.

0 (hodnocen0 x )

BK

New York : Springer, [2013]

xiii, 600 stran : ilustrace (některé barevné) ; 24 cm

ISBN 978-1-4614-6848-6 (vázáno)

Obsahuje bibliografii na stranách 569-587, bibliografické odkazy a rejstříky

Popsáno podle 5. dotisku z roku 2016

001458627

Contents // 1 Introduction... 1 // 1.1 Prediction Versus Interpretation... 4 // 1.2 Key Ingredients of Predictive Models ... 5 // 1.3 Terminology... 6 // 1.4 Example Data Sets and Typical Data Scenarios... 7 // 1.5 Overview... 14 // 1.6 Notation... 15 // Part I General Strategies // 2 A Short Tour of the Predictive Modeling Process... 19 // 2.1 Case Study: Predicting Fuel Economy... 19 // 2.2 Themes... 24 // 2.3 Summary... 26 // 3 Data Pre-processing... 27 // 3.1 Case Study: Cell Segmentation in High-Content Screening ... 28 // 3.2 Data Transformations for Individual Predictors... 30 // 3.3 Data Transformations for Multiple Predictors... 33 // 3.4 Dealing with Missing Values... 41 // 3.5 Removing Predictors... 43 // 3.6 Adding Predictors... 47 // 3.7 Binning Predictors... 49 // 3.8 Computing... 51 // Exercises ... 58 // 4 Over-Fitting and Model Tuning... 61 // 4.1 The Problem of Over-Fitting... 62 // 4.2 Model Tuning... 64 // 4.3 Data Splitting ... 67 // 4.4 Resampling Techniques... 69 // ІХ // x Contents // 4.5 Case Study: Credit Scoring... 73 // 4.6 Choosing Final Tuning Parameters... 74 // 4.7 Data Splitting Recommendations... 77 // 4.8 Choosing Between Models... 78 // 4.9 Computing... 80 // Exercises... 89 // Part II Regression Models // 5 Measuring Performance in Regression Models... 95 // 5.1 Quantitative Measures of Performance... 95 // 5.2 The Variance-Bias Trade-off... 97 // 5.3 Computing... 98 // 6 Linear Regression and Its Cousins...101 // 6.1 Case Study: Quantitative

Structure-Activity Relationship // Modeling...102 // 6.2 Linear Regression ...105 // 6.3 Partial Least Squares...112 // 6.4 Penalized Models...122 // 6.5 Computing...128 // Exercises ...137 // 7 Nonlinear Regression Models...141 // 7.1 Neural Networks...141 // 7.2 Multivariate Adaptive Regression Splines...145 // 7.3 Support Vector Machines...151 // 7.4 AT-Nearest Neighbors ...159 // 7.5 Computing...161 // Exercises ...168 // 8 Regression Trees and Rule-Based Models...173 // 8.1 Basic Regression Trees...175 // 8.2 Regression Model Trees...184 // 8.3 Rule-Based Models...190 // 8.4 Bagged TYees...192 // 8.5 Random Forests...198 // 8.6 Boosting...203 // 8.7 Cubist...208 // 8.8 Computing...212 // Exercises ...218 // Contents Xl // 9 A Summary of Solubility Models...221 // 10 Case Study: Compressive Strength of Concrete // Mixtures ...225 // 10.1 Model Building Strategy...229 // 10.2 Model Performance...230 // 10.3 Optimizing Compressive Strength...233 // 10.4 Computing...236 // Part III Classification Models // 11 Measuring Performance in Classification Models...247 // 11.1 Class Predictions...247 // 11.2 Evaluating Predicted Classes...254 // 11.3 Evaluating Class Probabilities...262 // 11.4 Computing...266 // 12 Discriminant Analysis and Other Linear Classification Models...275 // 12.1 Case Study: Predicting Successful Grant Applications...275 // 12.2 Logistic Regression...282 // 12.3 Linear Discriminant Analysis...287 // 12.4 Partial Least Squares Discriminant Analysis...297

12.5 Penalized Models...302 // 12.6 Nearest Shrunken Centroids...306 // 12.7 Computing...308 // Exercises ...326 // 13 Nonlinear Classification Models...329 // 13.1 Nonlinear Discriminant Analysis...329 // 13.2 Neural Networks...333 // 13.3 Flexible Discriminant Analysis...338 // 13.4 Support Vector Machines...343 // 13.5 /C-Nearest Neighbors...350 // 13.6 Naive Bayes...353 // 13.7 Computing...358 // Exercises ...366 // 14 Classification Trees and Rule-Based Models...369 // 14.1 Basic Classification Trees...370 // 14.2 Rule-Based Models...383 // 14.3 Bagged Trees...385 // 14.4 Random Forests...386 // 14.5 Boosting...389 // 14.6 C5.0...392 // 4.7 Comparing Two Encodings of Categorical Predictors...400 // 4.8 Computing...400 // Ixercises ...411 //  Summary of Grant Application Models ...415 // temedies for Severe Class Imbalance...419 // 6.1 Case Study: Predicting Caravan Policy Ownership...419 // 6.2 The Effect of Class Imbalance...420 // 6.3 Model Tuning...423 // 6.4 Alternate Cutoffs...423 // 6.5 Adjusting Prior Probabilities...426 // 6.6 Unequal Case Weights...426 // 6.7 Sampling Methods...427 // .6.8 Cost-Sensitive Training ...429 // .6.9 Computing...435 // Sxercises ...442 // Hase Study: Job Scheduling ...445 // 17.1 Data Splitting and Model Strategy...450 // 17.2 Results...454 // 17.3 Computing...457 // IV Other Considerations // Measuring Predictor Importance...463 // 18.1 Numeric Outcomes...464 // 18.2 Categorical Outcomes...468 // 18.3 Other Approaches...472 // 18.4 Computing...478

// Exercises ...484 // An Introduction to Feature Selection...487 // 19.1 Consequences of Using Non-informative Predictors...488 // 19.2 Approaches for Reducing the Number of Predictors...490 // 19.3 Wrapper Methods...491 // 19.4 Filter Methods...499 // 19.5 Selection Bias...500 // 19.6 Case Study: Predicting Cognitive Impairment ...502 // 19.7 Computing...511 // Exercises ...518 // 20 Factors That Can Affect Model Performance...521 // 20.1 Type III Errors...522 // 20.2 Measurement Error in the Outcome...524 // 20.3 Measurement Error in the Predictors...527 // 20.4 Discretizing Continuous Outcomes...531 // 20.5 When Should You Trust Your Model’s Prediction?...534 // 20.6 The Impact of a Large Sample...538 // 20.7 Computing...541 // Exercises ...542 // Appendix // A A Summary of Various Models ...549 // В An Introduction to R...551 // B.l Start-Up and Getting Help...551 // B.2 Packages...552 // B.3 Creating Objects...553 // B.4 Data Types and Basic Structures...554 // B.5 Working with Rectangular Data Sets...558 // B.6 Objects and Classes...560 // B.7 R Functions...561 // B.8 The Three Faces of =...562 // B.9 The AppliedPredictiveModeling Package...562 // B.10 The caret Package...563 // B.ll Software Used in this Text...565 // С Interesting Web Sites...567 // References...569 // Indicies // Computing...591 // General...595