Necati Aras

This course will examine methods that proved to be useful in recognizing patterns and making predictions. We will review applications and provide an opportunity for hand-on experimentation with data mining algorithms. At the end of the course students will have developed an understanding of the strengths and limitations of popular methods to solve regression and classification problems in data mining.

Introduction to heuristic methods; classical construction heuristics; classical improvement heuristics; Lagrangean relaxation, simulated annealing, tabu search. Neural networks, genetic algorithms, ant colony optimization.

Economics and engineering decisions; principles of decision theory; generation and evaluation of alternatives; unconstrained and constrained optimization; duality and sensitivity analysis; application of LP; network models; simulation; case studies.

Integer programming: modelling, branch-and-bound method, cutting plane algorithm, linearization, preprocessing and node/variable selection in branch-and-bound; deterministic dynamic programming; nonlinear programming: modeling and examples, unconstrained and constrained optimization; Markov chains; Poisson process and exponential distribution; queueing theory.

Dynamic programming; introduction to stochastic processes; queueing models; reliability models; Markov decision processors; value iteration method; successive approximations and policy improvement algorithms; maintenance and reliability models; game theory.

Basic topics in parametric statistics; estimation, confidence intervals, and hypothesis testing; analysis of variance, regression and correlation analysis; goodness of fit tests; application in statistical quality control, demand forecasting, and other IE/OR topics; elementary design of experiments and data collection; computer implementations using available up-to-date statistical software.

Modelling concepts; linear programming models; simplex and dual simplex methods; duality and sensitivity analysis; transportation, transshipment and assignment problems; basic problems in network theory: minimum spanning tree, shortest path, maximum flow; integer linear programming: modeling