Data Mining Methods and Models

DANIEL T. LAROSE, PhD, received his PhD in statistics from the University of Connecticut. An associate professor of statistics at Central Connecticut State University, he developed and directs Data Mining@CCSU, the world's first online master of science program in data mining. He has also worked as a data mining consultant for Connecticut-area companies. He is the author of Discovering Knowledge in Data: An Introduction to Data Mining (Wiley), and is currently working on the third book of his three-volume set on data mining: Data Mining the Web: Uncovering Patterns in Web Content, Structure, and Usage (with Zdravko Markov, PhD), scheduled to be published by Wiley in 2006.

Preface. 1. Dimension Reduction Methods. Need for Dimension Reduction in Data Mining. Principal Components Analysis. Factor Analysis. User-Defined Composites. 2. Regression Modeling. Example of Simple Linear Regression. Least-Squares Estimates. Coefficient or Determination. Correlation Coefficient. The ANOVA Table. Outliers, High Leverage Points, and Influential Observations. The Regression Model. Inference in Regression. Verifying the Regression Assumptions. An Example: The Baseball Data Set. An Example: The California Data Set. Transformations to Achieve Linearity. 3. Multiple Regression and Model Building. An Example of Multiple Regression. The Multiple Regression Model. Inference in Multiple Regression. Regression with Categorical Predictors. Multicollinearity. Variable Selection Methods. An Application of Variable Selection Methods. Mallows' C p Statistic. Variable Selection Criteria. Using the Principal Components as Predictors in Multiple Regression. 4. Logistic Regression. A Simple Example of Logistic Regression. Maximum Likelihood Estimation. Interpreting Logistic Regression Output. Inference: Are the Predictors Significant? Interpreting the Logistic Regression Model. Interpreting a Logistic Regression Model for a Dichotomous Predictor. Interpreting a Logistic Regression Model for a Polychotomous Predictor. Interpreting a Logistic Regression Model for a Continuous Predictor. The Assumption of Linearity. The Zero-Cell Problem. Multiple Logistic Regression. Introducing Higher Order terms to Handle Non-Linearity. Validating the Logistic Regression Model. WEKA: Hands-On Analysis Using Logistic Regression. 5. Naive Bayes and Bayesian Networks. The Bayesian Approach. The Maximum a Posteriori (MAP) Classification. The Posterior Odds Ratio. Balancing the Data. Naive Bayes Classification. Numeric Predictors for Naive Bayes Classification. WEKA: Hands-On Analysis Using Naive Bayes. Bayesian Belief Networks. Using the Bayesian Network to Find Probabilities. WEKA: Hands-On Analysis Using Bayes Net. 6. Genetic Algorithms. Introduction to Genetic Algorithms. The Basic Framework of a Genetic Algorithm. A Simple Example of Genetic Algorithms at Work. Modifications and Enhancements: Selection. Modifications and enhancements: Crossover. Genetic Algorithms for Real-Valued Variables. Using Genetic Algorithms to Train a Neural Network. WEKA: Hands-On Analysis Using Genetic Algorithms. 7. Case Study: Modeling Response to Direct-Mail Marketing. The Cross-Industry Standard Process for Data Mining: CRISP-DM. Business Understanding Phase. Data Understanding and Data Preparation Phases. The Modeling Phase and the Evaluation Phase. Index.