CSE 5331/7331 Fall 2008
AN INTRODUCTION TO DATA MINING AND RELATED TOPICS

Course Syllabus

Book:

Data Mining Introductory and Advanced Topics by Margaret H. Dunham Prentice Hall, 2003 Book Web Page

Class Material

• Slides
• WEKA slides

Homework

• HW 1
  Articles from Homework:
  • “Gene Analysis Boosts Sidelined Heart Drug” (Benjamin Koopferstock)
  • “Retail Group Forecasts Anemic 2.2% Rise in Holiday Sales” (Kathryn Patterson)
  • “Mark Smolinski: Detect Epidemics Before They Start” (Kristina Kraemer)
  • “Adaptation of Bayesian Data Mining Algorithms to Longitudinal Claims Data: Coxib Safety as an Example” (Mallik Kotamarti)
  • “US to create 'risk assessments' of air passengers” (Chris Garcia)
  • “Mining the Voice of the Customer to Improve Enterprise Performance and the customer Experience” (Liang Wang)
  • “New Thoughts on Language Acquisition: Toddlers as Data Miners” (Yu Su)
  • “Association between paracetamol use in infancy and childhood, and risk of asthma, rhinoconjunctivitis, and eczema in children aged 6-7 years: analysis from Phase Three of the ISAAC programme.” (Aaron Smith)
  • “Sexist Men Earn More Money: Study ” (Bryan Palmer)
  • “Data Mining for Cyber Security” (Thomas Shadle)
  • “Campaign Finance Tracker” (Karl Burkheimer)
  • “Employment Situation Summary” (Jennifer Lewis)
  • “Plan’s Mystery: What’s All This Stuff Worth?” (Steven Weber)
  • “Update: Google Guy's Words Get Quick Response” (Daniel Liu)
• HW 2
  Data for HW 2
• HW 3
  Data for HW 3
• HW 4

Projects:

• Project I
• Project II
• Data for Project II
• Project III
• Data for Project III

Supplementary Class Material

• Introduction
  • CRISP-DM
  • The Chi-Square Test
  • Chi-Square Probabilitites
  • Privacy:
    • Privacy Cartoon
    • Government Wiretapping
  • Visualization:
    • Box Plot
    • Lift Chart
    • Data Modeling and Mining: Why Lift? by David S. Coppock
    • ROC Curve
    • ROC Curve ROC Example profided by Dr. Monnie McGee
• Techniques
  • Neuron
  • Mind and Machine Biology of the Brain
  • Neural Impulse 3D Animation (Downloaded from YouTube)
  • Genetic Algorithm Example
• Classification
  • Nonlinear Regression
  • Decision Trees:
    • Example 4.9
    • Gain vs. Gain-Ratio
    • Excerpt from Classification and Regression Trees by leo Breiman, Jerome H. Friedman, Richard A. Olshen, and Charles J. Stone
    • (submitted by Nathan Banek and his brother) Twenty Questions Web Site
  • Neural Networks:
    • Neural Networks by Christos Stergiou and Dimitrios Siganos
    • Artificial Neural Networks Technology by Dave Anderson and George McNeil
    • Derivation of weight change calculation for gradient descent
    • How many hidden units should I use?
    • Kohonen Networks
    • SOM Demonstration
    • SOM of the Market
    • Momentum and Learning Rate Adaptation
    Metrics:
    • Kappa Statistic
    • Kappa Statistic
    • Area under ROC
• Clustering:
  • Original BIRCH paper
  • Original DBSCAN paper
  • Crime Data Visualization and Spatial Database Management System
  • Original CURE paper
  • COBWEB on Wikipedia
  • Original COBWEB paper
  • K-Means and EM Clustering Algorithms
  • Dimentionality Reduction
    • Reducing Data Dimension from CMU Machine Learning Class
    • Wikipedia Principal Components Analysis
    • Principal Components Analysis Tutorial by Lindsay Smith
• Dimensional Modeling:
  • Dimensional Modeling Example
  • Decision Support by R. Ramakrishnan and J. Gehrke
  • The OLAP Report
  • OLAP Analysis
  Data Warehouse:
  • CDC Data Warehouse
  • The Data Warehouse Institute
  • The Data Warehousing Information Center
  • The Kimball Group
  • DB2 Data Warehouse Edition
  • Data Warehousing and OLAP by Hector Garcia-Molina
  • Introduction to Tivoli Enterprise Data Warehouse
  • Advances in Data WareHouse Perforamnce I/O Elimination in DB2, Winter Corporation

  Sample Applications:

  • Sports (8/26/08)
    • Beijing 2008
    • Olympic IDashboard
  • Hurricane Prediction (8/26/08,8/28/08)
    • NOAA National Hurricane Center
  • Subprime Mortgage Foreclosures (8/26/08)
    • The New York Times, In the Shadow of Foreclosures
  • Data Mining Applications Overview (8/28/08)
    • SEDE '08 Keynote
  • Biodegradation Prediction (Enviroinformatics) (9/4/08)
    • SMU BDMINE
  • Nuclear Test Detection (9/23/08)
    • CTBTO
    • Spatiotemporal Stream Mining Applied to Seismic+ Data
  • Data Mining and Elections (11/4/08)
    • Data Mining Techniques to Study Voting Patterns in the US
    • Dow does CNN make election projections?
    • Edison Media Research Exit Poll
    • Election Projection
    • Election-Projection.net
    • Projection Accuracy Competition

    Tools:

    • Winplot (Peanut Software)
    • Zoomit XLMiner
    • WEKA
      • WEKA
      • WEKA Wiki
      • WEKA Page by Hakan Kjellerstrand
      • WEKA Machine Learning Algorithms in Java by Ian H. Witten and Eibe Frank
    • Stuttgart Neural Network Simulator
    • Rulequest
    • ODBCMine Decision Trees
    • Salford Systems (CART)
    • Hierarchical Clustering Explorer

Resources:

• Statsoft Electronic Textbook
• Mathworld
• Statistics Primer (Revised)
• KDNuggets
• ACM SIGKDD
• The Data Mine
• SMU Library User Education, Citing Sources: APA Style
• IEEE Citation Style