Transcript CS4256 Business Intelligence Systems

IS5152 Decision Making Technologies
Semester 2, 2010/11.
Tuesdays, 6.30-8.30 pm, COM1/204.
Instructor: Dr. Rudy Setiono
Contact: [email protected],
[email protected]
Office: COM2 04-13
IS5152 Decision Making Technologies

Course objective: to introduce students to decision making technologies
that can support decision making in the financial, operational, marketing and
other strategic areas.

Description: Over the past two decades, increasing research efforts have
been directed at finding new machine learning (ML) techniques for decision
making and their possible application in solving practical problems. ML
techniques such as artificial neural network methods have been proven to be
powerful tools for business decision making. Among the application
problems where ML techniques outperform traditional decision making
methods such as statistical methods are credit rating, bankruptcy analysis,
foreign exchange rate predictions and many others.
IS5152 Decision Making Technologies

Topics covered:
The techniques covered in this course include neural networks for
classification/regression/clustering, genetic algorithm for optimization,
decision tree methods, support vector machine, data envelopment
analysis and data mining.

Journal articles that present new techniques for decision making and/or
describe successful application of the existing methods in solving
practical problems will be discussed in class.
IS5152 Decision Making Technologies
This course requires the students to have some background knowledge in:

Calculus

Simple linear algebra

Basic probability and statistics
No computer programming skill is required.
IS5152 Decision Making Technologies
Tentative schedule:
Week 1
January 11, 2011
Introduction and class administration
Week 2
January 18, 2011
Decision making under uncertainty
Week 3
January 25, 2011
Optimization and decision making
Week 4
February 1, 2011
Support vector machines
Week 5
February 8, 2011
Decision making with multiple objectives
Week 6
February 15, 2011
Data envelopment analysis
February 22, 2011
No lecture. Mid-semester break
Week 7
March 1, 2011
Mid-semester exam.
Week 8
March 8, 2011
Decision making with decision trees and rules
Week 9
March 15, 2011
Neural networks for decision making (Part 1)
Week 10
March 22, 2011
Neural networks for decision making (Part 2)
Week 11
March 29, 2011
Rule generation from neural networks
Week 12
April 5, 2011
Genetic algorithms for decision making
Week 13
April 12, 2011
Project presentation
IS5152 Decision Making Technologies
References: Available in the RBR sections of Central Library and HSS
Business Library.
1.
2.
3.
Neural networks: A comprehensive foundation
Author: Haykin, Simon S
Machine Learning
Author: Mitchell, Tom M
Operations research : applications and algorithms
Author: Winston, Wayne L
IS5152 Decision Making Technologies
Grading:
1. Continual assessment (50%):
• Midterm Exam (20%)
• Class project (30%):
o
o
o
20% for the project work, and
10% for project report and presentation.
Projects are to be carried out in teams consisting n students.
2. Final exam on 6 May pm: 50%.
Both midterm exam and final exam are open-book examinations.
IS5152 Decision Making Technologies
Class project:
- Identify an interesting problem/topic to test one or more of the
techniques for decision making discussed in class.
- Search/find/collect relevant data.
Use an available software to analyze the data.
- Software will be provided or they can be obtained via the internet.
- Write a (max) 20 page report.
- Present the project in class (duration: 20 minutes).
- More detailed instructions about the project will be given later in the
semester.
IS5152 Decision Making Technologies
IVLE:
1. Do check IVLE for this course regularly for announcements,
updates, etc.
2. All lecture materials will be placed in the workbin.
3. Message from Students Against the Violation of the Earth
(SAVE):
•
the Office of Provost had approved the submission of all academic
assignments for undergraduate and graduate studies on doublesided print or through electronic submission
•
you are encouraged to print your lecture notes on both sides on the
paper. If possible and depending on the layout of the notes, also
encourage them to print 4 to 6 pages on a side