Feature Selection and Ensemble Methods for Bioinformatics : Algorithmic Classification and Implementations

Machine learning is the branch of artificial intelligence whose goal is to develop algorithms that add learning capabilities to computers. Ensembles are an integral part of machine learning. A typical ensemble includes several algorithms performing the task of prediction of the class label or the degree of class membership for a given input presented as a set of measurable characteristics, often called features. Feature Selection and Ensemble Methods for Bioinformatics: Algorithmic Classification and Implementations offers a unique perspective on machine learning aspects of microarray gene expression based cancer classification. This multidisciplinary text is at the intersection of computer science and biology and, as a result, can be used as a reference book by researchers and students from both fields. Each chapter describes the process of algorithm design from beginning to end and aims to inform readers of best practices for use in their own research. Title......Page 2 Copyright Page......Page 3 Table of Contents......Page 4 Preface......Page 9 Biological Background......Page 16 Gene Expression Data Sets......Page 21 Introduction to Data Classification......Page 25 Naïve Bayes......Page 28 Nearest Neighbor......Page 47 Classification Tree......Page 68 Support Vector Machines......Page 83 Introduction to Feature and Gene Selection......Page 132 Feature Selection Based on Elements of Game Theory......Page 138 Kernel-Based Feature Selection with the Hilbert-Schmidt Independence Criterion......Page 155 Extreme Value Distribution Based Gene Selection......Page 174 Evolutionary Algorithm for Identifying Predictive Genes......Page 192 Redundancy-Based Feature Selection......Page 218 Unsupervised Feature Selection......Page 238 Differential Evolution for Finding Predictive Gene Subsets......Page 251 Ensembles of Classifiers......Page 267 Classifier Ensembles Built on Subsets of Features......Page 275 Bagging and Random Forests......Page 311 Boosting and AdaBoost......Page 329 Ensemble Gene Selection......Page 344 Introduction to Classification Error Estimation......Page 349 ROC Curve, Area under it, other Classification Performance Characteristics and Statistical Tests......Page 356 Bolstered Resubstitution Error......Page 398 Performance Evaluation......Page 421 Application Examples......Page 429 End Remarks......Page 451 About the Contributors......Page 454 Index......Page 455 "This book offers a unique perspective on machine learning aspects of microarray gene expression based cancer classification, combining computer science, and biology"--Provided by publisher.