Machine Learning for Practical Decision Making: A Multidisciplinary Perspective with Applications from Healthcare, Engineering and Business Analytics Christo El Morr pdf download
Machine Learning for Practical Decision Making: A Multidisciplinary Perspective with Applications from Healthcare, Engineering and Business Analytics Christo El Morr pdf download
https://ebookmeta.com/product/machine-learning-for-business-
analytics-concepts-techniques-and-applications-with-jmp-pro-2nd-
edition-galit-shmueli/
https://ebookmeta.com/product/machine-learning-for-business-
analytics-concepts-techniques-and-applications-with-jmp-pro-2nd-
edition-galit-shmueli-2/
https://ebookmeta.com/product/business-analytics-a-data-driven-
decision-making-approach-for-business-volume-i-1st-edition-amar-
sahay/
https://ebookmeta.com/product/fashion-and-its-multi-cultural-
facets-1st-edition-patricia-hunt-hurst/
Ergodic Theoretic Methods in Group Homology A
Minicourse on L2 Betti Numbers in Group Theory Clara
Löh
https://ebookmeta.com/product/ergodic-theoretic-methods-in-group-
homology-a-minicourse-on-l2-betti-numbers-in-group-theory-clara-
loh/
https://ebookmeta.com/product/no-mercy-valerie-law-fbi-suspense-
thriller-1-1st-edition-blake-pierce/
https://ebookmeta.com/product/the-definitive-guide-to-hotwire-
and-django-1st-edition-michael-yin/
https://ebookmeta.com/product/film-and-the-anarchist-
imagination-2nd-edition-richard-porton/
https://ebookmeta.com/product/the-vanquishers-the-
vanquishers-1-1st-edition-kalynn-bayron/
Deep Learning Approaches to Cloud Security Deep
Learning Approaches for Cloud Security 1st Edition
Pramod Singh Rathore Editor Vishal Dutt Editor Rashmi
Agrawal Editor Satya Murthy Sasubilli Editor Srinivasa
Rao Swarna Editor
https://ebookmeta.com/product/deep-learning-approaches-to-cloud-
security-deep-learning-approaches-for-cloud-security-1st-edition-
pramod-singh-rathore-editor-vishal-dutt-editor-rashmi-agrawal-
editor-satya-murthy-sasubilli-editor-sr/
International Series in
Operations Research & Management Science
Christo El Morr
Manar Jammal
Hossam Ali-Hassan
Walid EI-Hallak
Machine Learning
for Practical
Decision Making
A Multidisciplinary Perspective
with Applications from Healthcare,
Engineering and Business Analytics
International Series in Operations Research &
Management Science
Founding Editor
Frederick S. Hillier, Stanford University, Stanford, CA, USA
Volume 334
Series Editor
Camille C. Price, Department of Computer Science, Stephen F. Austin State Uni-
versity, Nacogdoches, TX, USA
Associate Editor
Joe Zhu, Foisie Business School, Worcester Polytechnic Institute, Worcester, MA, USA
The book series International Series in Operations Research and Management
Science encompasses the various areas of operations research and management
science. Both theoretical and applied books are included. It describes current
advances anywhere in the world that are at the cutting edge of the field. The series
is aimed especially at researchers, advanced graduate students, and sophisticated
practitioners.
The series features three types of books:
• Advanced expository books that extend and unify our understanding of partic-
ular areas.
• Research monographs that make substantial contributions to knowledge.
• Handbooks that define the new state of the art in particular areas. Each
handbook will be edited by a leading authority in the area who will organize a
team of experts on various aspects of the topic to write individual chapters. A
handbook may emphasize expository surveys or completely new advances (either
research or applications) or a combination of both.
The series emphasizes the following four areas:
Mathematical Programming: Including linear programming, integer program-
ming, nonlinear programming, interior point methods, game theory, network opti-
mization models, combinatorics, equilibrium programming, complementarity
theory, multiobjective optimization, dynamic programming, stochastic program-
ming, complexity theory, etc.
Applied Probability: Including queuing theory, simulation, renewal theory,
Brownian motion and diffusion processes, decision analysis, Markov decision
processes, reliability theory, forecasting, other stochastic processes motivated by
applications, etc.
Production and Operations Management: Including inventory theory, produc-
tion scheduling, capacity planning, facility location, supply chain management,
distribution systems, materials requirements planning, just-in-time systems, flexible
manufacturing systems, design of production lines, logistical planning, strategic
issues, etc.
Applications of Operations Research and Management Science: Including
telecommunications, health care, capital budgeting and finance, economics, market-
ing, public policy, military operations research, humanitarian relief and disaster
mitigation, service operations, transportation systems, etc.
This book series is indexed in Scopus.
Christo El Morr • Manar Jammal •
Hossam Ali-Hassan • Walid El-Hallak
Machine Learning
for Practical Decision Making
A Multidisciplinary Perspective with
Applications from Healthcare, Engineering
and Business Analytics
Christo El Morr Manar Jammal
School of Health Policy and Management School of Information Technology
York University York University
Toronto, ON, Canada Toronto, ON, Canada
© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland
AG 2022
This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether
the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of
illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and
transmission or information storage and retrieval, electronic adaptation, computer software, or by
similar or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or
the editors give a warranty, expressed or implied, with respect to the material contained herein or for any
errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
To our families for their love and support. To
our students, they are our inspiration.
Preface
This book fills a gap in the machine learning literature. It presents the machine
learning concepts in a very simple and practical way starting with a tangible example
and building on it the theory. Each machine learning chapter starts with a presenta-
tion of the problem to solve, followed by a practical example on how the algorithm
that solves the problem works, then comes the presentation of the machine learning
algorithm theory and closing remarks (e.g., pitfalls, advantages limitations). This is
followed by a set of key terms, a set of questions to test your understanding of the
chapter material, a set of references to read more about the subject and hands-on lab
exercises. The lab exercises allow readers to apply the chapter’s concepts and gain
needed skills. To maximize the benefit for readers and to expose them to a myriad of
machine learning languages and frameworks, the lab exercises (and sometimes the
chapter’s material) are built around Python mainly, followed by R (and R Studio),
and Weka. One of the strengths of this book is that it can be used by people who are
exposed to programming or would like to learn how to program and those who prefer
not to program but to solve decision-making problems with machine learning using
simple graphical user interface. The former can use Python, the machine learning
language par excellence, throughout all chapters, or R (limited to Chaps. 1–4, and 6).
The use of Tableau, a visual analytics platform, is reserved to Chap. 5, Data
Visualization, while Weka is explained and used in Chaps. 4 and 6–12; given the
simplicity of Weka, we believe that there was no necessity to add Weka-based lab
exercises for Chaps. 13–15.
This introductory textbook to machine learning for decision making can be used
by students in Computer Science, Information Technology, Health Informatics, and
Business fields. Depending on the students’ level of study and exposure to technol-
ogy, either Weka or Python can be used. However, given the pervasive use of Python
in the market we advise students of all sorts to get exposure to Python and how it
works. Teaching the whole Python language is beyond the scope of this book;
however, we cover Python’s libraries related to machine leaning (e.g., Scikit-
Learn, TensorFlow, Keras) and many Python programming concepts.
vii
viii Preface
Another strength of the book is its focus on the necessary content for an
introductory course to machine learning while providing enough complexity without
being complex or introducing heavy mathematical formulation; the exception being
neural networks where we considered that providing less simple mathematical
formulations was necessary as an illustration but skipping them would not be a
problem to understand the algorithm. There is no mathematical knowledge needed to
read and use this book. Chapter 2 provides, in a simple manner coupled with many
examples, the main mathematical concepts needed to understand the chapters.
A final strength of this book is the use of a variety of datasets from several
domains (e.g., health, business, social media, census, survey) which provide a good
exposure to the myriad types of applications in which machine learning can be used.
The book is organized in three parts: Part 1 is an introduction section that
encompasses Chaps. 1–5, it introduces machine learning fundamentals and allows
for installing the different software tools and the introduction to Python, R, and
Weka. The machine learning algorithms and corresponding lab exercises are covered
in Chaps. 6–15, and future perspectives are provided in Chap. 16.
Professors who adopt this book have flexibility in the way they want to teach the
material; it all depends on the objectives of their course. Some can use to teach
machine learning using Python and hence need to cover Chaps. 1, 4 and 6–16; this
could be true for courses related to practical machine learning. Professors who are
interested in teaching Analytics and (some) machine learning can cover Chaps. 1–5
and then some of the Chaps. 6–16. It is still feasible to cover all chapters in 12 weeks,
Chaps. 1 and 2 are an introduction and a quick overview that can be covered with
their labs in one session; Chaps. 3 (or some of it) and Chap. 4 are possible to
combine in one session. Chapters 5 and 16 can be covered in one session; then each
Chaps. 6–14 in one session; and final Chaps. 15 and 16 in one session. This is not to
impose a single way of approaching the textbook but to provide examples of
alternatives and demonstrate flexibility. The book provides you with flexibility to
be adopted in several contexts. The datasets used in the lab exercise touch upon
many domains: health, information technology, business, and engineering.
We hope that this first edition of the book will enrich the readers’ knowledge and
skills and we welcome their comments and suggestions. Readers can access the
Python code for the Chap. 6 to Chap. 15 lab exercises on GitHub https://github.com/
christoelmorr/ML-4-Practical-DM.git.
ix
x Contents
2.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
2.6 Key Terms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
2.7 Test Your Understanding . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
2.8 Read More . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
2.9 Lab . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
2.9.1 Working Example in R . . . . . . . . . . . . . . . . . . . . . . 55
2.9.2 Working Example in Python . . . . . . . . . . . . . . . . . . 57
2.9.3 Do It Yourself . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
2.9.4 Do More Yourself (Links to Available Datasets for
Use) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
3 Overview of Machine Learning Algorithms . . . . . . . . . . . . . . . . . . . 61
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
3.2 Data Mining . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
3.3 Analytics and Machine Learning . . . . . . . . . . . . . . . . . . . . . . 63
3.3.1 Terminology Used in Machine Learning . . . . . . . . . 64
3.3.2 Machine Learning Algorithms: A Classification . . . . 65
3.4 Supervised Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
3.4.1 Multivariate Regression . . . . . . . . . . . . . . . . . . . . . 68
3.4.2 Decision Trees . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
3.4.3 Artificial Neural Networks . . . . . . . . . . . . . . . . . . . 72
3.4.4 Naïve Bayes Classifier . . . . . . . . . . . . . . . . . . . . . . 75
3.4.5 Random Forest . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
3.4.6 Support Vector Machines (SVM) . . . . . . . . . . . . . . . 78
3.5 Unsupervised Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
3.5.1 K-Means . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
3.5.2 K-Nearest Neighbors (KNN) . . . . . . . . . . . . . . . . . . 82
3.5.3 AdaBoost . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
3.6 Applications of Machine Learning . . . . . . . . . . . . . . . . . . . . . 85
3.6.1 Machine Learning Demand Forecasting and Supply
Chain Performance . . . . . . . . . . . . . . . . . . . . . . . . . 85
3.6.2 A Case Study on Cervical Pain Assessment with
Motion Capture . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
3.6.3 Predicting Bank Insolvencies Using Machine
Learning Techniques . . . . . . . . . . . . . . . . . . . . . . . . 89
3.6.4 Deep Learning with Convolutional Neural Network
for Objective Skill Evaluation in Robot-Assisted
Surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
3.7 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
3.8 Key Terms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
3.9 Test Your Understanding . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
3.10 Read More . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
3.11 Lab . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
3.11.1 Machine Learning Overview in R . . . . . . . . . . . . . . 94
Contents xi
"Ei äitiä! An… anteeksi, pyydän anteeksi, mutta .. mutta eikö lady
Alfreton ole äitinne?"
"Oh, niin, minä ymmärrän. Mutta et sinä heitä löydä. Ihmiset eivät
koskaan löydä toisiaan tällaisissa paikoissa. Parasta, että jäät minun
seuraani — jäisit ainakin siksi, kunnes kohtaat ne toiset…"
"Neiti Exeter tuo heidät tänne tänään, ja minä lupasin pitää varani
ollakseni valmis heitä puhuttelemaan ja neuvomaan heille, mitä
katsella ja minne mennä…"
Mennä asumaan tädin luo? Kummallista kyllä, tämä ajatus tuli nyt
vasta ensi kertaa Elman mieleen. Tavuakaan siitä eivät hänen
kuultensa olleet hiiskuneet hänen sukulaisensa enempää kuin neiti
Exeterkään puolestaan, ja se tunne, minkä tämä hänen
ajattelemattoman toverinsa keveä vihjaus hänen rinnassaan synnytti,
oli niin valtava, että se hetkiseksi karkoitti kaikki muut ajatukset.
"Vielä mitä! Luuletko niin? Ajattelitko niin heti vai nyt vasta?"
"Ei. Neiti Exeter sen tilasi; enkä luule… olen melkein varma, ettei
täti siitä oikein pitänyt. Minä näin hänen vilkaisevan minun puvustani
sinun pukuusi", ja Elma-parka nielaisi jotakin kurkustaan. Hän oli
vihdoinkin tulkinnut oikein tätinsä kasvoilla väikkyneen epäilyttävän
ilmeen tässä suhteessa.
Mitäpä lady Mabel oli Elmalle, tai Elma hänelle? Mitäpä merkitsi,
mitä tuo lavertelija ajatteli tai sanoi? Mitäpä haittasi, että
herttuattaren tyttären pitsipuku nyt uusien tulokkaiden
jakapäiväisemmistä aineksista tehtyjen ja karkeavärisempien
pukimien rinnalla muuttui välkkyväksi suviseitiksi? Elma ei välittäisi,
hän kieltäysi säikähtymästä ja masensi sydämessään heräävän
paholaisen — mutta muutamien ohimenevien, nopeasti haihtuvien
tuokioiden kuluessa hän oli kokenut jotakin, mitä ei koskaan ollut
unohtava. Aina elämänsä loppuun asti hän muistikin tuon
pysähdyksen siellä puiden siimeksessä ja oli tuntevinaan ison,
suloisen kukkasen tuoksun vieressään sellaisena kuin tuulahdus
lennätti sen hänen nenäänsä hänen kohottaessaan ylpeän nuoren
päänsä taistellakseen ystäviensä puolesta. Jos ken muu tahansa
heidät pettikin, hän ei sitä tekisi. Jos joku häpesi heitä tai piti
hauskaa heidän kustannukseltaan, hän ei tekisi kumpaakaan. Vaikka
Mabel ehkä nauraisi, — mutta Mabel ei nauranut: saattoihan olla
mahdollista, että hänkin ajatteli hänen tovereistaan parempaa kuin
ensimäisen vaikutelmansa vallassa ja oli taipuvainen ystävällisyyteen
ja suopeuteen. Olihan hän puhunut Elmalle siihen suuntaan, että
"tytöt" pääsisivät hänen tuttavuuteensa, ja siis…
Tosiasia oli, että lady Alfreton oli väleen väistänyt häntä; mutta
seuraavassa tuokiossa nähtiin Elman täti ja Mabelin äiti ryhmän
reunassa, ja kaikkien silmät kohtasivat toisensa erehtymättömästi.
"Ooh?" äännähti neiti Exeter käsittäen, että suuri hetki oli tullut.
"Minä kai menen nyt äitini luo", ilmoitti lady Mabel silmäisten
tyynesti ympärilleen. "Hän kutsui minua katseillaan, ja luullakseni
tätisi kutsui sinua", virkkoi hän Elmalle. "Varmaan tapaamme
toisemme vielä", jatkoi nuori aatelisneiti, keveästi puikahtaen
mustanappisen viitan ohi, "hyvästi nyt!" Sitten muutamaa askelta
kauempaa: "Elma, Elma!"
"Totta. Mene, rakas lapsi", pisti tässä eräs toinen ääni väliin.
"Mene, rakas Elma, nuori ystäväsi on oikeassa; eikä lady Alfreton" —
neiti Exeter koetti puhua huolettomasti — "eikä lady Alfreton
hyväksy, että liityt mihinkään muuhun seurueeseen kuin hänen
omaansa".
"Niin, näetkös, Elma, he… mutta olen pahoillani, etten ollut heille
kohtelias. Varmaan aioin olla kohtelias" — jokseenkin
hämmentyneenä — "mutta kun…"
Mutta vasta jonkun ajan kuluttua hän sai selville, mitä tuo "jokin"
oli.
"Mutta mitä?"
Elma epäröi. Sitten hän äkkiä puhkesi puhumaan: "Et saa luulla,
että se oli Selina-tädin syy; tätihän ei tiennyt, ei voinut otaksuakaan,
ei voinut vähääkään aavistaa, miltä se neiti Exeteristä tuntuisi. Neiti
Exeter pitää Selina-tätiä ja… ja teitä kaikkia niin suuressa arvossa; ja
tavatessaan tädin täällä minun kanssani neiti Exeter odotti, tarkoitan,
piti luultavana, että… että häntä edes puhuteltaisiin… ja… ja…"
"No?"