Practical Machine Learning Cookbook

图书信息

作者	Atul Tripathi
出版社	Packt Publishing
ISBN	9781785286537
出版时间	2017-04-14
字数	271.1万
分类	进口书,外文原版书,电脑,网络

读书简介

Machine learning has become the new black. The challenge in today's world is the explosion of data from existing legacy data and incoming new structured and unstructured data. The complexity of discovering, understanding, performing analysis, and predicting outcomes on the data using machine learning algorithms is a challenge. This cookbook will help solve everyday challenges you face as a data scientist. The application of various data science techniques and on multiple data sets based on real-world challenges you face will help you appreciate a variety of techniques used in various situations. The first half of the book provides recipes on fairly complex machine-learning systems, where you'll learn to explore new areas of applications of machine learning and improve its efficiency. That includes recipes on classifications, neural networks, unsupervised and supervised learning, deep learning, reinforcement learning, and more. The second half of the book focuses on three different machine learning case studies, all based on real-world data, and offers solutions and solves specific machine-learning issues in each one. What You Will Learn ?Get equipped with a deeper understanding of how to apply machine-learning techniques ?Implement each of the advanced machine-learning techniques ?Solve real-life problems that are encountered in order to make your applications produce improved results ?Gain hands-on experience in problem solving for your machine-learning systems ?Understand the methods of collecting data, preparing data for usage, training the model, evaluating the model's performance, and improving the model's performance

目录

Practical Machine Learning Cookbook

Practical Machine Learning Cookbook

Credits

About the Author

About the Reviewer

www.PacktPub.com

Why subscribe?

Customer Feedback

Preface

What this book covers

What you need for this book

Who this book is for

Sections

Getting ready

How to do it…

Conventions

Reader feedback

Customer support

Downloading the example code

Downloading the color images of this book

Errata

Piracy

Questions

1. Introduction to Machine Learning

What is machine learning?

An overview of classification

An overview of clustering

An overview of supervised learning

An overview of unsupervised learning

An overview of reinforcement learning

An overview of structured prediction

An overview of neural networks

An overview of deep learning

2. Classification

Introduction

Discriminant function analysis - geological measurements on brines from wells

Getting ready

Step 1 - collecting and describing data

How to do it...

Step 2 - exploring data

Step 3 - transforming data

Step 4 - training the model

Step 5 - classifying the data

Step 6 - evaluating the model

Multinomial logistic regression - understanding program choices made by students

Getting ready

Step 1 - collecting data

How to do it...

Step 2 - exploring data

Step 3 - training the model

Step 4 - testing the results of the model

Step 5 - model improvement performance

Tobit regression - measuring the students' academic aptitude

Getting ready

Step 1 - collecting data

How to do it...

Step 2 - exploring data

Step 3 - plotting data

Step 4 - exploring relationships

Step 5 - training the model

Step 6 - testing the model

Poisson regression - understanding species present in Galapagos Islands

Getting ready

Step 1 - collecting and describing the data

How to do it...

Step 2 - exploring the data

Step 3 - plotting data and testing empirical data

Step 4 - rectifying discretization of the Poisson model

Step 5 - training and evaluating the model using the link function

Step 6 - revaluating using the Poisson model

Step 7 - revaluating using the linear model

3. Clustering

Introduction

Hierarchical clustering - World Bank sample dataset

Getting ready

Step 1 - collecting and describing data

How to do it...

Step 2 - exploring data

Step 3 - transforming data

Step 4 - training and evaluating the model performance

Step 5 - plotting the model

Hierarchical clustering - Amazon rainforest burned between 1999-2010

Getting ready

Step 1 - collecting and describing data

How to do it...

Step 2 - exploring data

Step 3 - transforming data

Step 4 - training and evaluating model performance

Step 5 - plotting the model

Step 6 - improving model performance

Hierarchical clustering - gene clustering

Getting ready

Step 1 - collecting and describing data

How to do it...

Step 2 - exploring data

Step 3 - transforming data

Step 4 - training the model

Step 5 - plotting the model

Binary clustering - math test

Getting ready

Step 1 - collecting and describing data

How to do it...

Step 2 - exploring data

Step 3 - training and evaluating model performance

Step 4 - plotting the model

Step 5 - K-medoids clustering

K-means clustering - European countries protein consumption

Getting ready

Step 1 - collecting and describing data

How to do it...

Step 2 - exploring data

Step 3 - clustering

Step 4 - improving the model

K-means clustering - foodstuff

Getting ready

Step 1 - collecting and describing data

How to do it...

Step 2 - exploring data

Step 3 - transforming data

Step 4 - clustering

Step 5 - visualizing the clusters

4. Model Selection and Regularization

Introduction

Shrinkage methods - calories burned per day

Getting ready

Step 1 - collecting and describing data

How to do it...

Step 2 - exploring data

Step 3 - building the model

Step 4 - improving the model

Step 5 - comparing the model

Dimension reduction methods - Delta's Aircraft Fleet

Getting ready

Step 1 - collecting and describing data

How to do it...

Step 2 - exploring data

Step 3 - applying principal components analysis

Step 4 - scaling the data

Step 5 - visualizing in 3D plot

Principal component analysis - understanding world cuisine

Getting ready

Step 1 - collecting and describing data

How to do it...

Step 2 - exploring data

Step 3 - preparing data

Step 4 - applying principal components analysis

5. Nonlinearity

Generalized additive models - measuring the household income of New Zealand

Getting ready

Step 1 - collecting and describing data

How to do it...

Step 2 - exploring data

Step 3 - setting up the data for the model

Step 4 - building the model

Smoothing splines - understanding cars and speed

How to do it...

Step 1 - exploring the data

Step 2 - creating the model

Step 3 - fitting the smooth curve model

Step 4 - plotting the results

Local regression - understanding drought warnings and impact

Getting ready

Step 1 - collecting and describing data

How to do it...

Step 2 - collecting and exploring data

Step 3 - calculating the moving average

Step 4 - calculating percentiles

Step 5 - plotting results

6. Supervised Learning

Introduction

Decision tree learning - Advance Health Directive for patients with chest pain

Getting ready

Step 1 - collecting and describing the data

How to do it...

Step 2 - exploring the data

Step 3 - preparing the data

Step 4 - training the model

Step 5- improving the model

Decision tree learning - income-based distribution of real estate values

Getting ready

Step 1 - collecting and describing the data

How to do it...

Step 2 - exploring the data

Step 3 - training the model

Step 4 - comparing the predictions

Step 5 - improving the model

Decision tree learning - predicting the direction of stock movement

Getting ready

Step 1 - collecting and describing the data

How to do it...

Step 2 - exploring the data

Step 3 - calculating the indicators

Step 4 - preparing variables to build datasets

Step 5 - building the model

Step 6 - improving the model

Naive Bayes - predicting the direction of stock movement

Getting ready

Step 1 - collecting and describing the data

How to do it...

Step 2 - exploring the data

Step 3 - preparing variables to build datasets

Step 4 - building the model

Step 5 - creating data for a new, improved model

Step 6 - improving the model

Random forest - currency trading strategy

Getting ready

Step 1 - collecting and describing the data

How to do it...

Step 2 - exploring the data

Step 3 - preparing variables to build datasets

Step 4 - building the model

Support vector machine - currency trading strategy

Getting ready

Step 1 - collecting and describing the data

How to do it...

Step 2 - exploring the data

Step 3 - calculating the indicators

Step 4 - preparing variables to build datasets

Step 5 - building the model

Stochastic gradient descent - adult income

Getting ready

Step 1 - collecting and describing the data

How to do it...

Step 2 - exploring the data

Step 3 - preparing the data

Step 4 - building the model

Step 5 - plotting the model

7. Unsupervised Learning

Introduction

Self-organizing map - visualizing of heatmaps

How to do it...

Step 1 - exploring data

Step 2 - training the model

Step 3 - plotting the model

Vector quantization - image clustering

Getting ready

Step 1 - collecting and describing data

How to do it...

Step 2 - exploring data

Step 3 - data cleaning

Step 4 - visualizing cleaned data

Step 5 - building the model and visualizing it

8. Reinforcement Learning

Introduction

Markov chains - the stocks regime switching model

Getting ready

Step 1 - collecting and describing the data

How to do it...

Step 2 - exploring the data

Step 3 - preparing the regression model

Step 4 - preparing the Markov-switching model

Step 5 - plotting the regime probabilities

Step 6 - testing the Markov switching model

Markov chains - the multi-channel attribution model

Getting ready

How to do it...

Step 1 - preparing the dataset

Step 2 - preparing the model

Step 3 - plotting the Markov graph

Step 4 - simulating the dataset of customer journeys

Step 5 - preparing a transition matrix heat map for real data

Markov chains - the car rental agency service

How to do it...

Step 1 - preparing the dataset

Step 2 - preparing the model

Step 3 - improving the model

Continuous Markov chains - vehicle service at a gas station

Getting ready

How to do it...

Step 1 - preparing the dataset

Step 2 - computing the theoretical resolution

Step 3 - verifying the convergence of a theoretical solution

Step 4 - plotting the results

Monte Carlo simulations - calibrated Hull and White short-rates

Getting ready

Step 1 - installing the packages and libraries

How to do it...

Step 2 - initializing the data and variables

Step 3 - pricing the Bermudan swaptions

Step 4 - constructing the spot term structure of interest rates

Step 5 - simulating Hull-White short-rates

9. Structured Prediction

Introduction

Hidden Markov models - EUR and USD

Getting ready

Step 1 - collecting and describing data

How to do it...

Step 2 - exploring data

Step 3 - turning data into a time series

Step 4 - building the model

Step 5 - displaying the results

Hidden Markov models - regime detection

Getting ready

Step 1 - collecting and describing data

How to do it...

Step 2 - exploring data

Step 3 - preparing the model

10. Neural Networks

Introduction

Modelling SP 500

Getting ready

Step 1 - collecting and describing data

How to do it...

Step 2 - exploring data

Step 3 - calculating the indicators

Step 4 - preparing data for model building

Step 5 - building the model

Measuring the unemployment rate

Getting ready

Step 1 - collecting and describing data

How to do it...

Step 2 - exploring data

Step 3 - preparing and verifying the models

Step 4 - forecasting and testing the accuracy of the models built

11. Deep Learning

Introduction

Recurrent neural networks - predicting periodic signals

Getting ready...

How to do it...

12. Case Study - Exploring World Bank Data

Introduction

Exploring World Bank data

Getting ready...

Step 1 - collecting and describing data

How to do it...

Step 2 - downloading the data

Step 3 - exploring data

Step 4 - building the models

Step 5 - plotting the models

13. Case Study - Pricing Reinsurance Contracts

Introduction

Pricing reinsurance contracts

Getting ready...

Step 1 - collecting and describing the data

How to do it...

Step 2 - exploring the data

Step 3 - calculating the individual loss claims

Step 4 - calculating the number of hurricanes

Step 5 - building predictive models

Step 6 - calculating the pure premium of the reinsurance contract

14. Case Study - Forecast of Electricity Consumption

Introduction

Getting ready

Step 1 - collecting and describing data

How to do it...

Step 2 - exploring data

Step 3 - time series - regression analysis

Step 4 - time series - improving regression analysis

Step 5 - building a forecasting model

Step 6 - plotting the forecast for a year