Intelligent Decision Support Systems (IDSS)


Master in Artificial Intelligence (UPC-URV-UB)

Spring semester, Course 2012/2013

(4.5 ECTS, 36 hours)


Responsible Professor:

KEMLG


Miquel Sànchez i Marrè
Knowledge Engineering & Machine Learning Group
Dept. de Llenguatges i Sistemes Informàtics
Universitat Politècnica de Catalunya
Campus Nord-Building K2M, 2nd floor, office 202c
miquel@lsi.upc.edu



Professors:

Professor  Office
E-mail  Phone
Tutoring
Karina Gibert
C5 building,
  2nd floor, office 219
Campus Nord
karina.gibert@upc.edu 93 401 73 23
To convene by e-mail
Miquel Sànchez i Marrè K2M building,
  2nd floor, office 202c
Campus Nord 
miquel@lsi.upc.edu 93 413 78 41
To convene by e-mail

Timetable:

Day Hour Room
Professor
Friday
10-13 BUILDING A5, A5203
M. Sànchez/K. Gibert

Course schedule:

Lecturing periodFrom February 11th 2013 to May 24th 2013
Holidays: March 29th 2013 (Easter), April 26th 2013 (FIB’s Spring Festivity)

Contents:

lIntroducction
Complexity of real-world systems or domains
The need of decision support tools

Decisions
Decision Theory
Modelling of Decision Process

Evolution of Decision Support Systems 
Historical perspective of Management Information Systems
Decision Support Systems (DSS) 
Advanced Decision Support Systems
Intelligent Decision Support Systems (IDSS)

Intelligent Decision Support Systems (IDSS)
IDSS Architecture
IDSS Analysis and Design
Requirements, advantages and drawbacks of IDSS 
IDSS Validation
Implementation of an IDSS in a computer: Environmental Systems Management (an example)
l
Knowledge Discovery in a IDSS: from Data to Models 
Introducction
Knowledge Discovery and Data Mining
Artificial Intelligence and Statistics
Data Structure
Data Filtering
Knowledge Models, Uncertainty Models and Data Mining Techniques
Descriptive Models 
(AI) Conceptual Clustering Techniques
(Stats) Statistical Clustering Techniques
(AI&Stats) AI&Stats Hybrid Techniques: Classification Based on Rules (ClBR)
uAssociative Models 
(AI): Association Rules, Model-Based Reasoning, Qualitative Reasoning
(Stats): Component Principal Analysis (CPA), Simple Correspondence Analysis (SCA), Multiple Correspondence Analysis (MCA)
(AI&Stats) Bayesian (Belief) Networks
Discriminant Models 
Rule-based Reasoning 
(AI) Decision Trees 
(AI) Classification Rules 
(Stats) Discriminant Analysis
(AI&Stats) Box-plot Based Induction Rules
Case-Based Reasoning 
(AI) Instance-Based Learning (IBL)
Bayesian Learning
uPredictive Models
(AI) Connexionist Models (ANN), Evolutionary Computation, Collaborative Resolution (“Swarm intelligence“)
(Stats) Multiple sand Simple Linear Regression Models, Variance Analysis, Time Series Models
(AI&Stats) Regression Trees / Model Trees
Uncertainty Models
(Stats) Pure Probablistic Model
(AI&Stats) Bayesian Networks
(AI) Certainity Factor Model
(AI) Possibilistic Model (Fuzzy Logic)
l

Postprocessing and Model Validation
        Postprocessing Techniques
        Validation
    Graphical Validation Tools
    Validation methods for Predictive and Discriminant Models
    Validation methods for Descriptive and Associative Models
Statistical Methods for Hypotheses Verification

Software Tools for IDSS Development 
GESCONDA
KLASS
WEKA
System R
DAVIS

Application of IDSS to real-world problems
DAI-DEPUR+
PSARU

Future Trends in IDSS and Conclusions

Learning Resources
Literature References
Internet Resources



Course material

The “Racó de la FIB” is the Virtual Campus for you. There, you will find any announcement, change in the schedule, the practical works, the results of the evaluation of the works and final marks, etc. There you will also find a pointer to this WEB page of the course where you could find some other material (slides, internet references, etc.) in electronic format

2011/2012 Sylabus
Part 1 - Introduction/Decision Theory
Part 2 - Evolution of DSS/IDSS
        Part 3 - KDD in IDSS/Preprocessing
Part 4 - DM Methods/Descriptive & Associative Models
        Part 5 - DM Methods/Discriminant & Predictive Models
Part 6 - Uncertainty Models
        Part 7 - PostProcessing and Model Validation
        Part 8 - Tools and Applications
Part 9 - Future Trends in IDSS/Conclusions
2012/2013 Sylabus
Part 1 - Introduction/Decision Theory
Part 2 - Evolution of DSS/IDSS
        Part 3 - KDD in IDSS/Preprocessing
Part 4 - DM Methods/Descriptive & Associative Models
        Part 5 - DM Methods/Discriminant & Predictive Models
Part 6 - Uncertainty Models
        Part 7 - PostProcessing and Model Validation
        Part 8 - Tools and Applications
Part 9 - Future Trends in IDSS/Conclusions

        Case Study 1a - Implementation of an IDSS in a computer (I) (Customer Fidelization Case)
Case Study 1b - Implementation of an IDSS in a computer (II) (Optimal Wastewater Treatment System Assignment Case)
        Case Study 2 - The use of Descriptive Models for Decision Support (Health/Medical Case)
Case Study 3 - The use of Associative Models for Decision Support (I)
        Case Study 4 - The use of Predictive Models for Decision Support (Breast Cancer Case)
        Case Study 5 - The use of Discriminant Models for Decision Support (I) (Wastewater Treatment Plant Management Case)
        Case Study 6 - The use of Discriminant Models for Decision Support (II) (The Textile Industry Case)
Case Study 7 - The use of Associative Models for Decision Support (II) (Ecological Interrelationships Case)

       
Practical exercises

Three practical works must be done by all students in the course (PW1, PW2 and PW3). The practical works will be:

PW1 (Individual work) - Modelling a decision making process in a concrete domain.
PW1 will be given on: February 15th, 2013
PW1 is due on: October 22nd, 2013

PW2 (Individual work) - Reviewing the state of the art of Intelligent Decision Support Systems.
PW2 will be given on: February 22nd, 2013
PW2 is due on: March 8th, 2013

PW3 (Group work) - A research work, including some practical software development, about several features of Intelligent Decision Support Systems, such as in some data mining models.
PW3 will be given on: February 22nd, 2013
PW3 is due on: May 24th, 2013

The work delivery and oral exposition of the PW3 will be done on Friday 24th May 2013 (20 min. approx. for each group)

Evaluation
Evaluation of the Knowledge and skills obtained by the students will be assessed through the 3 practical Works. The final grade will be the weighted mean of the grade of each practical work.  Each practical work will have the following weights:

PW1 -> 10%
PW2 -> 20%
PW3 -> 70%

Thus, the final grade will be computed as follows:

FinalGrade = 0.1*PW1Grade + 0.2*PW2Grade + 0.7*PW3Grade

Evaluation of the course performance will be based on three marks coming from the three exercises. The first exercise about decision modelling will have a 10% weight, the second one about the state of the art in IDSS will have a 20% weight and the research and software development part will have a 70% weight in the final evaluation.

The first exercise will be evaluated by means of its quality and its justified explanation in the document. The second exercise will be evaluated according to its accuracy and completeness.The third exercise will be evaluated through the quality of the research undertaken, the software and/or documentation delivered, and the quality of the oral exposition (both presentation and content assessed).


Honour Code (adapted from Stanford University Honor Code)

A.    The Honor Code is an undertaking of the students, individually and collectively:
  1. that they will not give or receive aid in examinations; that they will not give or receive unpermitted aid in class work, in the preparation of reports, or in any other work that is to be used by the instructor as the basis of grading;
  2. that they will do their share and take an active part in seeing to it that others as well as themselves uphold the spirit and letter of the Honor Code.
B.    The faculty on its part manifests its confidence in the honor of its students by refraining from proctoring examinations and from taking unusual and unreasonable precautions to prevent the forms of dishonesty mentioned above. The faculty will also avoid, as far as practicable, academic procedures that create temptations to violate the Honor Code.

C.    While the faculty alone has the right and obligation to set academic requirements, the students and faculty will work together to establish optimal conditions for honorable academic work.

Examples of conduct which have been regarded as being in violation of the Honor Code include:

•    Copying from another’s examination paper or allowing another to copy from one’s own paper
•    Unpermitted collaboration
•    Plagiarism
•    Revising and resubmitting a quiz or exam for regrading, without the instructor’s knowledge and consent
•    Giving or receiving unpermitted aid on a take-home examination
•    Representing as one’s own work the work of another
•    Giving or receiving aid on an academic assignment under circumstances in which a reasonable person should have known that such aid was not permitted

In recent years, most student disciplinary cases have involved Honor Code violations; of these, the most frequent arise when a student submits another’s work as his or her own, or gives or receives unpermitted aid. The standard penalty for a first offense or multiple violations will be proposed by the teachers in charge of the course, and approved by the Academic Commission of the Master MAI.



Resources for student learning

Literature

greenball Basic References

greenball Other IDSS references and IDSS Applications

2012

2011


2010

2009

2008

2007


2006

 

2005

 

2004


2003
  • J.L. Arcos. Development of an Industrial Design Support System for Air Pollution Treatment. Procc. of IJCAI’03 Workshop on Environmental Decision Support Systems (EDSS’03), pp. 1-8, Acapulco, 2003.
  • J. Hastings, K. Branting, J. Lockwood and S. Schell. CARMA+: A General Architecture for Pest Management. Procc. of IJCAI’03 Workshop on Environmental Decision Support Systems (EDSS’03), pp. 18-21, Acapulco, 2003.
  • J. Wiese, S. Schmitt, R. Bergmann and T. G. Schmitt. A Case-Based Predictive Sequencing Batch Reactor Controller. Procc. of IJCAI’03 Workshop on Environmental Decision Support Systems (EDSS’03), pp. 22-28, Acapulco, 2003.
  • R. Srinivasan and I. Halim. Multi-perspective Models for Diagnosing Waste Generation in Chemical Processes. Procc. of IJCAI’03 Workshop on Environmental Decision Support Systems (EDSS’03), pp. 44-49, Acapulco, 2003.
  • P. Struss, M. Bendati, E. Lersch, W. Roque and P. Salles. Design of a Model-based Decision Support System for Water Treatment. Procc. of IJCAI’03 Workshop on Environmental Decision Support Systems (EDSS’03), pp. 50-59, Acapulco, 2003.
  • M. Martínez, M. Sànchez-Marrè, J. Comas and I. Rodríguez-Roda. Defining a Decision Support System to Manage Filamentous Bulking Episodes in Activated Sludge Systems. Procc. of IJCAI’03 Workshop on Environmental Decision Support Systems (EDSS’03), pp. 69-76, Acapulco, 2003.

2002

2000

1999

1998

Internet Resources

Catalan Statistics Institute (Institut d'Estadística de Catalunya)
http://www.idescat.net 
European Statistics Office
http://www.europa.eu.int/comm/eurostat
UCI Machine Learning Repository
http://archive.ics.uci.edu/ml/
Knowledge Discovery Network of Excellence
http://www.kdnet.org/
Data Mining Resources
http://www.scd.ucar.edu/hps/GROUPS/dm/dm.html
Data Mining Sites
http://www.cacs.louisiana.edu/~arlab/repos/sortedSites.html
GESCONDA software
http://www.lsi.upc.edu/~webia/KEMLG/projects/gesconda.html 
WEKA Software
http://www.cs.waikato.ac.nz/~ml/
R Software 
http://www.r-project.org/
DAVIS Software
http://stat.skku.ac.kr/myhuh/DAVIS.html