Hierarchical Feature Selection For Knowledge Discovery

This book is the first work that systematically describes the procedure of data mining and knowledge discovery on Bioinformatics databases by using the state-of-the-art hierarchical feature selection algorithms. The novelties of this book are three-fold. To begin with, this book discusses the hierarchical feature selection in depth, which is generally a novel research area in Data Mining/Machine Learning. Seven different state-of-the-art hierarchical feature selection algorithms are discussed and evaluated by working with four types of interpretable classification algorithms (i.e. three types of Bayesian network classification algorithms and the k-nearest neighbours classification algorithm). Moreover, this book discusses the application of those hierarchical feature selection algorithms on the well-known Gene Ontology database, where the entries (terms) are hierarchically structured. Gene Ontology database that unifies the representations of gene and gene products annotation provides the resource for mining valuable knowledge about certain biological research topics, such as the Biology of Ageing. Furthermore, this book discusses the mined biological patterns by the hierarchical feature selection algorithms relevant to the ageing-associated genes. Those patterns reveal the potential ageing-associated factors that inspire future research directions for the Biology of Ageing research.

As computer power grows and data collection technologies advance, a plethora of data is generated in almost every field where computers are used. The com puter generated data should be analyzed by computers; without the aid of computing technologies, it is certain that huge amounts of data collected will not ever be examined, let alone be used to our advantages. Even with today's advanced computer technologies (e. g. , machine learning and data mining sys tems), discovering knowledge from data can still be fiendishly hard due to the characteristics of the computer generated data. Taking its simplest form, raw data are represented in feature-values. The size of a dataset can be measUJ·ed in two dimensions, number of features (N) and number of instances (P). Both Nand P can be enormously large. This enormity may cause serious problems to many data mining systems. Feature selection is one of the long existing methods that deal with these problems. Its objective is to select a minimal subset of features according to some reasonable criteria so that the original task can be achieved equally well, if not better. By choosing a minimal subset offeatures, irrelevant and redundant features are removed according to the criterion. When N is reduced, the data space shrinks and in a sense, the data set is now a better representative of the whole data population. If necessary, the reduction of N can also give rise to the reduction of P by eliminating duplicates.

Due to increasing demands for dimensionality reduction, research on feature selection has deeply and widely expanded into many fields, including computational statistics, pattern recognition, machine learning, data mining, and knowledge discovery. Highlighting current research issues, Computational Methods of Feature Selection introduces the basic concepts and principles, state-of-the-art algorithms, and novel applications of this tool. The book begins by exploring unsupervised, randomized, and causal feature selection. It then reports on some recent results of empowering feature selection, including active feature selection, decision-border estimate, the use of ensembles with independent probes, and incremental feature selection. This is followed by discussions of weighting and local methods, such as the ReliefF family, k-means clustering, local feature relevance, and a new interpretation of Relief. The book subsequently covers text classification, a new feature selection score, and both constraint-guided and aggressive feature selection. The final section examines applications of feature selection in bioinformatics, including feature construction as well as redundancy-, ensemble-, and penalty-based feature selection. Through a clear, concise, and coherent presentation of topics, this volume systematically covers the key concepts, underlying principles, and inventive applications of feature selection, illustrating how this powerful tool can efficiently harness massive, high-dimensional data and turn it into valuable, reliable information.

This textbook comprehensively covers both fundamental and advanced topics related to data science. Data science is an umbrella term that encompasses data analytics, data mining, machine learning, and several other related disciplines. The chapters of this book are organized into three parts: The first part (chapters 1 to 3) is a general introduction to data science. Starting from the basic concepts, the book will highlight the types of data, its use, its importance and issues that are normally faced in data analytics, followed by presentation of a wide range of applications and widely used techniques in data science. The second part, which has been updated and considerably extended compared to the first edition, is devoted to various techniques and tools applied in data science. Its chapters 4 to 10 detail data pre-processing, classification, clustering, text mining, deep learning, frequent pattern mining, and regression analysis. Eventually, the third part (chapters 11 and 12) present a brief introduction to Python and R, the two main data science programming languages, and shows in a completely new chapter practical data science in the WEKA (Waikato Environment for Knowledge Analysis), an open-source tool for performing different machine learning and data mining tasks. An appendix explaining the basic mathematical concepts of data science completes the book. This textbook is suitable for advanced undergraduate and graduate students as well as for industrial practitioners who carry out research in data science. They both will not only benefit from the comprehensive presentation of important topics, but also from the many application examples and the comprehensive list of further readings, which point to additional publications providing more in-depth research results or provide sources for a more detailed description of related topics. "This book delivers a systematic, carefully thoughtful material on Data Science." from the Foreword by Witold Pedrycz, U Alberta, Canada.

This book is a collection of representative and novel works done in Data Mining, Knowledge Discovery, Clustering and Classification that were originally presented in French at the EGC'2012 Conference held in Bordeaux, France, on January 2012. This conference was the 12th edition of this event, which takes place each year and which is now successful and well-known in the French-speaking community. This community was structured in 2003 by the foundation of the French-speaking EGC society (EGC in French stands for ``Extraction et Gestion des Connaissances'' and means ``Knowledge Discovery and Management'', or KDM). This book is intended to be read by all researchers interested in these fields, including PhD or MSc students, and researchers from public or private laboratories. It concerns both theoretical and practical aspects of KDM. The book is structured in two parts called ``Knowledge Discovery and Data Mining'' and ``Classification and Feature Extraction or Selection''. The first part (6 chapters) deals with data clustering and data mining. The three remaining chapters of the second part are related to classification and feature extraction or feature selection.

With advanced computer technologies and their omnipresent usage, data accumulates in a speed unmatchable by the human's capacity to process data. To meet this growing challenge, the research community of knowledge discovery from databases emerged. The key issue studied by this community is, in layman's terms, to make advantageous use of large stores of data. In order to make raw data useful, it is necessary to represent, process, and extract knowledge for various applications. Feature Selection for Knowledge Discovery and Data Mining offers an overview of the methods developed since the 1970s and provides a general framework in order to examine these methods and categorize them. This book employs simple examples to show the essence of representative feature selection methods such as Bayes Feature Selector, Class Association rule-Information Gain feature selector and Bayes Theorem-Information Gain Feature Selector and compares them using data sets with combination of intrinsic properties according to the objective of feature selection. In addition, the book suggests guidelines on how to use different methods under various circumstances.

This SpringerBrief covers the technical material related to large scale hierarchical classification (LSHC). HC is an important machine learning problem that has been researched and explored extensively in the past few years. In this book, the authors provide a comprehensive overview of various state-of-the-art existing methods and algorithms that were developed to solve the HC problem in large scale domains. Several challenges faced by LSHC is discussed in detail such as: 1. High imbalance between classes at different levels of the hierarchy 2. Incorporating relationships during model learning leads to optimization issues 3. Feature selection 4. Scalability due to large number of examples, features and classes 5. Hierarchical inconsistencies 6. Error propagation due to multiple decisions involved in making predictions for top-down methods The brief also demonstrates how multiple hierarchies can be leveraged for improving the HC performance using different Multi-Task Learning (MTL) frameworks. The purpose of this book is two-fold: 1. Help novice researchers/beginners to get up to speed by providing a comprehensive overview of several existing techniques. 2. Provide several research directions that have not yet been explored extensively to advance the research boundaries in HC. New approaches discussed in this book include detailed information corresponding to the hierarchical inconsistencies, multi-task learning and feature selection for HC. Its results are highly competitive with the state-of-the-art approaches in the literature.