. Students complete hands-on projects that translate statistical methodology into efficient, reproducible code for real research problems. Prerequisite: DATA 5301* Foundations of Data Science and DATA 5302* Probability & Statistics for Data Science, or equivalent preparation, and graduate standing in the Division of Data Science.