Instructor: Randy S. Weinberg
Office: PH 223G
Office Hours: T 2:00 - 4:00, W 11:00-1:30, Th 10:00 - 12:00, others by appt.
Phone: 268-3228
Email: rweinberg@cmu.edu

Required Text: Seven Methods for Transforming Corporate Data into Business Intelligence by Vasant Dhar and Roger Stein

Other readings to be announced.

Recommended for Additional Depth:
- How to Solve It : Modern Heuristics by Zbigniew Michalewicz and David B. Fogel (for those wishing a more rigorous introduction)
- Genetic Algorithms + Data Structures = Evolution Programs by Zbigniew Michalewicz (excellent intro to GAs if you can find it)

References and Resources:
ACM Digital Library :Communications of the ACM (Association for Computing Machinery), all ACM journals and conference proceedings
IEEE Xplore : All IEEE (Institute for Electrical and Electronics Engineers) journals, magazines (including IEEE Intelligent Systems, formerly IEEE Expert) and conference proceedings
Electronic Collections Online (ECO) CMU library online reference to journals, etc.
Many other books, journals, magazines, conference proceedings in CMU libraries and via CMU electronic subscriptions
Yahoo's AI Page: Numerous Interesting Links
About.Com AI Site: Comprehensive Index
On-Line Demos of many technologies by Pacific Northwest National Laboratories
PC AI: "Intelligent Solutions for Today's Computers"

Intelligent Decision Support Methods is a course about emerging and advanced computer based applications for management support. You will see how these technologies can help people make better decisions.

The impact of technology on many organizations is increasing and profound at all levels. People and technology interact now, more than ever, to address organizational issues. With the rapid acceleration of the pace of change and increasing complexity, managers need better tools to help them make quality decisions. Among the most interesting and highest ROI (return on investment) IT projects are those that help managers make better decisions. Technologies that help managers retrieve, analyze, summarize, and understand their data and organizational intelligence (along with the use of the Internet and intranets) are becoming the cornerstone of modern management.

Traditional MIS and Decision Support solutions such as relational database, management science and operations research approaches have played, and continue, to play an important role in informing managerial decisions. However, various methodologies – including data warehousing, data mining, and various Artificial Intelligence (AI) based techniques - have emerged in recent years as powerful, cost effective solutions. Significant recent advances in the price, performance, and usability of all of these knowledge intensive information technologies have enabled their adoption in organizations of all sizes. There is evidence that these technologies are changing the manner in which organizations are managed, structured and reengineered.

We will examine some of the most important and promising technologies in use today:

Data Warehousing: Discovering the real value of the organizational databases

Expert Systems: Computer programs that perform comparably to human experts in a problem domain

Machine Learning: Automating knowledge acquisition from existing data

Case Based Reasoning: Using history as a guide

Artificial Neural Networks: Pattern recognition based on real world data

Genetic Algorithms: Evolving solutions to complex decision problems

Fuzzy Logic and Linguistics: Extending logic to allow for partial, or continuous truths

Intelligent Software Agents: Automating the search for information

You will learn about these important technologies through a combination of readings, case analyses, short projects, software demonstrations, guest speakers, and a term project. Keep in mind, however, that these technologies (while of interest in themselves) are actually of secondary importance to us; our main concern is their impact and relevance to organizational decision making. Learning which tools to apply in which situations is an important part of our agenda. I hope that what you learn this term will excite your imagination!

Normally, I will spend class time discussing cases, technology, and current practice with you. Since we're on a tight schedule, I generally won't be going over the material that you can easily read and learn on your own. Therefore, I expect you to read the assigned materials before class begins and be prepared to discuss them if called upon. From week 2 through week 7 there will be a short in-class writing assignment (a paragraph or two) based on the assigned reading material for the week. You cannot make up any in class writing assignment you miss either through absence or lateness; however, I will drop your lowest writing assignment score of the six given throughout the term.

I will assume that you are well prepared for class and will feel free to call upon you to discuss the readings unless you notify Sharon Blazevich (PH 223E) by 4:00pm of class days that you are not prepared. You may do this once during the term. If I call on you and you are obviously not prepared, you will get no credit for class participation for the evening.

Note carefully our calendar. We have no Tuesday evenings off for holidays, term breaks, and special events.

We will be covering a wide variety of topics this mini session and moving at a fast pace. Fortunately, our textbook is exceptionally well written; Dhar and Stein serve as remarkable tour guides. At all times, keep in mind that this is not a computer science course. We are primarily interested in the pragmatic decision and management support aspects of these technologies.

In addition to the readings and in-class writing assignments, you will be asked to complete the following tasks during this course:

· A 15-20 minute presentation of one substantial article, case study, business analysis, software product review, local company profile, or the like, related to course topics. Accompanying the presentation must be a one page abstract/summary of your findings.

· A team term project.

· Final Examination. Details to be discussed in class.

COURSE GRADING

A composite of these components will be used to determine your final grade:
- Term Project: 30%
- Final Examination: 30%
- In Class Writing (based on assigned readings): 20%
- Individual Presentation: 10%
- Class participation: 10%

Thursday March 8, Mini Course Final Exam Day (also mid semester break)

Selected Readings for more information:

Data Warehousing / Online Analytical Processing

White Papers and Short Cases From CIO Magazine
Data Warehousing Information Center
Assessing Loan Risks: A Datamining Case Study by Rob Gerritsen
"Machine Learning and Data Mining" by Tom Mitchell in Communications of the ACM
"Interactive Analysis of Computer Crimes" by Brown, Gunderson, Evans in IEEE IT Pro, August 2000
"Data Mining to Predict Aircraft Component Replacement" by Letorneau, Famili, Matwin in IEEE Intelligent Systems, Nov/Dec 1999
Various feature articles in Communications of the ACM, September, 1998


Artificial Neural Networks and Genetic Algorithms

Artificial Neural Networks at Pacific Northwest National Laboratory (lots of links)
The Basic Ideas in Neural Networks by Rumelhart, et. al. in Communications of the ACM
Neural Networks by Christos Stergiou and Dimitrios Siganos
Z Solutions: Downloadable Demos (and software) of high quality
QNet V2000: Neural Network software
Genetic Programming Notebook: Lots of interesting links for all sorts of AI
"Forecast: A Neural System for Diagnosis and Control of Highway Surfaces" by Luchetta, Manetti, Francini in IEEE Intelligent Systems, May/June 1998

Expert Systems and Analytic Hierarchy Process (AHP):

Acquired Intelligence Inc: Specialists in knowledge acquisition and expert system development (interesting demos)
OSHA's Expert Systems: OSHA's environmental and workplace Expert Systems
ExpertChoice: Expert Choice software (AHP)
Decisions by Objectives by Dr. Ernest H. Forman (online book on AHP)
"The use of the analytic hierarchy process in the selection of participants for a telecommuting pilot project" by Karsten and Garvin in ACM SIGCPR / SIGMIS '96
Attar Software: Rule Induction

Case Based Reasoning:

AI-CBR: Case Based Reasoning
Case Based Reasoning Resources
Case Based Reasoning on the Web
CBR Demos
"Cased-Based Reasoning Support for Online Catalog Sales" by Vollrath, Wilke, Bergmann in IEEE Internet Computing, Jul/Aug 1998

Fuzzy Decison Making:

Fuzzy Logic at Pacific Northwest National Laboratories
Fuzzy Net Online: Demos, Tutorials, Source Code
Fuzzy Judgment Maker: Fuzzy Logic DSS Tools
FuzzyTech: Fuzzy Logic Software (info and download)
Cognitive Fuzzy Modeling for Enhanced Risk Assessment in a Health Care Institution by Smith and Eloff in IEEE Intelligent Systems, Mar/Apr 2000

Intelligent Software Agents:

AgentBuilder: Very Readable White Papers, Examples, Downloadable Software
Yahoo Intelligent Software Agents
UMBC AgentWeb
Bits and Pixels: Tool to create Intelligent Agents
BotSpot: The Spot for All Bots on the Net
Intelligent Agents for Proactive Fault Detection by Hood and Ji
Protecting the Integrity of Agents: An Exploration into Letting Agents Loose in an Unpredictable World (Crossroads Magazine)
Using Intelligent Agents for Wireless e-commerce Applications: The Yellowstone Project

Text Analysis:

"Lightweight Document Matching for Help Desk Applications" by Weiss, White, Apt, Damerau in IEEE Intelligent Systems, Mar/Apr 2000
"Knowledge Retrieval and the World Wide Web" by Martin and Eklund in IEEE Intelligent Systems, May/June 2000
"Text mining: finding nuggets in mountains of textual data" by Dörre, Gerstl and Seiffert in ACM KDD-99
"Textual data mining of service center call records" by Tan, Blau, Harp and Goldman in ACM KDD 2000