Operations Research and Data Mining (IE 525)

2021 Fall
Faculty of Engineering and Natural Sciences
Industrial Engineering(IE)
Kemal Kılıç -kkilic@sabanciuniv.edu,
Doctoral, Master
Formal lecture
Interactive,Learner centered
Click here to view.


The course will address unsupervised learning, supervised learning, association rule mining and feature subset selection problems, focus on the optimization formulations of these problems, discuss various techniques proposed as solutions and present their implementation particularly in the context of operations management. Among others, probabilistic and statistical methods, possibilistic methods, clustering algorithms, decision trees, metaheuristics (such as genetic algorithms, simulated annealing, etc.) and mathematical programming will be covered as part of the toolbox that are widely utilized in data mining. Particular emphasis will be given to multi criteria decision making and multi objective optimization, and their usage in data mining.The course will include case studies from both manufacturing and service industries.


To teach the optimization formulations of various data mining techniques developed for the unsupervised learning, supervised learning, association rule mining and feature subset selection problems.


list the basic components of a data mining process.
model a data mining problem and decide which techniques are suitable for the business objective of the user.
understand the optimization formulation of the various data mining problems and derive the techniques that can be used in those problems.
correctly apply the steps of various feature selection techniques, supervised and unsupervised learning algorithms and association rule mining algorithm.
implement feature selection, supervised and unsupervised learning algorithms, association rule mining techniques with a data mining tool (e.g., WEKA) in order to determine the relations that are hidden in the data.


  Percentage (%)
Final 25
Midterm 25
Assignment 15
Term-Paper 35