Machine Learning in Action

Machine Learning in Action, best machine learning book.Machine learning in actionPETER HARRINGTONANNINGShelter islandFor online information and ordering of this and other Manning books, please visitwww.manning.com.ThepublisheroffersdiscountsonthisbookwhenorderedinquantityFor more information, please contactSpecial sales departmentManning publications co20 Baldwin RoadPO Box 261Shelter island. ny11964Emailorders@manning.com@2012 by manning publications Co. All rights reservedNo part of this publication may be reproduced, stored in a retrieval system, or transmitted,inany form or by means electronic, mechanical, photocopying, or otherwise, without prior writtenpermission of the publisher.Many of the designations used by manufacturers and sellers to distinguish their products arelaimed as trademarks. Where those designations appear in the book, and manningPublications was aware of a trademark claim, the designations have been printed in initial capsor all capsRecognizing the importance of preserving what has been written, it is Mannings policy to havethe books we publish printed on acid-free paper, and we exert our best efforts to that endRecognizing also our responsibility to conserve the resources of our planet, Manning books areprinted on paper that is at least 15 percent recycled and processed without the use of elementalchlorineManning Publications Co Development editor: Jeff Bleiel20 Baldwin Road Technical proofreaders: Tricia Hoffman, Alex OttPO BOX 261Copyeditor: Linda recktenwaldShelter island. ny11964Proofreader: Maureen SpencerTypesetter: Gordan SalinovicCover designer: Marija TudorISBN9781617290183Printed in the united states of america12345678910-MAL-171615141312To Joseph and milobrief contentsPART 1 CLASSIFICATIONI Machine learning basics 32 Classifying with k-Nearest Neighbors 188 Splitting datasets one feature at a time: decision trees 374 Classifying with probability theory: naive Bayes 615 Logistic regression 836 Support vector machines 101mproving classification with the Adaboostmeta-algorithm 129PaRT 2 FORECASTING NUMERIC VALUES WITH REGRESSION1518 Predicting numeric values: regression 153Tree-based regression 179PART 3 UNSUPERVISED LEARNING205Grouping unlabeled items using k-means clustering 207Association analysis with the Apriori algorithm 2242 Efficiently finding frequent itemsets with FP-growth 248BRIEF CONTENTSPART4 ADDITIONAL TOOLS。。。2678 Using principal component analysis to simplify data 26914 Simplifying data with the singular valuedecomposition 28015 Big data and MapReduce 299contentseface xunuacknowledgments xixabout this book xxiabout the author xxvabout the cover illustration xxviPARTI CLASSIFICATIONMachine learning basics 3What is machine learning? 5Sensors and the data deluge 6- Machine learning will be moreimportant in the future1.2 Key terminology 71. 3 Key tasks of machine learning 101.4 How to choose the right algorithm 111. 5 Steps in developing a machine learning application1.6 Why Python? 13Executable pseudo-code13· Python is potala13·WhatPython has that other languages dont have 14. Drawbacks 141.7 Getting started with the num Py library 15Summary