(eBook PDF) Handbook of Statistical Analysis and

Data Mining Applications 2nd Edition install

download

https://ebookluna.com/product/ebook-pdf-handbook-of-statistical-
analysis-and-data-mining-applications-2nd-edition/

Download more ebook from https://ebookluna.com

 Instant digital products (PDF, ePub, MOBI) ready for you
Download now and discover formats that fit your needs...

Handbook of statistical analysis and data mining

applications Second Edition Elder - eBook PDF

https://ebookluna.com/download/handbook-of-statistical-analysis-and-
data-mining-applications-ebook-pdf/

ebookluna.com

(eBook PDF) Data Mining and Predictive Analytics 2nd

Edition

https://ebookluna.com/product/ebook-pdf-data-mining-and-predictive-
analytics-2nd-edition/

ebookluna.com

Predictive Modeling in Biomedical Data Mining and Analysis

1st Edition- eBook PDF

https://ebookluna.com/download/predictive-modeling-in-biomedical-data-
mining-and-analysis-ebook-pdf/

ebookluna.com

(eBook PDF) Data Mining for Business Analytics: Concepts,

Techniques, and Applications in R

https://ebookluna.com/product/ebook-pdf-data-mining-for-business-
analytics-concepts-techniques-and-applications-in-r/

ebookluna.com
(eBook PDF) An Introduction to Statistical Methods & Data
Analysis 7th

https://ebookluna.com/product/ebook-pdf-an-introduction-to-
statistical-methods-data-analysis-7th/

ebookluna.com

(eBook PDF) Introduction to Data Mining, Global Edition

2nd Edition

https://ebookluna.com/product/ebook-pdf-introduction-to-data-mining-
global-edition-2nd-edition/

ebookluna.com

(eBook PDF) Data Mining for Business Analytics: Concepts,

Techniques, and Applications with JMP Pro

https://ebookluna.com/product/ebook-pdf-data-mining-for-business-
analytics-concepts-techniques-and-applications-with-jmp-pro/

ebookluna.com

Handbook of Thermal Analysis and Calorimetry: Recent

Advances, Techniques and Applications 2nd Edition Sergey
Vyazovkin - eBook PDF
https://ebookluna.com/download/handbook-of-thermal-analysis-and-
calorimetry-recent-advances-techniques-and-applications-ebook-pdf/

ebookluna.com

(eBook PDF) Data Mining Concepts and Techniques 3rd

https://ebookluna.com/product/ebook-pdf-data-mining-concepts-and-
techniques-3rd/

ebookluna.com
 vii
CONTENTS  

Application to Mixed Models 207 Postscript 288

Neural Nets for Prediction 208 References 288
Support Vector Machines (SVMS) and Other Kernel
Learning Algorithms 211 15. Fraud Detection
Postscript 212 Preamble 289
References 213 Issues With Fraud Detection 289
How Do You Detect Fraud? 292
11. Model Evaluation and Enhancement Supervised Classification of Fraud 293
Preamble 215 How Do You Model Fraud? 294
Evaluation and Enhancement: Part of the Modeling How Are Fraud Detection Systems Built? 295
Process 215 Intrusion Detection Modeling 296
Types of Errors in Analytical Models 216 Comparison of Models With and Without
Model Enhancement Techniques 227 Time-Based Features 297
Model Enhancement Checklist 231 Building Profiles 301
Postscript 232 Deployment of Fraud Profiles 302
References 232 Postscript 302
References 302
12. Predictive Analytics for Population
Health and Care
Preamble 235 III
The Future of Healthcare, and How Predictive TUTORIALS AND CASE
Analytics Fits 235
Predictive Analytics and Population Health 246 STUDIES
Predictive Analytics and Precision
Medicine 253 Tutorial A Example of Data Mining
Postscript 257 Recipes Using Windows 10 and
References 257 Statistica 13
Further Reading 258 LINDA A. MINER

13. Big Data in Education: New

Efficiencies for Recruitment, Learning, and Tutorial B Using the Statistica Data
Retention of Students and Donors Mining Workspace Method for Analysis of
ANDY PETERSON
Hurricane Data (Hurrdata.sta)
JEFF WONG
Preamble 259
Introduction 259
Industrial Integration of Educational Psychology and
Tutorial C Case Study—Using SPSS
Big Data Analytics 274 Modeler and STATISTICA to Predict
Postscript 275 Student Success at High-Stakes Nursing
References 276 Examinations (NCLEX)
Further Reading 277 GALINA BELOKUROVA, CHIARINA PIAZZA

14. Customer Response Modeling Introduction 335

Preamble 279 Decision Management in Nursing Education 336
Early CRM Issues in Business 279 Case Study 337
Knowing How Customers Behaved Before They Research Question 337
Acted 280 Literature Review 337
CRM in Business Ecosystems 281 Dataset and Expected Strength of Predictors 338
Conclusions 287 Data Mining With SPSS Modeler 339
viii CONTENTS

Data Mining With STATISTICA 348 Tutorial I Data Prep 1–2: Data
Conclusion 355 Description
References 356
ROBERTA BORTOLOTTI, MSIS, CBAP
Further Reading 357

Tutorial D Constructing a Histogram Tutorial J Data Prep 2-1: Data

in KNIME Using MidWest Company Cleaning and Recoding
Personality Data ROBERTA BORTOLOTTI
LINDA A. MINER

Tutorial K Data Prep 2-2: Dummy

Tutorial E Feature Selection in KNIME Coding Category Variables
BOB NISBET
ROBERTA BORTOLOTTI

Why Select Features? 377

Occam’s Razor—Simple, But Not
Simplistic 377 Tutorial L Data Prep 2-3: Outlier
Local Minimum Error 378 Handling
Moving Out of the Local Minimum 379 ROBERTA BORTOLOTTI
Strategies for Reduction of Dimensionality
in Predictive Analytics Available in
KNIME 379
Tutorial M Data Prep 3-1: Filling
Missing Values With Constants
Tutorial F Medical/Business Tutorial ROBERTA BORTOLOTTI
LINDA A. MINER

Tutorial N Data Prep 3-2: Filling Missing

Tutorial G A KNIME Exercise, Values With Formulas
Using Alzheimer’s Training Data of ROBERTA BORTOLOTTI
Tutorial F
LINDA A. MINER

Introduction 423 Tutorial O Data Prep 3-3: Filling

KNIME Project 423 Missing Values With a Model
Getting the Program to Open Microsoft ROBERTA BORTOLOTTI
Excel CSV File: Alzheimer Training
Data 426
Decision Trees Node 428 Tutorial P City of Chicago Crime
Linear Correlation Node 432
Map: A Case Study Predicting Certain
Conditional Box Plot Node 436
Decision Trees Again 438 Kinds of Crime Using Statistica
End Note 442 Data Miner and Text Miner
ENDRIN TUSHE

Tutorial H Data Prep 1-1: Merging Data Analysis 599

Data Sources Text Mining 606
ROBERTA BORTOLOTTI, MSIS, CBAP Boosted Trees 614
 ix
CONTENTS  

Tutorial Q Using Customer Churn Data Generalized Degrees of Freedom 711

to Develop and Select a Best Predictive Examples: Decision Tree Surface With Noise 712
Summary and Discussion 715
Model for Client Defection Using Postscript 716
STATISTICA Data Miner 13 64-bit for Acknowledgment 717
Windows 10 References 717
RICHARD PORTER WITH ASSISTANCE OF ROBERT NISBET, Further Reading 718
LINDA A. MINER, GARY MINER
17. The “Right Model” for the
About This Tutorial 627
Business Objectives 627
“Right Purpose”: When Less Is Good
Data Preparation 630 Enough
Feature Selection 642 Preamble 719
Building a Predictive Model With STATISTICA More Is Not Necessarily Better: Lessons From Nature
Data Miner DMRecipes 646 and Engineering 720
Model Evaluation 648 Postscript 726
References 726
Tutorial R Example With C&RT to
Predict and Display Possible Structural 18. A Data Preparation Cookbook
Preamble 727
Relationships Introduction 727
GREG ROBINSON, LINDA A. MINER, MARY A. MILLIKIN CRISP-DM—Business Understanding Phase 728
CRISP-DM—Data Understanding Phase 729
References 674 CRISP-DM—Data Preparation Phase 732
CRISP-DM—Modeling Phase 736
Tutorial S Clinical Psychology: Making 18 Common Mistakes in Data Preparation in
Decisions About Best Therapy for a Client Predictive Analytics Projects 736
LINDA A. MINER Postscript 739
References 740

19. Deep Learning

IV Preamble 741
The Guiding Concept of DL Technology—Human
MODEL ENSEMBLES, MODEL Cognition 742
COMPLEXITY; USING THE Early Artificial Neural Networks (ANNs) 743
RIGHT MODEL FOR THE How ANNs Work 745
More Elaborate Architectures—DL Neural
RIGHT USE, SIGNIFICANCE, Networks 746
ETHICS, AND THE FUTURE, Postscript 750
References 751
AND ADVANCED PROCESSES Further Reading 751
16. The Apparent Paradox of Complexity 20. Significance versus Luck in the
in Ensemble Modeling Age of Mining: The Issues of P-Value
JOHN ELDER, ANDY PETERSON
“Significance” and “Ways to Test
Preamble 705 Significance of Our Predictive Analytic
Introduction 706 Models”
Model Ensembles 706 Preamble 753
How Measure Model Complexity? 709 Introduction 753
x CONTENTS

The Problem of Significance in Traditional P-Value Ethics and Data Science for the Virtues of Personal
Statistical Analysis 754 Life (Existential-Motivational) 769
USUAL Data Mining/Predictive Analytic Combination: Right Standards, Right Goals,
Performance Measures—Terminology 759 and Personal Virtue (Normative, Situational,
Unique Ways to Test Accuracy (“Significance”) Existential) 770
of Machine Learning Predictive Models 760 Michael Sandel on “Doing The Right Thing” With
Compare Predictive Model Performance Against Data Analytics 770
Random Results With Lift Charts and Decile Discovering Data Ethics in an “Alignment
Tables 760 Methodology” 771
Evaluate the Validity of Your Discovery With Target References 772
Shuffling 762 Further Reading 772
Test Predictive Model Consistency With Bootstrap
Sampling 763 22. IBM Watson
Postscript 764 Preamble 773
References 765 Introduction 773
What Exactly Is Watson? 773
21. Ethics and Data Analytics Jeopardy! 774
Internal Features of Watson 774
ANDY PETERSON
Application Programming Interfaces (APIs) 776
Preamble 767 Software Development Kits (SDKs) 778
The Birthday Party—A Practical Example for Ethical Some Existing Applications of Watson
Action 767 Techology 778
Academic Secular Ethics 768 Ushering in the Cognitive Era 780
Ethics and Data Science for the Norms of Postscript 780
Government (Deontological-Normative) 769 Reference 781
Ethics and Data Science for the Goals in Business
(Situational-Teleological) 769 Index 783
 List of Tutorials on the Elsevier
Companion Web Page

Note: This list includes all the extra tuto- 8. TUTORIAL “W”—Diabetes Control in
rials published with the 1st edition of this Patients [Field: Medical Informatics]
handbook (2009). These can be considered 9. TUTORIAL “X”—Independent
“enrichment” tutorials for readers of this 2nd Component Analysis [Field: Separating
edition. Since the 1st edition of the handbook Competing Signals]
will not be available after the release of the 10. TUTORIAL “Y”—NTSB Aircraft
2nd edition, these extra tutorials are carried Accidents Reports [Field: Engineering—
over in their original format/versions of soft- Air Travel—Text Mining]
ware, as they are still very useful in learning 11. TUTORIAL “Z”—Obesity Control in
and understanding data mining and predic- Children [Field: Preventive Health Care]
tive analytics, and many readers will want to 12. TUTORIAL “AA”—Random Forests
take advantage of them. Example [Field: Statistics—Data Mining]
List of Extra Enrichment Tutorials that 13. TUTORIAL “BB”—Response
are only on the ELSEVIER COMPANION Optimization [Field: Data Mining—
web page, with data sets as appropriate, for Response Optimization]
downloading and use by readers of this 2nd 14. TUTORIAL “CC”—Diagnostic Tooling
edition of handbook: and Data Mining: Semiconductor Industry
[Field: Industry—Quality Control]
1. TUTORIAL “O”—Boston Housing
15. TUTORIAL “DD”—Titanic—Survivors
Using Regression Trees [Field:
of Ship Sinking [Field: Sociology]
Demographics]
16. TUTORIAL “EE”—Census Data
2. TUTORIAL “P”—Cancer Gene [Field:
Analysis [Field: Demography—Census]
Medical Informatics & Bioinformatics]
17. TUTORIAL “FF”—Linear & Logistic
3. TUTORIAL “Q”—Clustering of Shoppers
Regression—Ozone Data [Field:
[Field: CRM—Clustering Techniques]
Environment]
4. TUTORIAL “R”—Credit Risk
18. TUTORIAL “GG”—R-Language
using Discriminant Analysis [Field:
Integration—DISEASE SURVIVAL
Financial—Banking]
ANALYSIS Case Study [Field: Survival
5. TUTORIAL “S”—Data Preparation and
Analysis—Medical Informatics]
Transformation [Field: Data Analysis]
19. TUTORIAL “HH”—Social Networks
6. TUTORIAL “T”—Model Deployment
Among Community Organizations
on New Data [Field: Deployment of
[Field: Social Networks—Sociology &
Predictive Models]
Medical Informatics]
7. TUTORIAL “V”—Heart Disease Visual
20. TUTORIAL “II”—Nairobi, Kenya
Data Mining Methods [Field: Medical
Baboon Project: Social Networking
Informatics]

xi
xii LIST OF TUTORIALS ON THE ELSEVIER COMPANION WEB PAGE

Among Baboon Populations in Kenya 22. TUTORIAL “KK”—Dahlia Mosaic

on the Laikipia Plateau [Field: Social Virus: A DNA Microarray Analysis of 10
Networks] Cultivars from a Single Source: Dahlia
21. TUTORIAL “JJ”—Jackknife and Garden in Prague, Czech Republic
Bootstrap Data Miner Workspace and [Field: Bioinformatics]
MACRO [Field: Statistics Resampling
Methods]

The final companion site URL will be https://www.elsevier.com/books-and-journals/

book-companion/9780124166325.
 Foreword 1 for 1st Edition

This book will help the novice user be- • Asking the wrong question—when
come familiar with data mining. Basically, looking for a rare phenomenon, it may
data mining is doing data analysis (or statis- be helpful to identify the most common
tics) on data sets (often large) that have been pattern. These may lead to complex
obtained from potentially many sources. As analyses, as in item 3, but they may also
such, the miner may not have control of the be conceptually simple. Again, you may
input data, but must rely on sources that have need to take care that you don't overfit
gathered the data. As such, there are prob- the data.
lems that every data miner must be aware of • Don't become enamored with the data.
as he or she begins (or completes) a mining There may be a substantial history from
operation. I strongly resonated to the mate- earlier data or from domain experts that
rial on “The Top 10 Data Mining Mistakes,” can help with the modeling.
which give a worthwhile checklist: • Be wary of using an outcome variable (or
• Ensure you have a response variable and one highly correlated with the outcome
predictor variables—and that they are variable) and becoming excited about the
correctly measured. result. The predictors should be “proper”
• Beware of overfitting. With scads of predictors in the sense that they (a) are
variables, it is easy with most statistical measured prior to the outcome and (b)
programs to fit incredibly complex are not a function of the outcome.
models, but they cannot be reproduced. It • Do not discard outliers without solid
is good to save part of the sample to use justification. Just because an observation
to test the model. Various methods are is out of line with others is insufficient
offered in this book. reason to ignore it. You must check the
• Don't use only one method. Using only circumstances that led to the value. In
linear regression can be a problem. any event, it is useful to conduct the
Try dichotomizing the response or analysis with the observation(s) included
categorizing it to remove nonlinearities and excluded to determine the sensitivity
in the response variable. Often, there are of the results to the outlier.
clusters of values at zero, which messes • Extrapolating is a fine way to go
up any normality assumption. This, of broke; the best example is the stock
course, loses information, so you may market. Stick within your data, and
want to categorize a continuous response if you must go outside, put plenty
variable and use an alternative to of caveats. Better still, restrain the
regression. Similarly, predictor variables impulse to extrapolate. Beware that
may need to be treated as factors rather pictures are often far too simple and
than linear predictors. A classic example we can be misled. Political campaigns
is using marital status or race as a linear oversimplify complex problems (“my
predictor when there is no order. opponent wants to raise taxes”; “my

xiii
xiv FOREWORD 1 FOR 1st EDITION

opponent will take us to war”) when using mean replacement, almost the
the realities may imply we have same set of predictor variables surfaced,
some infrastructure needs that can be but the residual sum of squares was 20.
handled only with new funding or we I then used multiple imputation and
have been attacked by some bad guys. found approximately the same set of
Be wary of your data sources. If you are predictors but had a residual sum of
combining several sets of data, they need squares (median of 20 imputations) of
to meet a few standards: 25. I find that mean replacement is rather
• The definitions of variables that are optimistic but surely better than relying
being merged should be identical. Often, on only complete cases. Using stepwise
they are close but not exact (especially regression, I find it useful to replicate
in metaanalysis where clinical studies it with a bootstrap or with multiple
may have somewhat different definitions imputations. However, with large data
due to different medical institutions or sets, this approach may be expensive
laboratories). computationally.
• Be careful about missing values. Often, To conclude, there is a wealth of material
when multiple data sets are merged, in this handbook that will repay study.
missing values can be induced: one
variable isn't present in another data set; Peter A. Lachenbruch
what you thought was a unique variable Oregon State University, Corvallis, OR,
name was slightly different in the two United States
sets, so you end up with two variables American Statistical Association,
that both have a lot of missing values. Alexandria, VA, United States
• How you handle missing values can be Johns Hopkins University, Baltimore,
crucial. In one example, I used complete MD, United States
cases and lost half of my sample; all UCLA, Los Angeles, CA, United States
variables had at least 85% completeness, University of Iowa, Iowa City, IA,
but when put together, the sample lost United States
half of the data. The residual sum of University of North Carolina, Chapel
squares from a stepwise regression was Hill, NC, United States
about 8. When I included more variables
 Foreword 2 for 1st Edition

A November 2008 search on https:// This book is different; it is organized

www.amazon.com/ for “data mining” with the practitioner in mind. The volume
books yielded over 15,000 hits—including 72 is divided into four parts. Part I provides
to be published in 2009. Most of these books an overview of analytics from a historical
either describe data mining in very technical perspective and frameworks from which to
and mathematical terms, beyond the reach approach data mining, including CRISP-DM
of most individuals, or approach data min- and SEMMA. These chapters will provide a
ing at an introductory level without suffi- novice analyst an excellent overview by de-
cient detail to be useful to the practitioner. fining terms and methods to use and will
The Handbook of Statistical Analysis and Data provide program managers a framework
Mining Applications is the book that strikes from which to approach a wide variety of
the right balance between these two treat- data mining problems. Part II describes algo-
ments of data mining. rithms, though without extensive mathemat-
This volume is not a theoretical treatment ics. These will appeal to practitioners who
of the subject—the authors themselves recom- are or will be involved with day-to-day an-
mend other books for this—but rather contains alytics and need to understand the qualita-
a description of data mining principles and tive aspects of the algorithms. The inclusion
techniques in a series of “knowledge-­transfer” of a chapter on text mining is particularly
sessions, where examples from real data mining timely, as text mining has shown tremen-
projects illustrate the main ideas. This aspect dous growth in recent years.
of the book makes it most valuable for practi- Part III provides a series of tutorials that
tioners, whether novice or more experienced. are both domain-specific and software-­
While it would be easier for everyone if specific. Any instructor knows that examples
data mining were merely a matter of find- make the abstract concept more concrete,
ing and applying the correct mathematical and these tutorials accomplish exactly that.
equation or approach for any given problem, In addition, each tutorial shows how the
the reality is that both “art” and “science” ­solutions were developed using popular data
are necessary. The “art” in data mining re- mining software tools, such as Clementine,
quires experience: when one has seen and Enterprise Miner, Weka, and STATISTICA.
overcome the difficulties in finding solutions The step-by-step specifics will assist practi-
from among the many possible approaches, tioners in learning not only how to approach
one can apply newfound wisdom to the next a wide variety of problems but also how
project. However, this process takes consid- to use these software products effectively.
erable time, and particularly for data min- Part IV presents a look at the future of data
ing novices, the iterative process inevitable mining, including a treatment of model
in data mining can lead to discouragement ensembles and “The Top 10 Data Mining
when a “textbook” approach doesn't yield a Mistakes,” from the popular presentation by
good solution. Dr. Elder.

xv
xvi FOREWORD 2 FOR 1st EDITION

However, the book is best read a few the ­excellent “History of Statistics and Data
chapters at a time while actively doing Mining” chapter and chapters 16, 17, and
the data mining rather than read cover to 18. These are broadly applicable and should
cover (a daunting task for a book this size). be read by even the most experienced data
Practitioners will appreciate tutorials that miners.
match their business objectives and choose The Handbook of Statistical Analysis and
to ignore other tutorials. They may choose Data Mining Applications is an exceptional
to read sections on a particular algorithm to book that should be on every data miner's
increase insight into that algorithm and then bookshelf or, better yet, found lying open
decide to add a second algorithm after the next to their computer.
first is mastered. For those new to a partic-
ular software tool highlighted in the tutori- Dean Abbott
als section, the step-by-step approach will Abbott Analytics, San Diego, CA,
operate much like a user's manual. Many United States
chapters stand well on their own, such as
 Preface

Much has happened in the professional • Operations research—decision science

discipline known previously as data mining and optimization of business processes
since the first edition of this book was written • Stock market quants—focused on
in 2008. This discipline has broadened and stock market trading and portfolio
deepened to a very large extent, requiring a optimization.
major reorganization of its elements. A new • Data engineering—focused on
parent discipline was formed, data science, optimizing data flow through memories
which includes previous subjects and activi- and storage structures
ties in data mining and many new elements • Business intelligence—focused primarily
of the scientific study of data, including stor- on descriptive aspects of data but
age structures optimized for analytic use, predictive aspects are coming
data ethics, and performance of many ac- • Business analytics—focused primarily
tivities in business, industry, and education. on the predictive aspects of data but is
Analytic aspects that used to be included in merging with descriptives
data mining have broadened considerably to (based on an article by Vincent

include image analysis, facial recognition, in- Granville published in http://www.
dustrial performance and control, threat de- datasciencecentral.com/profiles/
tection, fraud detection, astronomy, national blogs/17-analytic-disciplines-compared.)
security, weather forecasting, and financial In this book, we will use the terms “data
forensics. Consequently, several subdisci- mining” and “predictive analytics” synony-
plines have been erected to contain various mously, even though data mining includes
specialized data analytic applications. These many descriptive operations also.
subdisciplines of data science include the Modern data mining tools, like the ones
following: featured in this book, permit ordinary busi-
• Machine learning—analytic algorithm ness analysts to follow a path through the
design and optimization data mining process to create models that
• Data mining—generally restricted in are “good enough.” These less-than-optimal
scope now to pattern recognition apart models are far better in their ability to lever-
from causes and interpretation age faint patterns in databases to solve prob-
• Predictive analytics—using algorithms to lems than the ways it used to be done. These
predict things, rather than describe them tools provide default configurations and
or manage them automatic operations, which shield the user
• Statistical analysis—use of parametric from the technical complexity underneath.
statistical algorithms for analysis and They provide one part in the crude analogy
prediction to the automobile interface. You don't have
• Industrial statistical analysis—analytic to be a chemical engineer or physicist who
techniques to control and direct industrial understands moments of force to be able to
operations operate a car. All you have to do is learn to

xvii
xviii PREFACE

turn the key in the ignition, step on the gas is enough here to permit you to construct
and the brake at the right times, and turn the “smart enough” business operations with a
wheel to change direction in a safe manner, relatively small amount of the right informa-
and voilà, you are an expert user of the very tion. James Taylor developed this concept
complex technology under the hood. The for automating operational ­decision-making
other half of the story is the instruction man- in the area of enterprise decision man-
ual and the driver's education course that agement (Raden and Taylor, 2007). Taylor
help you to learn how to drive. recognized that companies need decision-
This book provides the instruction man- making systems that are automated enough
ual and a series of tutorials to train you how to keep up with the volume and time-­critical
to do data mining in many subject areas. We nature of modern business operations.
provide both the right tools and the right These decisions should be deliberate, pre-
intuitive explanations (rather than formal cise, and consistent across the enterprise;
mathematical definitions) of the data mining smart enough to serve immediate needs
process and algorithms, which will enable appropriately; and agile enough to adapt
even beginner data miners to understand the to new opportunities and challenges in the
basic concepts necessary to understand what company. The same concept can be applied
they are doing. In addition, we provide many to nonoperational systems for customer re-
tutorials in many different industries and lationship management (CRM) and market-
businesses (using many of the most common ing support. Even though a CRM model for
data mining tools) to show how to do it. cross sell may not be optimal, it may enable
several times the response rate in prod-
uct sales following a marketing campaign.
OVERALL ORGANIZATION Models like this are “smart enough” to drive
OF THIS BOOK companies to the next level of sales. When
models like this are proliferated through-
We have divided the chapters in this book out the enterprise to lift all sales to the next
into four parts to guide you through the as- level, more refined models can be developed
pects of predictive analytics. Part I covers the to do even better. This e­ nterprise-wide “lift”
history and process of predictive analytics. in intelligent operations can drive a com-
Part II discusses the algorithms and methods pany through evolutionary rather than rev-
used. Part III is a group of tutorials, which olutionary changes to reach long-term goals.
serve in principle as Rome served—as the Companies can leverage “smart enough”
central governing influence. Part IV presents decision systems to do likewise in their pur-
some advanced topics. The central theme of suit of optimal profitability in their business.
this book is the education and training of Clearly, the use of this book and these tools
beginning data mining practitioners, not the will not make you experts in data mining.
rigorous academic preparation of algorithm Nor will the explanations in the book per-
scientists. Hence, we located the tutorials in mit you to understand the complexity of the
the middle of the book in Part III, flanked by theory behind the algorithms and methodol-
topical chapters in Parts I, II, and IV. ogies so necessary for the academic student.
This approach is “a mile wide and an inch But we will conduct you through a relatively
deep” by design, but there is a lot packed into thin slice across the wide practice of data
that inch. There is enough here to stimulate mining in many industries and disciplines.
you to take deeper dives into theory, and there We can show you how to create powerful
 PREFACE xix
predictive models in your own organization Coauthor Gary Miner wishes to thank his
in a relatively short period of time. In addi- wife, Linda A. Winters-Miner, PhD, who has
tion, this book can function as a springboard been working with Gary on similar books over
to launch you into higher-level studies of the the past 30 years and wrote several of the tu-
theory behind the practice of data mining. torials included in this book, using ­real-world
If we can accomplish those goals, we will data. Gary also wishes to thank the following
have succeeded in taking a significant step in people from his office who helped in various
bringing the practice of data mining into the ways, including Angela Waner, Jon Hillis, Greg
mainstream of business analysis. Sergeant, and Dr. Thomas Hill, who gave per-
The three coauthors could not have done mission to use and also edited a group of the
this book completely by themselves, and tutorials that had been written over the years
we wish to thank the following individuals, by some of the people listed as guest authors in
with the disclaimer that we apologize if, by this book. Dr. Dave Dimas, of the University of
our neglect, we have left out of this “thank- California—Irvine, has also been very helpful
you list” anyone who contributed. in providing suggestions for enhancements for
Foremost, we would like to thank ac- this second edition—THANK YOU DAVE !!!
quisitions editor (name to use?) and others Without all the help of the people men-
(names). Bob Nisbet would like to honor tioned here and maybe many others we failed
and thank his wife, Jean Nisbet, PhD, who to specifically mention, this book would never
blasted him off in his technical career by re- have been completed. Thanks to you all!
typing his PhD dissertation five times (be-
fore word processing) and assumed much
of the family's burdens during the writing Bob Nisbet
of this book. Bob also thanks Dr. Daniel B. Gary Miner
Botkin, the famous global ecologist, for in- Ken Yale
troducing him to the world of modeling and
exposing him to the distinction between
viewing the world as machine and viewing Reference
it as ­organism. And thanks are due to Ken Raden, N., Taylor, J., 2007. Smart Enough Systems: How to
Reed, PhD, for inducting Bob into the prac- Deliver Competitive Advantage by Automating Hidden
Decisions. Prentice Hall, NJ, ISBN: 9780132713061.
tice of data mining.
 Introduction

Often, data analysts are asked, “What very different ways of arriving at the same
are statistical analysis and data mining?” In conclusion, a decision. We will introduce
this book, we will define what data mining some basic analytic history and theory in
is from a procedural standpoint. But most Chapters 1 and 2.
people have a hard time relating what we The basic process of analytic modeling is
tell them to the things they know and under- presented in Chapter 3. But it may be diffi-
stand. Before moving on into the book, we cult for you to relate what is happening in
would like to provide a little background for the process without some sort of tie to the
data mining that everyone can relate to. The real world that you know and enjoy. In many
Preface describes the many changes in ac- ways, the decisions served by analytic mod-
tivities related to data mining since the first eling are similar to those we make every day.
edition of this book was published in 2009. These decisions are based partly on patterns
Now, it is time to dig deeper and discuss the of action formed by experience and partly by
differences between statistical analysis and intuition.
data mining (aka predictive analytics).
Statistical analysis and data mining are PATTERNS OF ACTION
two methods for simulating the unconscious
operations that occur in the human brain to A pattern of action can be viewed in
provide a rationale for decision-making and terms of the activities of a hurdler on a
actions. Statistical analysis is a very directed race track. The runner must start success-
rationale that is based on norms. We all think fully and run to the first hurdle. He must
and make decisions on the basis of norms. decide very quickly how high to jump to
For example, we consider (unconsciously) clear the hurdle. He must decide when and
what the norm is for dress in a certain situa- in what sequence to move his legs to clear
tion. Also, we consider the acceptable range the hurdle with minimum effort and with-
of variation in dress styles in our culture. out knocking it down. Then, he must run
Based on these two concepts, the norm and a specified distance to the next hurdle and
the variation around that norm, we render do it all over again several times, until he
judgments like “that man is inappropriately crosses the finish line. Analytic modeling is
dressed.” Using similar concepts of mean a lot like that.
and standard deviation, statistical analy- The training of the hurdler's “model” of
sis proceeds in a very logical way to make action to run the race happens in a series of
very similar judgments (in principle). On operations:
the other hand, data mining learns case by
case and does not use means or standard • Run slow at first.
deviations. Data mining algorithms build • Practice takeoff from different positions
patterns, clarifying the pattern as each case to clear the hurdle.
is submitted for processing. These are two • Practice different ways to move the legs.

xxi
xxii INTRODUCTION

• Determine the best ways to do each activity. two classes (for dichotomous keys) and those
• Practice the best ways for each activity who don't. Along with this joke is a similar
over and over again. recognition from the outside that taxono-
mists are divided also into two classes: the
This practice trains the sensory and motor
“lumpers” (who combine several species into
neurons to function together most efficiently.
one) and the “splitters” (who divide one spe-
Individual neurons in the brain are “trained”
cies into many). These distinctions point to
in practice by adjusting signal strengths and
a larger dichotomy in the way people think.
firing thresholds of the motor nerve cells. The
In ecology, there used to be two schools
performance of a successful hurdler follows
of thought: autoecologists (chemistry, phys-
the “model” of these activities and the process
ics, and mathematics explain all) and the
of coordinating them to run the race. Creation
synecologists (organism relationships in
of an analytic “model” of a business process to
their environment explain all). It wasn't until
predict a desired outcome follows a very simi-
the 1970s that these two schools of thought
lar path to the training regimen of a hurdler. We
learned that both perspectives were needed
will explore this subject further in Chapter 3
to understand the complexities in ecosys-
and apply it to develop a data mining process
tems (but more about that later). In business,
that expresses the basic activities and tasks per-
there are the “big picture” people versus
formed in creating an analytic model.
“detail” people. Some people learn by fol-
lowing an intuitive pathway from general to
HUMAN INTUITION specific (deduction). Often, we call them “big
picture” people. Other people learn by fol-
In humans, the right side of the brain is lowing an intuitive pathway from specific to
the center for visual and esthetic sensibil- general (inductive). Often, we call them “de-
ities. The left side of the brain is the center tail” people. Similar distinctions are reflected
for quantitative and time-regulated sensi- in many aspects of our society. In Chapter 1,
bilities. Human intuition is a blend of both we will explore this distinction to a greater
sensibilities. This blend is facilitated by the depth in regards to the development of sta-
neural connections between the right side tistical and data mining theory through time.
of the brain and the left side. In women, the Many of our human activities involve
number of neural connections between the finding patterns in the data input to our sen-
right and left sides of the brain is 20% greater sory systems. An example is the mental pat-
(on average) than in men. This higher con- tern that we develop by sitting in a chair in
nectivity of women's brains enables them to the middle of a shopping mall and making
exercise intuitive thinking to a greater extent some judgment about patterns among its cli-
than men. Intuition “builds” a model of re- entele. In one mall, people of many ages and
ality from both quantitative building blocks races may intermingle. You might conclude
and visual sensibilities (and memories). from this pattern that this mall is located in
an ethnically diverse area. In another mall,
you might see a very different pattern. In
PUTTING IT ALL one mall in Toronto, a great many of the
TOGETHER stores had Chinese titles and script on the
windows. One observer noticed that he was
Biological taxonomy students claim (in the only non-Asian seen for a half hour. This
jest) that there are two kinds of people in led to the conclusion that the mall catered
­taxonomy—those who divide things up into to the Chinese community and was owned
 INTRODUCTION xxiii
(probably) by a Chinese company or person. of the model. We will discuss this approach in
Statistical methods employed in testing this more detail in Chapter 1. Data mining doesn't
“hypothesis” would include the following: start with a model; it builds a model with the
data. Thus, statistical analysis uses a model to
• Performing a survey of customers to gain
characterize a pattern in the data; data mining
empirical data on race, age, length of time
uses the pattern in the data to build a model.
in the United States, etc.
This approach uses deductive reasoning,
• Calculating means (averages) and
following an Aristotelian approach to truth.
standard deviations (an expression of the
From the “model” accepted in the beginning
average variability of all the customers
(based on the mathematical distributions as-
around the mean).
sumed), outcomes are deduced. On the other
• Using the mean and standard deviation
hand, data mining methods discover patterns
for all observations to calculate a metric
in data inductively, rather than deductively,
(e.g., student's t-value) to compare with
following a more Platonic approach to truth.
standard tables.
We will unpack this distinction to a much
• If the metric exceeds the standard table
greater extent in Chapter 1.
value, this attribute (e.g., race) is present
Which is the best way to do it? The an-
in the data at a higher rate than expected
swer is it depends. It depends on the data.
at random.
Some data sets can be analyzed better with
More advanced statistical techniques can statistical analysis techniques, and other data
accept data from multiple attributes and sets can be analyzed better with data mining
process them in combination to produce a techniques. How do you know which ap-
metric (e.g., average squared error), which proach to use for a given data set? Much ink
reflects how well a subset of attributes (se- has been devoted to paper to try to answer
lected by the processing method) predict that question. We will not add to that effort.
desired outcome. This process “builds” an Rather, we will provide a guide to general an-
analytic equation, using standard statistical alytic theory (Chapter 2) and broad analytic
methods. This analytic “model” is based on procedures (Chapter 3) that can be used with
averages across the range of variation of the techniques for either approach. For the sake
input attribute data. This approach to finding of simplicity, we will refer to the joint body of
the pattern in the data is basically a deduc- techniques as analytics. In Chapter 4, we in-
tive, top-down process (general to specific). troduce some of the many data preparation
The general part is the statistical model em- procedures for analytics.
ployed for the analysis (i.e., normal paramet- Chapter 5 presents various methods for
ric model). This approach to model building selecting candidate predictor variables to
is very “Platonic.” In Chapter 1, we will ex- be used in a statistical or machine-learning
plore the distinctions between Aristotelian model (differences between statistical and
and Platonic approaches for understanding machine-­learning methods of model build-
truth in the world around us. ing are discussed in Chapter 1). Chapter 6 in-
Part I—Introduction and overview of data troduces accessory tools and some advanced
mining processes. features of many data mining tools.
Both statistical analysis and data mining Part II—Basic and advanced algoithms,
algorithms operate on patterns: statistical and their application to common problems.
analysis uses a predefined pattern (i.e., the Chapters 7 and 8 discuss various basic
parametric model) and compares some mea- and advanced algorithms used in data min-
sure of the observations to standard metrics ing modeling applications. Chapters 9 and 10
xxiv INTRODUCTION

discuss the two general types of models, clas- significance, “luck” and ethics in data mining
sification and prediction. Chapter 11 presents applications. The book ends with Chapter 22,
some methods for evaluating and refining which gives an overview of the IBM Watson
analytic models. Chapters 12–15 describe technology, which IBM is trying to leverage
how data mining methods are applied to to solve many analytic problems. It is likely
four common applications. Part III contains that even these new processing strategies are
a group of tutorials that show how to apply not the end of the line in data mining devel-
various data mining tools to solve common opment. Chapter 1 ends with the statement
problems. Part IV discusses various issues of that we will discover increasingly novel and
model complexity, ethical use, and advanced clever ways to mimic the most powerful pat-
processes. Chapter 16 describes the para- tern recognition engine in the universe, the
dox of complexity. Chapter 17 introduces human brain.
the principle of “good-enough” models. One step further in the future could be to
Chapter 18 presents a list of data preparation drive the hardware supporting data mining
activities in the form of a cookbook, along to the level of portable devices like phones
with some caveats of using data mining (pre- and medical data loggers, even to smaller
dictive analytics) methods. Chapter 19 intro- applications in nanotechnology. In power-
duces one of the newest development areas, ful biological quantum computers, the size
deep learning. Some practitioners think that of pin heads (and smaller) may be the next
many data mining analyses will move in the wave of technological development to drive
direction of using deep learning algorithms. data mining advances. Rather than the sky,
Chapters 20 and 21 present various issues of the atom is the limit.

Bob Nisbet
September, 2017
Other documents randomly have
different content
was attached to the battalion commanded by Lieut.-Col. De
Salaberry, and on the latter retiring from the service, Captain
Thomson was promoted to the lieutenant-colonelcy, and given
command of the battalion. He was out during the Fenian troubles,
and subsequently retired retaining his rank. For a number of years
Colonel Thomson has been actively engaged, and still continues, in
the export trade of timber to Britain, etc. He took a prominent part
in the creation of the Union Bank of Canada, at Quebec, as well as in
the formation of several commercial companies connected with river
navigation. At present he is a director of the Union Bank of Canada;
vice-president of the Quebec Steamship Company; president of the
Saguenay & Lake St. John Railway Company; director of the St.
Lawrence Steamboat Company; and director of the Quebec Ferry
Company. A lover of sport Col. Thomson divides his summer vacation
between the secluded salmon pools of the Murray river, leased to
him, and his rustic cottage at Pointe-à-Pic, Murray Bay. Later on one
may meet him scouring for cariboo, with an Indian guide, the snow-
clad heights in rear of Baie St. Paul, known as Les Jardins. In politics
the colonel may be counted among the Liberal-Conservatives, and in
religion an adherent of the Presbyterian church. He is married to
Annie Atkinson, niece of the late Henry Atkinson, of Spencerwood.

Oulton, Alfred E., Dorchester, Judge of Probate for the

county of Westmoreland, New Brunswick, was born in
Westmoreland, on the 2nd March, 1845. His parents were Thomas E.
Oulton and Elizabeth Carter, both natives of Westmoreland county,
whose ancestors came from Yorkshire, England, and settled in
Westmoreland in 1763. Mr. Oulton received his education at the
schools of his native place, and after attending in his father’s store
for a while, went to Sackville, New Brunswick, and spent three years
in Mount Allison Academy, taking a course of studies which
embraced the higher mathematics and the Latin language. He
adopted law as a profession, and pursued his studies in the office of
A. L. Palmer, now judge in equity of New Brunswick. He was
admitted to practice as an attorney in June, 1867, and as a barrister
in June the following year. He then went into a law partnership with
Mr. Palmer for three years, when Mr. Palmer removed to St. John and
the partnership was dissolved. Since then he has carried on business
on his own account, and we may say here that he has been a very
successful lawyer. His practice extends into all the courts in the
province, and also into the Supreme courts of the Dominion, and he
does a great deal of office work, such as the collecting of claims,
conveyancing, and general notarial work. Mr. Oulton was elected
secretary of the municipal council of Dorchester, N.B., on its
organization on the 7th June, 1887, and still holds the office. On the
death of Governor Chandler, he was appointed judge of probate,
August 1, 1878, and is considered practical and painstaking in the
discharge of his official duties, and gives great satisfaction to the
public. He is a commissioner for the Admiralty Court of New
Brunswick. He joined the Masonic order in 1866, and was for three
years in succession master of the Blue lodge at Dorchester. He is
also a Royal Arch Mason, being a member of the chapter held at
Moncton. In religion he is a member of the Church of England; and
in politics a Conservative. He was married in June, 1883, to Kate
Estabrook, daughter of the late G. B. Estabrook, of Sackville, N.B.,
and they have a family of three children, two boys and one girl.

McIsaac, Colin F., Barrister, Antigonish, Nova Scotia,

M.P.P. for Antigonish, was born at South River, Antigonish county, in
1856. He is of Scotch descent, his ancestors having come from
Inverness-shire, Scotland, many years ago, and settled in Antigonish
county. His brother, Angus McIsaac, now judge of the county court
of Antigonish, sat in the House of Commons at Ottawa from 1873 to
1885. Colin F. McIsaac was educated at St. François Xavier College,
in his native county. Having adopted law as a profession, he devoted
some years to study, and on 12th January, 1880, was admitted to
the bar of Nova Scotia, since which time he has successfully
practised his profession. In 1882 he was elected a governor of St.
François Xavier College, and has occupied this position ever since.
He entered political life in 1886, and at the general election held that
year was elected by a handsome majority to represent Antigonish in
the Nova Scotia legislature. In politics Mr. McIsaac is a Liberal; and in
religion is a member of the Roman Catholic church.

Philp, Rev. John, M.A., Pastor St. James Street Methodist

Church, Montreal, is a Canadian by birth, having been born in the
town of Cobourg, in the province of Ontario. His father, the Rev.
William Philp, a native of Cornwall, England, was for nearly forty
years a minister of the Wesleyan Methodist church in Canada. His
mother was a person of rare excellence of heart and life. Rev. Mr.
Philp, the subject of our sketch, was educated at Victoria University,
Cobourg, and graduated in the arts course in 1861, receiving the
B.A. degree, and taking the Hodgins prize. Three years later the
same university conferred upon him the M.A. degree. He entered the
ministry of the Wesleyan Methodist church in 1860, when in the
twentieth year of his age, and while yet at college. In June, 1861, he
received his first appointment. In June, 1865, he was married to
Miss Maggie Grafton, of Dundas, Ontario, and has a family of four
children, living, the eldest of whom will soon graduate in medicine in
McGill University, Montreal. His stations in order have been Oakville,
Dundas, St. Mary’s, Woodstock, Fairfield, Windsor, Paris, St. Mary’s,
Queen’s Avenue Church, London; Wesley Church, Hamilton; Carlton
Street Church, Toronto; St. James Street Church, Montreal, and few
ministers in the denomination can show more work done in the
Master’s vineyard during nearly twenty-eight years. Fairfield circuit,
near Brantford, was his first superintendency, and here extensive
revivals of religion took place. One of the finest country churches
was erected on what is known as Fairfield Plain. At Windsor during
his term, the small frame building in which the congregation had
long worshipped was superseded by a new and attractive brick
church. At Paris, the present beautiful sanctuary was built, and thus
the interests of Methodism there greatly furthered. At St. Mary’s, a
remarkable temperance movement took place, in which over two
thousand signed the pledge. In this he took an active part. During
his term in London, the Queen’s Avenue Methodist Church was
modernised and beautified, at a cost of $14,000. And since he took
charge of the St. James Street Church, Montreal, the congregation,
by their spirit and liberality, have begun a signally important and
greatly needed work—the erection of a representative church in a
more central part of the city; which, when it is completed will be the
most imposing and commodious religious edifice of Methodism in
Canada. Mr. Philp has received many marks of esteem and
confidence on the part of his brethren in the church. He has been
the secretary of the London Conference; secretary of the Examining
Board; delegate to the General Conference; Conference examiner for
the Theological College, Montreal; preacher of one of the
baccalaureate sermons for Victoria University, &c. Though greatly
devoted to his ministerial duties, he has found time to go abroad a
little, making a trip to Britain, and a tour through the Continent,
sailing up the Rhine, lingering amid the glories of Alpine scenery,
and viewing some of the principal cities of Europe. While in England,
nothing affected the reverend gentleman more than a visit to the
little Cornish town where his (now sainted) parents were born and
spent their early days. From that unpretentious centre (Lostwithiel
and its immediate vicinity) came eight young men who knew each
other in their boyhood, and who entered the ministry of the
Methodist church in Canada, doing blessed and effective service for
Christ. Most of them have since passed to their reward. Rev. Mr.
Philp is a firm believer in the great doctrines of truth as held by the
Methodist church, not because they are the creed of the church, but
because they are the vitalities of Christian life; preeminently, the
divinity of the Son of God, the vicarious character and sufficiency of
the atonement, free and full salvation alone through faith in Christ,
regeneration, the witness of the Spirit, the divine authority of the
Holy Scriptures. But, while strongly attached to the Methodist
church, he would scorn all narrowness of thought and view, all
servile devotion to mere dogma, all sectarian prejudice and caste,
and would most fervently pray with the Apostle, “Grace be with all
them that love our Lord Jesus Christ in sincerity.” He believes that
the pulpit should be progressive in its spirit and aim; abreast of the
times in sanctified scholarship and power to teach, wisely
conservative in its doctrinal tendency (by which he means not too
eager to hail and foster new things), gospel in its character, never
descending to the level of the mere lecture platform, or wasting its
energy in mere speculative enquiry. Its one mission should be to
preach Christ with all tenderness, simplicity, earnestness and
directness as the sinner’s hope, the world’s saviour. The Montreal
Daily Star, of the 24th October, 1887, thus kindly speaks of the Rev.
Mr. Philp:—“He is a comparatively young man, although his
ministerial work has been much greater and more varied than falls to
the lot of men of his years. In the pulpit he presents the appearance
of a man of great intellectual power, and his delivery bears out the
impression, as his discourses are logical and keenly analytic. His
elocution is easy, and increases in animation as he approaches the
conclusion and application of his arguments. Mr. Philp is noted as a
successful revivalist, and he has held in many places large meetings,
and by his earnest, self-denying labors in every station in which he
has labored caused large increases in the membership of his church.
While especially active in forwarding the advance of the spiritual
interests of his flock, Mr. Philp is not forgetful of the fact that the
handmaids of religion, architecture, music, etc., have also their
influence on the people. He has endeavored to promote their
cultivation in available forms, and his efforts in the direction of
improving the ecclesiastical structures over which he has had
control, and the erection of others, have been peculiarly successful.”

Paton, Hugh, General Manager and Secretary of the

Shedden Company, Montreal, was born at Johnstone, Renfrewshire,
Scotland, on the 5th October, 1852. His parents were William Paton
and Mary Shedden, of Kilbirnie, Ayrshire, Scotland. Mr. Paton
received his education in the Grammar School of Paisley, Scotland.
In 1871 he came out to Canada, to reside with his uncle, the late
John Shedden, railway contractor, Toronto. He entered Mr. Shedden’s
office in Toronto, and remained there until that gentleman’s untimely
death in 1873, he having been killed by a train on the Toronto and
Nipissing Railway, when celebrating the opening of that line, which
he had built. Mr. Paton then removed to Montreal, where he has
since resided, and became secretary-treasurer of the Shedden
Company, general forwarders and carriers, and cartage agents for
the Grand Trunk Railway, which succeeded to the business of the
deceased Mr. Shedden. This position he occupied until 1879, when
he became manager and secretary, and this office he still holds; and
we say here that Mr. Paton is now the principal proprietor of this
company. He was honorary secretary-treasurer of the Province of
Quebec Turf Club for four years; and honorary secretary-treasurer of
the Montreal Tandem Club for two years. From 1879 to 1886 he was
honorary secretary-treasuror of the Montreal Hunt; and this year
(1887) he was elected master of the fox hounds, and that position
he now holds. He has always taken an interest in racing and in
agricultural pursuits, and is the owner of a farm near the city of
Montreal where he resides during the summer months, and where
he indulges in his favorite pursuits. He has owned several “Queen’s
platers” and “steeple-chasers.” Since Mr. Paton took charge of the
Shedden Company he has considerably extended its ramifications
over the Dominion of Canada and the Western States of America,
carrying on the business of contractors, forwarders, and carriers,
owning about eight hundred horses, and grain and general storage
warehouses at several points. In 1879 he made an extended tour
over the continent of America, visiting nearly all the western states,
and among other places of note Salt Lake City, the Yosemite valley,
and San Francisco. Since then he has twice travelled over the
continent of Europe. Mr. Paton is a Liberal-Conservative in politics;
and in religion is an adherent of the Presbyterian church. He was
married in 1884 to Bella Robertson, daughter of Andrew Robertson,
formerly merchant, Montreal, and now chairman of the Montreal
Harbor Commission.
 De Wolfe, Charles Edgar, Windsor, Barrister, Judge of
Probate, and Revising Barrister for the county of Hants, Nova Scotia,
was born in the town of Windsor, the shire town of the county of
Hants, Nova Scotia, on the 22nd of July, 1845. His parents were
James Lovitt De Wolfe, and Margaret A., daughter of the late
Thomas Lovett, of Cornwallis, Kings county, Nova Scotia. Their
children were Charles Edgar, Sarah Frances, widow of the Rev. H. P.
Almon; Amelia Isabella; Benjamin Arthur, who died 17th February,
1845; James Lovitt, a doctor, residing in England; Benjamin Alfred,
who died 17th August, 1851; Perez Morton, head of the well-known
book firm of De Wolfe, Fiske & Co., Archway Book Store, 365
Washington street, Boston; Annie, wife of W. I. Fenwick, broker,
Montreal, and Mary Agnes. J. L. De Wolfe was a lawyer, studied law
in the office of the late Judge L. M. Wilkins; and subsequently, and
for many years, Mr. De Wolfe was engaged in mercantile business in
the town of Windsor, in the widely-known firm of B. De Wolfe & Son,
in which he was junior partner. He died on 16th April, 1863, and his
wife died on 23rd November, 1886. Benjamin De Wolfe, grandfather
of C. E. De Wolfe, was the senior partner in the before-mentioned
firm. He was member for Hants county in 1827-31. He married a
Miss Lovitt, of the city of St. John, New Brunswick. They had four
children, James Lovitt, Benjamin, lost at sea when acting as
supercargo of a vessel; George, a medical student, also dead; and
Sarah L., who now resides in Windsor. Benjamin De Wolfe, senior,
died 9th December, 1863. Loran De Wolfe, the father of Benjamin,
was born at Say Brook, Connecticut, 7th April, 1754. He resided in
Windsor, or about three miles from it, the greater portion of his life.
He married Mary Fox, of Cornwallis, Kings county. They had five
children, viz., Benjamin, Phœbe M., George, Hannah, and Isaac. In
1791 Loran De Wolfe was assessor for the town of Windsor. The
“Baptist Missionary Magazine” for April, 1835, in an obituary notice
of him remarks: “As an instance of the public confidence in this
worthy man, we may remark that he was elected in 1812 by
acclamation, to represent the township of Windsor in the General
Assembly. He retained his seat until ill health obliged him to retire
from public life.” Nathan De Wolfe, father of Loran, and great-great-
grandfather of C. E. De Wolfe, was born in Say Brook, Connecticut,
in 1720, graduated A.M. in 1743, at Yale College, New Haven, and
was engaged in the practice of law. He had previously “owned the
covenant,” or joined the Congregational church, 7th June, 1741. He
married, first, about 1748, Lydia Kirtland, daughter of John Kirtland.
His second wife was Lydia Beldon, born at Say Brook, October 28th,
1721. Their children were Lucilla; Edward, born 1752; Loran, born
7th April, 1754; Elisha, born 5th May, 1756, and Nathan. Nathan,
senr., came to Horton, Kings county, Nova Scotia, in 1760. He was a
successful farmer. The Nova Scotian census returns of 1770 give a
detailed account of his farm produce and stock. His residence was
on the east side of the main post road, opposite to the present
Baptist church, in Wolfville (1887). His legal practice did not interfere
with his agricultural pursuits. He was for many years senior justice of
the peace for Kings county. He was also registrar of probate, and
took an active part in public affairs. He died at Horton on the 21st of
March, 1789, aged sixty-nine years. Charles Edgar De Wolfe, the
subject of our sketch, after receiving the usual course of education,
entered upon the study of law in the office of the solicitor of the city
of Portland, Maine, U.S., and continued his studies in the office of S.
L. Morse, Q.C., Bridgetown, Annapolis county, Nova Scotia, and
completed them in the office of the Hon. James Macdonald, Halifax,
the present chief justice of Nova Scotia. He was admitted to the bar
on the 19th of July, 1870. In 1880 he was appointed judge of
probate for Hants county; and in 1886 was made revising barrister
for the same county, under the Electoral Franchise Act, and both
offices he still continues to hold. Previous to his becoming a law
student, Mr. De Wolfe owned and edited a weekly newspaper in
Bridgetown, and for five years he was proprietor and editor of the
Windsor Mail, published at Windsor, N.S. He has taken a deep
interest in the temperance movement, and on various occasions, and
in different places in his native province, delivered strong addresses
on its behalf. In politics he is a Liberal-Conservative, and has been
an active party man in his county. On the 12th of October, 1887, he
was married to Cassie H., daughter of Samuel Grey, of New Annan,
Colchester county, N.S. Mrs. De Wolfe was, before her marriage, a
captain in the Salvation Army, and a very zealous worker for God
and humanity in the provinces of Nova Scotia and Newfoundland,
but through ill health, had to retire from active service in the army.
Mrs. De Wolfe has a sister in India, working there as a missionary,
under the direction of the Baptist Board of Foreign Missions, of Nova
Scotia.

Killam, Amasa Emerson, Moncton, Manager of the St.

Martin’s and Upham Railway, M.P.P. for Westmoreland county, New
Brunswick, was born on the 25th of August, 1834. His parents were
born in New Brunswick, his father on the 26th of March, 1811, and
his mother on the 10th of May, 1812. His paternal grandfather was
an officer in the British army, and served during the American war of
independence, and on the declaration of peace came to New
Brunswick and settled at Sackville. His grandparents, on the
mother’s side, were U. E. loyalists, and also became settlers in the
Maritime provinces. Mr. Killam received his education at the common
schools of his native place. He held the position of postmaster for a
number of years, and is now manager of the St. Martin’s and Upham
Railway, and in 1884 purchased the Elgin, Petitcodiac and Havelock
railway, from Petitcodiac to Elgin, and in 1885 built the extension of
the road to Havelock, and became managing director, and in 1886
took an interest in building the Central Railway, from Norton to
Fredericton, and is managing director of the company: also
managing director of the Buctouche and Moncton railway. He first
entered the House of Assembly after the general election held in
1878, as representative of Westmoreland county. At the following
general election he failed to be returned; but in September, 1883, on
the resignation of P. A. Landry, who was elected to the House of
Commons at Ottawa, Mr. Killam was chosen to fill the vacancy. At
the general election held in 1886 he again came before his
constituents, and was once more chosen their representative in the
local house. In politics he is a Liberal-Conservative. On the 25th July,
1857, he was married, at Sackville, to Millicent Wheaton, and the
fruit of the union has been seventeen children.

Young, Sir William, LL.D., ex-Chief Justice of Nova

Scotia, Halifax.—The late Sir William Young, who was a Scotchman
by birth, was born at Falkirk, in 1799, and died at Halifax, on the 8th
of May, 1887. He was a son of John Young, of Falkirk, Stirlingshire,
Scotland, who, many years ago, emigrated to Nova Scotia, making
Halifax his home. His son William received his education at the
University of Glasgow, where he took honors. He then took up law
as a profession, was admitted to the bar of Nova Scotia in 1826, and
appointed Queen’s counsel in 1843. In commencing his career as a
lawyer, he had some advantages over most young men, in his family
connections, which were quite numerous. But he, wisely, did not too
largely depend on this for success; he was well-read, clear-headed,
energetic, and bound to get on through his own inherent powers
and perseverance. When he had established his reputation at the
bar, and became comparatively independent in circumstances, he
entered the Legislative Assembly of Nova Scotia, having been
returned in 1833 to represent the island of Cape Breton when it
formed an electoral district. Subsequently, when the island was
divided, he represented Inverness, extending over a period of
twenty-two years—from 1837 to 1859. In the latter year he
successfully contested the county of Cumberland against Dr. (now Sir
Charles) Tupper. Sir William being at that time leader of the Liberals,
or opposition, and Dr. Tupper that of the Conservatives, or
government party. Cumberland returned both these gentlemen to
the Legislature, there being three members, Sir William, however,
taking the lead. Shortly after getting into parliament Sir William
showed his boldness of spirit and manly independence by entering
his protest against the unjust coal mining monopoly then in
existence, which had been granted by the Crown to the creditors of
the late Duke of York, a monopoly which he and his brother George
were largely instrumental in having removed at a later date. In
1838, during the closing scenes of the Canadian rebellion of that
time, he was appointed as a delegate, with others, to meet Lord
Durham, and discuss the numerous grievances of which the French
population complained. The grievances of his own province he
exposed in a letter of vigorous remonstrance, which Lord Durham
afterwards annexed to his celebrated report. His associates on this
memorable occasion were Mather B. Almon, J. W. Johnson and Jas.
B. Uniacke, and sad to say the last of these delegates in the person
of Sir William Young has now passed away. They met Lord Durham
in Quebec, and in the several interviews with his lordship and his
suite, and representatives from the several other provinces, they laid
the foundations of the confederacy which in July, 1867, was
perfected. In 1839 Sir William Young and Herbert Huntington were
sent to Britain to impress upon the home government the removal of
grievances existing in Nova Scotia, such a delegation having been
found necessary, Lord Durham having thrown up his office, and
returned to England in disgust. These delegates showed a
considerable amount of tact and diplomatic skill, and their mission
advanced the interests of the people in many ways. Their report,
which shortly after their return was published, covered a wide field,
and exhibited an active correspondence with the several
departments of the Imperial government, from which valuable
concessions were obtained. During the long period Sir William
served in parliament he was a prominent figure in that body, acting
either as chairman or leading member on almost every important
committee. He became a member of the Executive Council in 1842.
In 1843 he was elected speaker of the Legislative Assembly, and
occupied this office for eleven consecutive years. In 1854 he became
leader of the government as well as attorney-general; and leader of
the opposition in 1857, a change of government having taken place.
In 1859 he was chosen president of the Executive Council. For all
this period, even when in the speaker’s chair, the impress of his
strong mind was visible in almost every important measure, as the
journals of the house amply testify, from the time he first attacked
the coal mine monopoly of the creditors of the Duke of York, to the
time of his retirement from the arena of politics. In 1851 he was
associated with Messrs. Ritchie and McCully, both of whom
afterwards were, like himself, made judges, in revising the statutes
of Nova Scotia; and on the floor of the house he was the recognized
spokesman of the agriculturists of the province, “a legacy,” which he
often jokingly remarked, “had possibly descended to him from his
father, the famous ‘Agricola,’” a then popular writer on agriculture
both as a science and as an art. In 1860 he retired from political life,
and was appointed chief justice of Nova Scotia, and this office he
resigned in 1881 on account of age. When appointed to the chief
justiceship he brought to the discharge of his high duties a clear
intellect, a sound understanding of law, and a well-trained judicial
mind, and during the time he sat on the bench he attended to its
duties faithfully. His quick apprehension of points of both law and
practice, his searching insight into all matters of a difficult or
abstract character, made him distinguished as a judge and respected
by the bar. In 1876 Sir William started on a six months’ tour in
Europe, and, just before he left, the bar of Nova Scotia, and the
mayor and corporation of Halifax presented him with addresses,
which bore feeling testimony to his eminent services in the
legislative halls, on the bench, and as a citizen in all the various
spheres of life. To these addresses he made an off-hand and very
happy response, showing the cordiality of his disposition and warmth
of heart, as well as his readiness and ability as a speaker. In 1868 he
received the honor of knighthood from her Majesty Queen Victoria;
and in 1881 the degree of LL.D. was conferred on him by Dalhousie
College. Sir William Young was married, in 1830, to Anne, daughter
of the late Hon. Michael Tobin, M.L.C. She died at Halifax on the
12th January, 1883, at the age of seventy-nine years. Few ladies in
Halifax were more generally known or more sincerely respected than
Lady Young. She was a lifelong contributor to all public charities of
the city, and in her more active years was prominently connected
with every benevolent undertaking. Sir William Young was possessed
of considerable means at the time of his death, and by his will he
left his possessions to various educational, charitable, and other
institutes in the city in which he had lived and been so benevolent
and public spirited a citizen for the greater part of a century.
 Cannon, Lawrence Ambrose, Quebec, Advocate, and
Clerk of the Corporation of Quebec city, was born at Quebec on the
20th March, 1814. His father, John Cannon, architect, was of Irish
parentage, and his mother, Angèle Griault dite Larivière, was of
French descent. Mr. Cannon, senr., was a member of the Legislative
Assembly of Lower Canada from 1824 to 1830, and represented the
county of Hampshire, then comprising the present counties of
Portneuf and Champlain. He was an Independent in politics, and
though not unfriendly to the powers that then ruled, was much
attached to the principles of the French Canadian party in the
Assembly. He was a strenuous supporter of the Autonomists, who, at
the time of the first proposed union of Lower and Upper Canada,
exerted themselves so strongly that they succeeded in defeating the
measure. Mr. Cannon was also above all an Irishman, and although
living in Canada, he deeply sympathized with every movement
calculated to advance the prosperity of the land of his birth. He, too,
helped his fellow-countrymen in the land of his adoption, and
contributed largely by his exertions and means to have erected in
Quebec St. Patrick’s Church, which stands to-day as a monument to
the religious ardor and generosity of the Irish race in the ancient
capital. He was twice married; first to Angèle Griault dite Larivière;
and the second time, in 1826, to widow Rosslewin, née Archange
Baby. Lawrence Ambrose Cannon, the subject of our sketch, was
educated first in private English schools, and afterwards in the
Quebec Seminary, where he prosecuted and terminated his classical
course of studies in 1833. He entered as a law student the office of
Hon. C. R. Ogden, then attorney-general of Lower Canada; and in
1836, that of Stuart and Black. When he had completed his legal
studies in 1838, he was called to the bar, and continued to practise
his profession until May, 1864. On the resignation of F. X. Garneau,
the Canadian historian, the city clerk of Quebec, through ill-health,
Mr. Cannon was elected to fill this important position, and he has
done it faithfully ever since. Some time after his appointment, the
charter of the city was materially amended, thus considerably
increasing his duties. Among other important changes, he was
charged with the preparing of the lists of the parliamentary electors,
and also of the persons qualified to be called upon to act as grand
and petit jurors. And by the Act 33 Vict., chap. 46, the sole
management of the municipal elections was conferred upon him. He
married in 1845, Mary Jane Cary, daughter of the late Thomas Cary,
then proprietor and publisher of the Quebec Mercury, and of Marie
Anne Dorion. He has three surviving children; one son, Lawrence
John Cannon, a practising barrister in Arthabaskaville, and two
daughters.

Torrance, David, Montreal.—Mr. Torrance, during his

lifetime one of Montreal’s most successful and distinguished
merchants, was of Scotch parentage. He was born in New York in
1805, and died in Montreal Jan. 29th, 1876. When yet a boy he
came to Kingston, Upper Canada (now Ontario), with his father,
James Torrance, who was then extensively engaged in business in
that town. In 1821 he removed to Montreal, and became a clerk
with his uncle, the late John Torrance, who kept a place of business
at the corner of St. Paul and St. Nicholas streets. By his close
attention to his duties, and aptitude to the work, he rapidly rose in
his employer’s estimation, which ended in his being taken into
partnership in 1833. During his clerkship the late Rev. Dr. Wilkes,
and the late Hon. John Young were engaged in the same
establishment. With the view of extending the business of the
concern, in 1835 Mr. Torrance entered into partnership with Mr.
Young, of Quebec, under the firm name of Torrance & Young; and
on the retirement of the late John Torrance, the senior member, the
firm’s name was changed to that of D. Torrance & Co., which
continued to the date of his demise, his partners being for many
years Thomas Cramp, and his son John Torrance. In 1826 the old
firm purchased a tug and passenger boat, the Hercules, from John
Handyside & Co., and placed it under the command of Captain
Brush, who afterwards became the proprietor of the Eagle Foundry,
Montreal. This was the first step towards the establishment of an
opposition line to the Molsons’ steamboats, then plying on the St.
Lawrence, and its stock was eventually merged in that of the
Richelieu Steamboat Company (now the Richelieu and Ontario
Navigation Company). Mr. Torrance was early alive to the great
future in store for Montreal, and was the first to launch out into
direct trade with the East Indies and China, and for over thirty years
the name of his firm has been well known in the great eastern
centres of commerce. As a business man he was remarkable for
great force of character and determination. This, combined with
unflinching industry and regular habits, made the immense business
of the firm move ahead with precision. An old friend of his once said
of him, “He was a model man in regard to his business and social
habits, and in the days of his prosperity was as regular in his
attendance at the counting house as when he first started in
business. His ambition was great, but tempered with prudence, and
though he engaged in commercial ventures in other cities than
Montreal, yet was uniformly successful.” Besides his promotion of
commerce and navigation, he likewise proved himself a stay to our
banking system, and after holding office for a long time as one of
the directors of the Bank of Montreal, he was in 1873 elected
president, which responsible position he held at the time of his
death. His firm was also one of the originators of the Dominion
Steamship Company. While largely engaged in ocean commerce, his
capital and resources were also devoted to the carrying on of our
inland forwarding trade. He was a diligent merchant, and did not
meddle much in public affairs, though he was a consistent Liberal in
politics throughout. To all benevolent and charitable schemes he was
a frequent and liberal giver. He was always ready to aid the
distressed and bring joy to those in want, and the main feature in
this regard was the unostentatious way in which he helped those in
need. He was a member of the St. James Street Methodist Church,
and at the time of his death was one of its trustees. He was, in fact,
the thorough type of a merchant prince, a representative of a class
which, unfortunately, is far too small in these latter days. He was
married to his cousin, the eldest daughter of the late John Torrance.
He was in feeble health for some years previous to his death, and
had only a few months before to forego active business, and when
death at last came he passed away quietly, surrounded by his
sorrowing family.

Skinner, Hon. Charles N., Q.C., St. John, ex-Judge of

Probate for the county of St. John, New Brunswick, was born in St.
John on the 12th March, 1833. His father, Samuel Skinner, was a
contractor and builder, and was a native of Nova Scotia. His mother,
Phœbe Sherwood, was a daughter of Robert Golding, whose
grandfather, Captain Golding, commanded a company of loyal
dragoons during the American revolutionary war. Both the Skinner
and Golding families were loyalists and emigrated from the New
England states—Mr. Skinner, the grandfather of the subject of our
sketch, a short time before the outbreak of the revolution, and Mr.
Golding after the war—and settled in the Maritime provinces. Charles
N. Skinner received his education in the common and grammar
schools of St. John. He studied law under Charles W. Stockton, of
that city; was admitted to practice in 1858, and called to the bar in
1860. Since then he has successfully practised his profession in his
native city. He is a well-read lawyer, a fluent, clear, and logical
speaker, and seldom fails to present his case in the best possible
light before a jury. His mind, too, is of a judicial cast; he is candid,
honest, and impartial, and is admirably fitted by nature for the
position he holds. When only about twenty-eight years of age he
entered the field of politics, and was elected to represent St. John in
the Legislative Assembly of New Brunswick in 1861. After being in
the house three years, the party with whom he was allied was
defeated on the question of confederation. In 1866 he again
appealed to his constituents, and was elected. During August of next
year he was appointed solicitor-general in the A. R. Wetmore
administration, and this office he held until March, 1868, when he
retired from political life, having been made a judge of probate. He
was also created a Queen’s counsel that year by the Provincial
government, and by the Dominion government in 1873. He was
elected to the Dominion parliament to represent the city and county
of St. John at the general election in 1887, having previously
resigned the office of Judge of Probate. He still practises at the bar
of St. John, and stands high among his confrères. For some years he
was a member of St. John city council, and took an active interest in
all matters brought forward for the benefit of the citizens. He is a
member of the brotherhood of Oddfellows. He belongs to the Baptist
denomination, and is considered a man of unblemished character
and liberal impulses. On the 12th January, 1865, he was married to
Eliza Jane, daughter of Daniel J. McLaughlan (then president of the
Commercial Bank of N.B.), of St. John, and the fruit of this union has
been a family of eight children.

Fenwick, George Edgeworth, M.D., C.M., Montreal,

was born in the city of Quebec, on the 8th October, 1825. His father,
Joseph Fenwick, in early life entered the East India Company’s
service, and subsequently, in command of his own ship, traded
between London and the port of Montreal. He was from Morpeth,
Northumberland, England. His mother, Margaret Elizabeth Greig, was
a native of Quebec, of Scotch descent. His grandfather belonged to
the landed gentry of Northumberland. Dr. Fenwick received his
education under the Rev. Mr. Ramsay, a clergyman of the Church of
England; and in June, 1841, began the study of medicine and
surgery in the Marine and Emigrant Hospital in his native city. His
brother, Dr. A. G. Fenwick, was at that time house-surgeon to that
institution, and he acted under him as house apothecary. He
remained in this position until November, 1842, when he entered the
medical department of McGill College, in Montreal. He successfully
passed his examination in May, 1846, but not being of age did not
receive his diploma until January, 1847, when a special convention of
the University was called for the purpose of conferring upon him the
degree of doctor in medicine and master in surgery. In May
following, Dr. Fenwick was appointed house-surgeon and apothecary
to the Montreal General Hospital, which office he filled until
December, 1848, when he commenced general practice in Montreal.
In 1849 he aided, in conjunction with Dr. Howard, the late Dr. G. D.
Gibb (afterwards Sir G. D. Gibb, baronet, M.D., of London, England),
and the late Drs. Pelletier, Boyer and Jones, in establishing the
Montreal Dispensary, and was one of the attending staff of that
institution until November, 1864, when, on the death of Dr. Thomas
Walter Jones, he received the appointment of attending surgeon of
the Montreal General Hospital. In 1867 he was appointed professor
of clinical surgery in McGill University, and held this position until
1876, when, on the resignation of the late Dr. George W. Campbell,
he was appointed professor of surgery, which chair he has filled to
this time. As a teacher Dr. Fenwick has had long experience in the
teaching of surgery. For many years professor of clinical surgery, his
lectures were all delivered in the General Hospital, and every student
who had the privilege of visiting the wards during his term of
service, knows well the keen interest he took in everything
concerning the cases in hand. Careful and painstaking himself, he
firmly exacted from his assistants, house officers and dressers, a like
degree of attention and carefulness in little things. After his
promotion to the chair of surgery his lectures were of a more
didactic nature, but to them he brought the same spirit of earnest
devotion to the cause of science, the same grasp of subject leading
to the formation of opinions strongly held, the same care for the
important minutiæ, and the same genial and impressive manner
which characterized his early teachings in the wards. In 1864 Dr.
Fenwick, with his colleague, Dr. F. W. Campbell, established the
Canada Medical Journal, which he continued to edit until 1879, when
he relinquished the editorial chair. As a medical writer he is probably
as well known as any in Canada. His articles upon surgical subjects
are all terse and logical, and carry the impress of a vigorous and
thoughtful mind. His most important papers are those upon
lithotomy, of which operation he has probably had a larger
experience than any other living surgeon in the Dominion. On
excision of bronchocele, his bold operations have commanded the
most wide-spread attention, and on excision of the knee-joint and
other major operations he has been remarkably successful. He holds
the degree of M.D., C.M. from his first university, and has never
sought medical honors from any institution abroad; nevertheless, he
has been considered worthy of them. He has been elected honorary
member of the New Brunswick Medical Society; of the Medical
Society of Nova Scotia, and of the Gynæcological Society of Boston.
For many years Dr. Fenwick represented the profession of Montreal
as one of the governors of the College of Physicians and Surgeons of
Quebec province. He has served as president of the Medico-
Chirurgical Society of Montreal; and was, at the meeting of the
Canada Medical Association, held at Ottawa in 1880, elected vice
president of that body for Quebec province, and in 1882 was elected
president of the association. Dr. Fenwick is an adherent of the
Episcopal church; and in politics a Conservative. In 1852 he was
married to Eliza Charlotte, daughter of the late Colonel de Hertel, of
St. Andrews, Argenteuil. The fruit of this union has been seven
children, only three of whom survive.

Adams, Rev. Thomas, M.A., Cambridge, D.C.L.,

Lennoxville, was born at Paramatta, New South Wales, on
September 14, 1847. His father, the Rev. Thomas Adams, was a
member of a family in Cornwall, England, of which the eldest brother
is J. C. Adams, F.R.S., the celebrated discoverer of the planet
Neptune, who, on the retirement of Sir G. B. Airy, declined the
position of Astronomer Royal of Great Britain, and is still director of
the Cambridge University. Another brother (W. G. Adams, F.R.S.), is
a leading authority on electricity and natural philosophy, and
occupies the professorial chair in King’s College, London, once held
by Wheatstone, and afterwards by Clerk Maxwell. The father of
Principal Adams became a missionary in the Friendly Islands (South
Pacific), and it was in Australia, on the way to that mission, that Dr.
Adams was born. Thomas Adams, sen., is chiefly noted for having
been the translator of a great portion of the Bible into Tonguese, and
for having been the first who issued a complete edition of the Sacred
Book in that language. His mother was Maria French, of Taunton,
Somerset. She accompanied his father into the mission field, and
gave her life to the work. She died in Vavau in February, 1860.
Professor Adams was educated first at Taunton, Somerset, at a large
proprietary school, under T. Sibly, B.A.; next at University College,
London, under the late Professor de Morgan, in mathematics, and
Professor J. R. Seeley, in classics. In November, 1867, he joined the
geological survey of England, under Sir A. C. Ramsay, but resigned
in April, 1869, owing to a severe sprain. In October, 1869, he
entered St. John’s College, Cambridge, and in January, 1873,
graduated as 19th wrangler in a first class of thirty-seven. After
acting temporarily as professor of mathematics in the Royal
Agricultural College, Cirencester, he was appointed mathematical and
science master in the Royal Grammar School, Lancaster, and in
August, 1874, he became senior mathematical master in the Royal
School of St. Peter’s, York. He was ordained deacon in 1874, and
priest in 1876, by the present archbishop of York. In 1881, on the
occasion of the jubilee meeting of the British Association in York, in
conjunction with Dr. T. Anderson, he became local secretary. In
December, 1882, he was elected, out of fifty-seven competitors, as
the first head master of the High School for boys, Gateshead-on-
Tyne, and left there a school of one hundred and fifty boys to accept
the position he now holds of principal of the University of Bishop’s
College, and rector of the College School, Lennoxville, province of
Quebec. He has held this position since August, 1885, and
succeeded Dr. Lobley in both offices. In July, 1878, he was married
to Annie Stanley, youngest daughter of the late T. Barnes, of
London, England.

Turnbull, Lieut.-Colonel James Ferdinand,

Commandant of the Royal School of Cavalry, Quebec city, was born
in London, England, on the 19th July, 1835, and baptized at
Westerham, in Kent, in the same font that had done duty to the ever
immortal General Wolfe. He is the eldest son of the late James