Real-World Machine Learning 无水印pdf

【实例简介】
Real-World Machine Learning 英文无水印pdf pdf所有页面使用FoxitReader和PDF-XChangeViewer测试都可以打开 本资源转载自网络，如有侵权，请联系上传者或csdn删除 本资源转载自网络，如有侵权，请联系上传者或csdn删除
Real-World Machine learning HENRIK BRINK JOSEPH W. RICHARDS MARK FETHEROLF MANNING SHELTER ISLAND For online information and ordering of this and other Manning books, please visit www.manning.com.Thepublisheroffersdiscountsonthisbookwhenorderedinquantity For more information, please contact Special sales department Manning publications co 20 Baldwin road PO BoX 761 Shelter island. ny11964 Email:orders@manning.com @2017 by manning publications Co. All rights reserved no part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by means electronic, mechanical, photocopying, or otherwise, without prior written permission of the publisher Many of the designations used by manufacturers and sellers to distinguish their products are claimed as trademarks. Where those designations appear in the book, and manning Publications was aware of a trademark claim, the designations have been printed in initial caps or all caps @o Recognizing the importance of preserving what has been written, it is Mannings policy to have the books we publish printed on acid-free paper, and we exert our best efforts to that end Recognizing also our responsibility to conserve the resources of our planet, Manning books are printed on paper that is at least 15 percent recycled and processed without the use of elemental chlorine Manning publications co Development editor: Susanna Kline 20 Baldwin road Technical development editor: Al Scherer POBoⅹ761 Review editors: Olivia booth. Ozren harlovic Shelter island. ny11964 Project editor: Kevin Sullivan Copyeditor: Sharon Wilkey Proofreader: Katie Tennant Technical proofreader: Valentin Crettaz Typesetter: Dennis Dalinnik Cover designer Marija Tudor ISBN:9781617291920 Printed in the united states of america 12345678910-EBM-212019181716 brief contents PART 1 THE MACHINE-LEARNING WORKFLOW 1 What is machine learning? 3 Real-world data 27 3 Modeling and prediction 52 4 Model evaluation and optimization 77 5 Basic feature engineering 106 PART 2 PRACTICAL APPLICATION 127 6 Example: Nyc taxi data 129 7 Advanced feature engineering 146 8 Advanced nlp example: movie review sentiment 172 Scaling machine-learning workflows 196 10 Example: digital display advertising 214 contents preface xu acknowledgments xvii this book about the authors xx about the cover illustration xxii PART 1 THE MACHINE-LEARNING WORKFLOW What is machine learning? 3 1.1 Understanding how machines learn 4 1.2 Using data to make decisions 7 Traditional approaches 8 The machine-leaning approach 11 Five advantages to machine learning 16 Challenges 16 1. 8 Following the Ml workflow: from data to deployment 17 Data collection and preparation 18 Learning a model from data 19. Evaluating model performance 20 Optimizing model performance 21 CoNTENTS 1. Boosting model performance with advanced techniques 22 Data preprocessing and feature engineering 22 Improving models continually with online methods 24 Scaling models with data volume and velocity 25 1.5 Summary 25 1.6 Terms from this chapter 25 Real-world data 27 2.1 Getting started: data collection 28 Which features should be included? 30. How can we obtain ground truth for the target variable? 32 How much training data is required? 33 Is the training set representative enough? 35 Preprocessing the data for modeling 36 Categorical features 36. Dealing with missing data 38 Simple feature engineering 40. Data normalization 42 2.3 Using data visualization 43 Mosaic plots44· Box plots46· Density plots8 Scatter plots 50 2.4 Summary 50 2.5 Terms from this chapter 51 Modeling and prediction 52 8.1 Basic machine-learning modeling 58 Finding the relationship between input and target 53 The purpose of finding a good model 55. Types of modeling methods 56 Supervised versus unsupervised learning 58 8.2 Classification: predicting into buckets 59 Building a classifier and making predictions 61 Classifying complex, nonlinear data 64 Classifying with multiple classes 66 3.3 Regression: predicting numerical values 68 Building a regressor and making predictions 69 Performing regression on complex, nonlinear data 73 3.4 Summary 74 3.5 Terms from this chapter 75 CONTENTS 4 Model evaluation and optimization 77 4.1 Model generalization: assessing predictive accuracy for new d 78 The problem: overfitting and model optimism 79 a The solution cross-validation 82 Some things to look out for when using cross-validation 86 4.2 Evaluation of classification models 87 Class-wise accuracy and the confusion matrix 89 Accuracy trade-offs and RoC curves 90 Multiclass classification 93 4.3 Evaluation of regression models 96 Using simple regression performance metrics 97 Examining residuals gg 4. Model optimization through parameter tuning 100 Ml algorithms and their tuning parameters 100 Grid search 101 4.5 Summary 104 4.6 Terms from this chapter 105 Basic feature engineering 106 5.1 Motivation: why is feature engineering useful? 107 What is feature engineering? 107. Five reasons to use feature engineering 107. Feature engineering and domain expertise 109 5.2 Basic feature-engineering processes 110 Example: event recommendation 110. Handling date and time features 112. Working with simple text features 114 5. 3 Feature selection 116 Forward selection and backward elimination 119- Feature selection for data exploration 121" Real-world feature le123 5. Summary 125 5.5 Terms from this chapter 126 【实例截图】
【核心代码】