Statistics for Data Science: Leverage the power of statistics for Data Analysis, Classification, Regression, Machine Learning, and Neural Networks

Statistics for Data Science

Leverage the power of statistics for Data Analysis, Classification, Regression, Machine Learning, and Neural Networks

James D. Miller

BIRMINGHAM - MUMBAI

Statistics for Data Science

Copyright © 2017 Packt Publishing

All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, without the prior written permission of the publisher, except in the case of brief quotations embedded in critical articles or reviews.

Every effort has been made in the preparation of this book to ensure the accuracy of the information presented. However, the information contained in this book is sold without warranty, either express or implied. Neither the author, nor Packt Publishing, and its dealers and distributors will be held liable for any damages caused or alleged to be caused directly or indirectly by this book.

Packt Publishing has endeavored to provide trademark information about all of the companies and products mentioned in this book by the appropriate use of capitals. However, Packt Publishing cannot guarantee the accuracy of this information.

First published: November 2017

Production reference: 1151117

Published by Packt Publishing Ltd.

Livery Place

35 Livery Street

Birmingham

B3 2PB, UK.

ISBN 978-1-78829-067-8

www.packtpub.com

Credits

About the Author

James D. Miller, is an IBM certified expert, creative innovator and accomplished Director, Sr. Project Leader and Application/System Architect with +35 years of extensive applications and system design and development experience across multiple platforms and technologies. Experiences include introducing customers to new and sometimes disruptive technologies and platforms, integrating with IBM Watson Analytics, Cognos BI, TM1 and web architecture design, systems analysis, GUI design and testing, database modelling and systems analysis, design and development of OLAP, client/server, web and mainframe applications and systems utilizing: IBM Watson Analytics, IBM Cognos BI and TM1 (TM1 rules, TI, TM1Web and Planning Manager), Cognos Framework Manager, dynaSight-ArcPlan, ASP, DHTML, XML, IIS, MS Visual Basic and VBA, Visual Studio, PERL, SPLUNK, WebSuite, MS SQL Server, ORACLE, SYBASE Server, and so on.

Responsibilities have also included all aspects of Windows and SQL solution development and design including analysis; GUI (and website) design; data modelling; table, screen/form and script development; SQL (and remote stored procedures and triggers) development/testing; test preparation and management and training of programming staff. Other experience includes the development of  Extract, Transform, and Load (ETL)  infrastructure such as data transfer automation between mainframe (DB2, Lawson, Great Plains, and so on.) systems and client/server SQL server and web-based applications and integration of enterprise applications and data sources.

Mr Miller has acted as Internet Applications Development Mgr. responsible for the design, development, QA and delivery of multiple websites including online trading applications, warehouse process control and scheduling systems, administrative and control applications. Mr Miller also was responsible for the design, development and administration of a web-based financial reporting system for a 450-million-dollar organization, reporting directly to the CFO and his executive team.

He has also been responsible for managing and directing multiple resources in various management roles including project and team leader, lead developer and applications development director.

He has authored the following books published by Packt:

Mastering Predictive Analytics with R – Second Edition 

Big Data Visualization 

Learning IBM Watson Analytics 

Implementing Splunk – Second Edition 

Mastering Splunk 

IBM Cognos TM1 Developer's Certification Guide 

He has also authored a number of whitepapers on best practices such as Establishing a Center of Excellence and continues to post blogs on a number of relevant topics based on personal experiences and industry best practices. 

He is a perpetual learner continuing to pursue experiences and certifications, currently holding the following current technical certifications:

IBM Certified Developer Cognos TM1

IBM Certified Analyst Cognos TM1

IBM Certified Administrator Cognos TM1

IBM Cognos TM1 Master 385 Certification

IBM Certified Advanced Solution Expert Cognos TM1

IBM OpenPages Developer Fundamentals C2020-001-ENU

IBM Cognos 10 BI Administrator C2020-622

IBM Cognos 10 BI Author C2090-620-ENU

IBM Cognos BI Professional C2090-180-ENU

IBM Cognos 10 BI Metadata Model Developer C2090-632

IBM Certified Solution Expert - Cognos BI

Specialties: The evaluation and introduction of innovative and disruptive technologies, cloud migration, IBM Watson Analytics, big data, data visualizations, Cognos BI and TM1 application design and development, OLAP, Visual Basic, SQL Server, forecasting and planning; international application, and development, business intelligence, project development, and delivery and process improvement.

To Nanette L. Miller:

Like a river flows surely to the sea, darling so it goes, some things are meant to be.

About the Reviewer

James Mott, Ph.D, is a senior education consultant with extensive experience in teaching statistical analysis, modeling, data mining and predictive analytics. He has over 30 years of experience using SPSS products in his own research including IBM SPSS Statistics, IBM SPSS Modeler, and IBM SPSS Amos. He has also been actively teaching these products to IBM/SPSS customers for over 30 years. In addition, he is an experienced historian with expertise in the research and teaching of 20th Century United States political history and quantitative methods. His specialties are data mining, quantitative methods, statistical analysis, teaching, and consulting.

www.PacktPub.com

For support files and downloads related to your book, please visit www.PacktPub.com.

Did you know that Packt offers eBook versions of every book published, with PDF and ePub files available? You can upgrade to the eBook version at www.PacktPub.com and as a print book customer, you are entitled to a discount on the eBook copy. Get in touch with us at [email protected] for more details.

At www.PacktPub.com, you can also read a collection of free technical articles, sign up for a range of free newsletters and receive exclusive discounts and offers on Packt books and eBooks.

https://www.packtpub.com/mapt

Get the most in-demand software skills with Mapt. Mapt gives you full access to all Packt books and video courses, as well as industry-leading tools to help you plan your personal development and advance your career.

Why subscribe?

Fully searchable across every book published by Packt

Copy and paste, print, and bookmark content

On demand and accessible via a web browser

Customer Feedback

Thanks for purchasing this Packt book. At Packt, quality is at the heart of our editorial process. To help us improve, please leave us an honest review on this book's Amazon page at https://www.amazon.com/dp/1788290674. If you'd like to join our team of regular reviewers, you can email us at [email protected]. We award our regular reviewers with free eBooks and videos in exchange for their valuable feedback. Help us be relentless in improving our products!

Table of Contents

Preface

What this book covers

What you need for this book

Who this book is for

Conventions

Reader feedback

Customer support

Downloading the example code

Downloading the color images of this book

Errata

Piracy

Questions

Transitioning from Data Developer to Data Scientist

Data developer thinking

Objectives of a data developer

Querying or mining

Data quality or data cleansing

Data modeling

Issue or insights

Thought process

Developer versus scientist

New data, new source

Quality questions

Querying and mining

Performance

Financial reporting

Visualizing

Tools of the trade

Advantages of thinking like a data scientist

Developing a better approach to understanding data

Using statistical thinking during program or database designing

Adding to your personal toolbox

Increased marketability

Perpetual learning

Seeing the future

Transitioning to a data scientist

Let's move ahead

Summary

Declaring the Objectives

Key objectives of data science

Collecting data

Processing data

Exploring and visualizing data

Analyzing the data and/or applying machine learning to the data

Deciding (or planning) based upon acquired insight

Thinking like a data scientist

Bringing statistics into data science

Common terminology

Statistical population

Probability

False positives

Statistical inference

Regression

Fitting

Categorical data

Classification

Clustering

Statistical comparison

Coding

Distributions

Data mining

Decision trees

Machine learning

Munging and wrangling

Visualization

D3

Regularization

Assessment

Cross-validation

Neural networks

Boosting

Lift

Mode

Outlier

Predictive modeling

Big Data

Confidence interval

Writing

Summary

A Developer's Approach to Data Cleaning

Understanding basic data cleaning

Common data issues

Contextual data issues

Cleaning techniques

R and common data issues

Outliers

Step 1 – Profiling the data

Step 2 – Addressing the outliers

Domain expertise

Validity checking

Enhancing data

Harmonization

Standardization

Transformations

Deductive correction

Deterministic imputation

Summary

Data Mining and the Database Developer

Data mining

Common techniques

Visualization

Cluster analysis

Correlation analysis

Discriminant analysis

Factor analysis

Regression analysis

Logistic analysis

Purpose

Mining versus querying

Choosing R for data mining

Visualizations

Current smokers

Missing values

A cluster analysis

Dimensional reduction

Calculating statistical significance

Frequent patterning

Frequent item-setting

Sequence mining

Summary

Statistical Analysis for the Database Developer

Data analysis

Looking closer

Statistical analysis

Summarization

Comparing groups

Samples

Group comparison conclusions

Summarization modeling

Establishing the nature of data

Successful statistical analysis

R and statistical analysis

Summary

Database Progression to Database Regression

Introducing statistical regression

Techniques and approaches for regression

Choosing your technique

Does it fit?

Identifying opportunities for statistical regression

Summarizing data

Exploring relationships

Testing significance of differences

Project profitability

R and statistical regression

A working example

Establishing the data profile

The graphical analysis

Predicting with our linear model

Step 1: Chunking the data

Step 2: Creating the model on the training data

Step 3: Predicting the projected profit on test data

Step 4: Reviewing the model

Step 4: Accuracy and error

Summary

Regularization for Database Improvement

Statistical regularization

Various statistical regularization methods

Ridge

Lasso

Least angles

Opportunities for regularization

Collinearity

Sparse solutions

High-dimensional data

Classification

Using data to understand statistical regularization

Improving data or a data model

Simplification

Relevance

Speed

Transformation

Variation of coefficients

Casual inference

Back to regularization

Reliability

Using R for statistical regularization

Parameter Setup

Summary

Database Development and Assessment

Assessment and statistical assessment

Objectives

Baselines

Planning for assessment

Evaluation

Development versus assessment

Planning

Data assessment and data quality assurance

Categorizing quality

Relevance

Cross-validation

Preparing data

R and statistical assessment

Questions to ask

Learning curves

Example of a learning curve

Summary

Databases and Neural Networks

Ask any data scientist

Defining neural network

Nodes

Layers

Training

Solution

Understanding the concepts

Neural network models and database models

No single or main node

Not serial

No memory address to store results

R-based neural networks

References

Data prep and preprocessing

Data splitting

Model parameters

Cross-validation

R packages for ANN development

ANN

ANN2

NNET

Black boxes

A use case

Popular use cases

Character recognition

Image compression

Stock market prediction

Fraud detection

Neuroscience

Summary

Boosting your Database

Definition and purpose

Bias

Categorizing bias

Causes of bias

Bias data collection

Bias sample selection

Variance

ANOVA

Noise

Noisy data

Weak and strong learners

Weak to strong

Model bias

Training and prediction time

Complexity

Which way?

Back to boosting

How it started

AdaBoost

What you can learn from boosting (to help) your database

Using R to illustrate boosting methods

Prepping the data

Training

Ready for boosting

Example results

Summary

Database Classification using Support Vector Machines

Database classification

Data classification in statistics

Guidelines for classifying data

Common guidelines

Definitions

Definition and purpose of an SVM

The trick

Feature space and cheap computations

Drawing the line

More than classification

Downside

Reference resources

Predicting credit scores

Using R and an SVM to classify data in a database

Moving on

Summary

Database Structures and Machine Learning

Data structures and data models

Data structures

Data models

What's the difference?

Relationships

Machine learning

Overview of machine learning concepts

Key elements of machine learning

Representation

Evaluation

Optimization

Types of machine learning

Supervised learning

Unsupervised learning

Semi-supervised learning

Reinforcement learning

Most popular

Applications of machine learning

Machine learning in practice

Understanding

Preparation

Learning

Interpretation

Deployment

Iteration

Using R to apply machine learning techniques to a database

Understanding the data

Preparing

Data developer

Understanding the challenge

Cross-tabbing and plotting

Summary

Preface

Statistics are an absolute must prerequisite for any task in the area of data science but may also be the most feared deterrent for developers entering into the field of data science. This book will take you on a statistical journey from knowing very little to becoming comfortable using various statistical methods for typical data science tasks.

What this book covers

Chapter 1: Transitioning from Data Developer to Data Scientist, sets the stage for the transition from data developer to data scientist. You will understand the difference between a developer mindset versus a data scientist mindset, the important difference between the two, and how to transition into thinking like a data scientist.

Chapter 2: Declaring the Objectives, introduces and explains (from a developer’s perspective) the basic objectives behind statistics for data science and introduces you to the important terms and keys that are used in the field of data science.

Chapter 3: A Developer's  Approach to Data Cleaning, discusses how a developer might understand and approach the topic of data cleaning using common statistical methods.

Chapter 4: Data Mining and the Database Developer, introduces the developer to mining data using R. You will understand what data mining is, why it is important, and feel comfortable using R for the most common statistical data mining methods: dimensional reduction, frequent patterns, and sequences.

Chapter 5: Statistical Analysis for the Database Developer, discusses the difference between data analysis or summarization and statistical data analysis and will follow the steps for successful statistical analysis of data, describe the nature of data, explore the relationships presented in data, create a summarization model from data, prove the validity of a model, and employ predictive analytics on a developed model.

Chapter 6: Database Progression to Database Regression, sets out to define statistical regression concepts and outline how a developer might use regression for simple forecasting and prediction within a typical data development project.

Chapter 7: Regularization for Database Improvement, introduces the developer to the idea of statistical regularization to improve data models. You will review what statistical regularization is, why it is important, and various statistical regularization methods.

Chapter 8: Data Development and Assessment, covers the idea of data model assessment and using statistics for assessment. You will understand what statistical assessment is, why it is important, and use R for statistical assessment.

Chapter 9: Databases and Neural Networks, defines the neural network model and draws from a developer’s knowledge of data models to help understand the purpose and use of neural networks in data science.

Chapter 10: Boosting and your Database, introduces the idea of using statistical boosting to better understand data in a database.

Chapter 11: Database Classification using Support Vector Machines, uses developer terminologies to define an SVM, identify various applications for its use and walks through an example of using a simple SVM to classify data in a database

Chapter 12: Database Structures and Machine Learning, aims to provide an explanation of the types of machine learning and shows the developer how to use machine learning processes to understand database mappings and identify patterns within the data.

What you need for this book

This book is intended for those with a data development background who are interested in possibly entering the field of data science and are looking for concise information on the topic of statistics with the help of insightful programs and simple explanation. Just bring your data development experience and an open mind!

Who this book is for

This book is intended for those developers who are interested in entering the field of data science and are looking for concise information on the topic of statistics with the help of insightful programs and simple explanation.

Conventions

In this book, you will find a number of text styles that distinguish between different kinds of information. Here are some examples of these styles and an explanation of their meaning.

Code words in text, database table names, folder names, filenames, file extensions, pathnames, dummy URLs, user input, and Twitter handles are shown as follows: In statistics, a boxplot is a simple way to gain information regarding the shape, variability, and center (or median) of a statistical data set, so we'll use the boxplot with our data to see if we can identify both the median Coin-in and if there are any outliers.

A block of code is set as follows:

MyFile <-C:/GammingData/SlotsResults.csv

MyData <- read.csv(file=MyFile, header=TRUE, sep=,)

New terms and important words are shown in bold. 

Warnings or important notes appear like this.

Tips and tricks appear like this.

Reader feedback

Feedback from our readers is always welcome. Let us know what you think about this book-what you liked or disliked. Reader feedback is important for us as it helps us develop titles that you will really get the most out of. To send us general feedback, simply email [email protected], and mention the book's title in the subject of your message. If there is a topic that you have expertise in and you are interested in either writing or contributing to a book, see our author guide at www.packtpub.com/authors.

Customer support

Now that you are the proud owner of a Packt book, we have a number of things to help you to get the most from your purchase.

Downloading the example code

You can download the example code files for this book from your account at http://www.packtpub.com. If you purchased this book elsewhere, you can visit http://www.packtpub.com/support and register to have the files emailed directly to you. You can download the code files by following these steps:

Log in or register to our website using your email address and password.

Hover the mouse pointer on the SUPPORT tab at the top.

Click on Code Downloads & Errata.

Enter the name of the book in the Search box.

Select the book for which you're looking to download the code files.

Choose from the drop-down menu where you purchased this book from.

Click on Code Download.

Once the file is downloaded, please make sure that you unzip or extract the folder using the latest version of:

WinRAR / 7-Zip for Windows

Zipeg / iZip / UnRarX for Mac

7-Zip / PeaZip for Linux

The code bundle for the book is also hosted on GitHub at https://github.com/PacktPublishing/Statistics-for-Data-Science. We also have other code bundles from our rich catalogue of books and videos available at https://github.com/PacktPublishing/. Check them out!

Downloading the color images of this book

We also provide you with a PDF file that has color images of the screenshots/diagrams used in this book. The color images will help you better understand the changes in the output. You can download this file from https://www.packtpub.com/sites/default/files/downloads/StatisticsforDataScience_ColorImages.pdf.

Errata

Although we have taken every care to ensure the accuracy of our content, mistakes do happen. If you find a mistake in one of our books-maybe a mistake in the text or the code-we would be grateful if you could report this to us. By doing so, you can save other readers from frustration and help us improve subsequent versions of this book. If you find any errata, please report them by visiting http://www.packtpub.com/submit-errata, selecting your book, clicking on the Errata Submission Form link, and entering the details of your errata. Once your errata are verified, your submission will be accepted and the errata will be uploaded to our website or added to any list of existing errata under the Errata section of that title. To view the previously submitted errata, go to https://www.packtpub.com/books/content/support and enter the name of the book in the search field. The required information will appear under the Errata section.

Piracy

Piracy of copyrighted material on the internet is an ongoing problem across all media. At Packt, we take the protection of our copyright and licenses very seriously. If you come across any illegal copies of our works in any form on the internet, please provide us with the location address or website name immediately so that we can pursue a remedy. Please