Syllabus

I will follow the main text book Tan/Steinbach/Kumar closely:

1. 08.07 Lecture 1

   Introduction (chap 1)

   • History
   • Definitions
   • Data Mining Tasks
   • Data Mining standards
2. 08.14 Lecture 2

   Data (chap 2)

   • Data Attributes
   • Data Preprocessing
   • Similarity Measures
3. 08.21 Lecture 2

   Data (chap 2)

   • Similarity Measures

   Guest Lecture by Mr. Keith Gile, Senior Advisor to CSO, Business Objects

   

   • Title: Business Intelligence 2.0 - The future of BI
   • Bio: Mr. Gile is an internationally recognized expert on Business Intelligence (BI), information management, reporting and analysis, and end-user segmentation. He currently works directly with Bernard Liautaud – the founder and Chief Strategy Officer at Business Objects as an advisor, and is a frequent speaker at industry events such as Computerworld’s BI Perspectives. Keith, has published over 200 research papers including several in leading magazines such as DMReview, and CIO while an analyst with Forrester Research, and consulted with leading corporations and government organizations about BI industry trends, BI architecture, best-practices, technology selection, acquisitions, and product development. Prior to joining Forrester Research and GiGa Information Group, Mr. Gile spent over fifteen years in the business intelligence and data warehousing industries, managing the development of complex decision support systems and technology education products and services.
4. 08.28 Lecture 3

   Exploring Data (chap 3)

   • Summary Statistics
   • Visualization
   • OLAP and Multidimensional Data Analysis
5. 09.04 Lecture 4

   Overview of Predictive Learning & Function Approximation (tentative)

   • Based on Friedman's 1994 paper of the same name
6. 09.11 Lab 1 @ N4-b3c-14

   Lab 1 of 2 (self-paced)

   • 2 registered students to share one machine
     • Sorry, due to limited machines, we can only accomodate registered students
   • Meet @ LT18 (optional for those who do not know the way)
   • Walk to N4-b3c-14 (CAIS) @ 6:45 pm
   • Getting familar with SAS Enterprise Miner
     • Download (and optionally print & bind) from SAS the following 150+ page document Getting Started with SAS Enterprise Miner 5.2
     • Also download the corresponding dataset during the lab
     • We will be following the tutorial in this guide.
7. 09.18 Lecture 5

   Classification 1 (chap 4)

   • Decision Trees
   • Model Overfitting
   • Bias/Variance
   • Comparing Classifiers
8. 09.25 Term Break

   No Lecture
9. 10.02 Lecture 6

   Classification 2 (chap 5)

   • Rule-Based Classifier
   • Nearest Neighbour Classifier
   • Bayesisan Classifiers
   • Artificial Neural Network (ANN)
   • Support Vector Machine (SVM)
   • Ensemble Methods
   • Class Imbalance Problem
10. 10.09 Lecture 7

    Cluster Analysis 1 (chap 8)

    • K-Means
    • Agglomerative Hierarchical Clustering
    • DBSCAN
    • Cluster Evaluation
11. 10.16 Lecture 8

    Cluster Analysis 2 (chap 9)

    • Prototype-Based Clustering
    • Density-Based Clustering
    • Graph-Based Clustering
    • Scalable Clustering Algorithms

    Anomaly Detection (chap 10)

    • Statistical Approaches
    • Proximity-Based Outlier Detection
    • Density-Based Outlier Detection
    • Clustering-Based Techniques
12. 10.23 Lecture 9

    Association Analysis 1 (chap 6)

    • Frequent Itemset Generation
    • Rule Generation
    • Compact Representation
    • Evaluation

    Association Analysis 2 (chap 7)

    • Handling Categorical, Continous attributes
    • Handling concept hierarchy
    • Sequential patterns
    • subgraph patterns
    • infrequent patterns
13. 10.30 Lab 2 @ N4-b3c-14 (CAIS)

    Lab 2 of 2

    • Cluster Analysis
    • Association Rules
    • Download Lab Material