Data Preparation for Mining; DM Techniques: Market Basket Analysis
The Knowledge Discovery Process
What Can Data Mining Do
Example: Moviegoer Database
Data Preprocessing
Data Cleaning
Dealing with Missing Values
Smoothing Noisy Data
Data Integration
Normalization
Data Reduction
Data Cube Aggregation
Dimensionality Reduction
Decision Tree Induction
Numerocity Reduction
Discretization
Dealing with Nominal Attributes
Dividing the Data
Training Sets
Test and Evaluation Sets
Cross Validation
Bootstrap Validation
Measuring Effectiveness
Ordinal or Nominal Outputs
Measuring Effectiveness: Lift
Market Basket Analysis
What Is Association Mining?
Basic Concepts
Different Kinds of Association Rules
Support and Confidence
Improvement
Key Steps in Association Rule Discovery
Apriori Algorithm
Apriori Algorithm - An Example
Generating Association Rules from Frequent Itemsets
Multiple-Level Association Rules
Mining Multi-Level Associations
Quantitative Association Rules
Mapping Quantitative to Boolean
MBA in Text / Web Content Mining
MBA in Web Usage Mining
Web Usage Mining: Example
Tools: Magnum Opus
Magnum Opus (cont.)
Tools: Weka Package
Weka: ARFF Format
PPT Slide
Weka Discretization Filter
Weka Association Rules
Email: mobasher@cs.depaul.edu
Home Page: http://maya.cs.depaul.edu/~mobasher/classes/cs589
Download presentation source