Supervised Machine Learning for Text Analysis in

R 1st Edition Emil Hvitfeldt pdf download

https://ebookname.com/product/supervised-machine-learning-for-text-analysis-in-r-1st-edition-emil-
hvitfeldt/

★★★★★ 4.6/5.0 (27 reviews) ✓ 107 downloads ■ TOP RATED

"Excellent quality PDF, exactly what I needed!" - Sarah M.

DOWNLOAD EBOOK
 Supervised Machine Learning for Text Analysis in R 1st
Edition Emil Hvitfeldt pdf download

TEXTBOOK EBOOK EBOOK GATE

Available Formats

■ PDF eBook Study Guide TextBook

EXCLUSIVE 2025 EDUCATIONAL COLLECTION - LIMITED TIME

INSTANT DOWNLOAD VIEW LIBRARY

Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...

Text Mining with Machine Learning Principles and

Techniques 1st Edition Jan Zizka

https://ebookname.com/product/text-mining-with-machine-learning-
principles-and-techniques-1st-edition-jan-zizka/

Hands On Machine Learning with R 1st Edition Brad

Boehmke (Author)

https://ebookname.com/product/hands-on-machine-learning-
with-r-1st-edition-brad-boehmke-author/

Phase Transitions in Machine Learning 1st Edition

Lorenza Saitta

https://ebookname.com/product/phase-transitions-in-machine-
learning-1st-edition-lorenza-saitta/

SAS 9 2 Language Reference Dictionary 1st Edition Sas

Publishing

https://ebookname.com/product/sas-9-2-language-reference-
dictionary-1st-edition-sas-publishing/
Science of the Sages Scientists Encountering Nonduality
from Quantum Physics to Cosmology to Consciousness
Robert Wolfe

https://ebookname.com/product/science-of-the-sages-scientists-
encountering-nonduality-from-quantum-physics-to-cosmology-to-
consciousness-robert-wolfe/

Signs of Cherokee Culture Sequoyah s Syllabary in

Eastern Cherokee Life Margaret Bender

https://ebookname.com/product/signs-of-cherokee-culture-sequoyah-
s-syllabary-in-eastern-cherokee-life-margaret-bender/

The Sky of Our Manufacture The London Fog in British

Fiction from Dickens to Woolf Jesse Oak Taylor

https://ebookname.com/product/the-sky-of-our-manufacture-the-
london-fog-in-british-fiction-from-dickens-to-woolf-jesse-oak-
taylor/

Craft Hope Handmade Crafts for a Cause 1st Edition Jade

Sims

https://ebookname.com/product/craft-hope-handmade-crafts-for-a-
cause-1st-edition-jade-sims/

The Facts on File Companion to the American Short Story

2nd Edition Companion to Literature Series Abby H. P.
Werlock

https://ebookname.com/product/the-facts-on-file-companion-to-the-
american-short-story-2nd-edition-companion-to-literature-series-
abby-h-p-werlock/
Frommer s Italy 2013 8th Edition Donald Strachan

https://ebookname.com/product/frommer-s-italy-2013-8th-edition-
donald-strachan/
Supervised Machine
Learning for Text
Analysis in R
Supervised Machine
Learning for Text
Analysis in R

Emil Hvitfeldt
Julia Silge
First edition published 2022
by CRC Press
6000 Broken Sound Parkway NW, Suite 300, Boca Raton, FL 33487-2742

and by CRC Press

2 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN

© 2022 Taylor & Francis Group, LLC

CRC Press is an imprint of Taylor & Francis Group, LLC

Reasonable efforts have been made to publish reliable data and information, but the author and pub-
lisher cannot assume responsibility for the validity of all materials or the consequences of their use.
The authors and publishers have attempted to trace the copyright holders of all material reproduced
in this publication and apologize to copyright holders if permission to publish in this form has not
been obtained. If any copyright material has not been acknowledged please write and let us know so
we may rectify in any future reprint.

Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced,
transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or
hereafter invented, including photocopying, microfilming, and recording, or in any information stor-
age or retrieval system, without written permission from the publishers.

For permission to photocopy or use material electronically from this work, access www.copyright.
com or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA
01923, 978-750-8400. For works that are not available on CCC please contact mpkbookspermis-
[email protected]

Trademark notice: Product or corporate names may be trademarks or registered trademarks and are
used only for identification and explanation without intent to infringe.

ISBN: 978-0-367-55418-7 (hbk)

ISBN: 978-0-367-55419-4 (pbk)
ISBN: 978-1-003-09345-9 (ebk)

DOI: 10.1201/9781003093459

Typeset in LMR10 font

by KnowledgeWorks Global Ltd.
 In loving memory of my mother-in-law Lisa, who was the first soul to hear
about and fully encourage the idea that eventually became this book —E.H.

For Grace, Violet, and Lewis, who (thanks to the pandemic and remote
school) had a front row seat to most of my work on this book —J.S.
Contents

Preface xiii

I Natural Language Features 1

1 Language and modeling 3
1.1 Linguistics for text analysis . . . . . . . . . . . . . . . . . . . 3
1.2 A glimpse into one area: morphology . . . . . . . . . . . . . 5
1.3 Different languages . . . . . . . . . . . . . . . . . . . . . . . 6
1.4 Other ways text can vary . . . . . . . . . . . . . . . . . . . . 7
1.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.5.1 In this chapter, you learned: . . . . . . . . . . . . . . . 8

2 Tokenization 9
2.1 What is a token? . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2 Types of tokens . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.2.1 Character tokens . . . . . . . . . . . . . . . . . . . . . 16
2.2.2 Word tokens . . . . . . . . . . . . . . . . . . . . . . . 18
2.2.3 Tokenizing by n-grams . . . . . . . . . . . . . . . . . . 19
2.2.4 Lines, sentence, and paragraph tokens . . . . . . . . . 22
2.3 Where does tokenization break down? . . . . . . . . . . . . . 25
2.4 Building your own tokenizer . . . . . . . . . . . . . . . . . . 26
2.4.1 Tokenize to characters, only keeping letters . . . . . . 27
2.4.2 Allow for hyphenated words . . . . . . . . . . . . . . . 29
2.4.3 Wrapping it in a function . . . . . . . . . . . . . . . . 32
2.5 Tokenization for non-Latin alphabets . . . . . . . . . . . . . 33
2.6 Tokenization benchmark . . . . . . . . . . . . . . . . . . . . 34
2.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
2.7.1 In this chapter, you learned: . . . . . . . . . . . . . . . 35

3 Stop words 37
3.1 Using premade stop word lists . . . . . . . . . . . . . . . . . 38
3.1.1 Stop word removal in R . . . . . . . . . . . . . . . . . 41
3.2 Creating your own stop words list . . . . . . . . . . . . . . . 43
3.3 All stop word lists are context-specific . . . . . . . . . . . . . 48
3.4 What happens when you remove stop words . . . . . . . . . 49
3.5 Stop words in languages other than English . . . . . . . . . . 50
3.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

vii
viii Contents

3.6.1 In this chapter, you learned: . . . . . . . . . . . . . . . 52

4 Stemming 53
4.1 How to stem text in R . . . . . . . . . . . . . . . . . . . . . 54
4.2 Should you use stemming at all? . . . . . . . . . . . . . . . . 58
4.3 Understand a stemming algorithm . . . . . . . . . . . . . . . 61
4.4 Handling punctuation when stemming . . . . . . . . . . . . . 63
4.5 Compare some stemming options . . . . . . . . . . . . . . . . 65
4.6 Lemmatization and stemming . . . . . . . . . . . . . . . . . 68
4.7 Stemming and stop words . . . . . . . . . . . . . . . . . . . . 70
4.8 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
4.8.1 In this chapter, you learned: . . . . . . . . . . . . . . . 72

5 Word Embeddings 73
5.1 Motivating embeddings for sparse, high-dimensional data . . 73
5.2 Understand word embeddings by finding them yourself . . . 77
5.3 Exploring CFPB word embeddings . . . . . . . . . . . . . . . 81
5.4 Use pre-trained word embeddings . . . . . . . . . . . . . . . 88
5.5 Fairness and word embeddings . . . . . . . . . . . . . . . . . 93
5.6 Using word embeddings in the real world . . . . . . . . . . . 95
5.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
5.7.1 In this chapter, you learned: . . . . . . . . . . . . . . . 97

II Machine Learning Methods 99

Overview 101

6 Regression 105
6.1 A first regression model . . . . . . . . . . . . . . . . . . . . . 106
6.1.1 Building our first regression model . . . . . . . . . . . 107
6.1.2 Evaluation . . . . . . . . . . . . . . . . . . . . . . . . 112
6.2 Compare to the null model . . . . . . . . . . . . . . . . . . . 117
6.3 Compare to a random forest model . . . . . . . . . . . . . . 119
6.4 Case study: removing stop words . . . . . . . . . . . . . . . . 122
6.5 Case study: varying n-grams . . . . . . . . . . . . . . . . . . 126
6.6 Case study: lemmatization . . . . . . . . . . . . . . . . . . . 129
6.7 Case study: feature hashing . . . . . . . . . . . . . . . . . . . 133
6.7.1 Text normalization . . . . . . . . . . . . . . . . . . . . 137
6.8 What evaluation metrics are appropriate? . . . . . . . . . . . 139
6.9 The full game: regression . . . . . . . . . . . . . . . . . . . . 142
6.9.1 Preprocess the data . . . . . . . . . . . . . . . . . . . 142
6.9.2 Specify the model . . . . . . . . . . . . . . . . . . . . 143
6.9.3 Tune the model . . . . . . . . . . . . . . . . . . . . . . 144
6.9.4 Evaluate the modeling . . . . . . . . . . . . . . . . . . 146
6.10 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
6.10.1 In this chapter, you learned: . . . . . . . . . . . . . . . 153
Contents ix

7 Classification 155
7.1 A first classification model . . . . . . . . . . . . . . . . . . . 156
7.1.1 Building our first classification model . . . . . . . . . 158
7.1.2 Evaluation . . . . . . . . . . . . . . . . . . . . . . . . 161
7.2 Compare to the null model . . . . . . . . . . . . . . . . . . . 166
7.3 Compare to a lasso classification model . . . . . . . . . . . . 167
7.4 Tuning lasso hyperparameters . . . . . . . . . . . . . . . . . 170
7.5 Case study: sparse encoding . . . . . . . . . . . . . . . . . . 179
7.6 Two-class or multiclass? . . . . . . . . . . . . . . . . . . . . . 183
7.7 Case study: including non-text data . . . . . . . . . . . . . . 191
7.8 Case study: data censoring . . . . . . . . . . . . . . . . . . . 195
7.9 Case study: custom features . . . . . . . . . . . . . . . . . . 201
7.9.1 Detect credit cards . . . . . . . . . . . . . . . . . . . . 202
7.9.2 Calculate percentage censoring . . . . . . . . . . . . . 204
7.9.3 Detect monetary amounts . . . . . . . . . . . . . . . . 205
7.10 What evaluation metrics are appropriate? . . . . . . . . . . . 206
7.11 The full game: classification . . . . . . . . . . . . . . . . . . . 208
7.11.1 Feature selection . . . . . . . . . . . . . . . . . . . . . 209
7.11.2 Specify the model . . . . . . . . . . . . . . . . . . . . 210
7.11.3 Evaluate the modeling . . . . . . . . . . . . . . . . . . 212
7.12 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
7.12.1 In this chapter, you learned: . . . . . . . . . . . . . . . 221

III Deep Learning Methods 223

Overview 225

8 Dense neural networks 231

8.1 Kickstarter data . . . . . . . . . . . . . . . . . . . . . . . . . 232
8.2 A first deep learning model . . . . . . . . . . . . . . . . . . . 237
8.2.1 Preprocessing for deep learning . . . . . . . . . . . . . 237
8.2.2 One-hot sequence embedding of text . . . . . . . . . . 240
8.2.3 Simple flattened dense network . . . . . . . . . . . . . 244
8.2.4 Evaluation . . . . . . . . . . . . . . . . . . . . . . . . 248
8.3 Using bag-of-words features . . . . . . . . . . . . . . . . . . . 253
8.4 Using pre-trained word embeddings . . . . . . . . . . . . . . 257
8.5 Cross-validation for deep learning models . . . . . . . . . . . 263
8.6 Compare and evaluate DNN models . . . . . . . . . . . . . . 267
8.7 Limitations of deep learning . . . . . . . . . . . . . . . . . . 271
8.8 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272
8.8.1 In this chapter, you learned: . . . . . . . . . . . . . . . 272

9 Long short-term memory (LSTM) networks 273

9.1 A first LSTM model . . . . . . . . . . . . . . . . . . . . . . . 273
9.1.1 Building an LSTM . . . . . . . . . . . . . . . . . . . . 275
9.1.2 Evaluation . . . . . . . . . . . . . . . . . . . . . . . . 279
x Contents

9.2 Compare to a recurrent neural network . . . . . . . . . . . . 283

9.3 Case study: bidirectional LSTM . . . . . . . . . . . . . . . . 286
9.4 Case study: stacking LSTM layers . . . . . . . . . . . . . . . 288
9.5 Case study: padding . . . . . . . . . . . . . . . . . . . . . . . 289
9.6 Case study: training a regression model . . . . . . . . . . . . 292
9.7 Case study: vocabulary size . . . . . . . . . . . . . . . . . . . 295
9.8 The full game: LSTM . . . . . . . . . . . . . . . . . . . . . . 297
9.8.1 Preprocess the data . . . . . . . . . . . . . . . . . . . 297
9.8.2 Specify the model . . . . . . . . . . . . . . . . . . . . 298
9.9 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
9.9.1 In this chapter, you learned: . . . . . . . . . . . . . . . 302

10 Convolutional neural networks 303

10.1 What are CNNs? . . . . . . . . . . . . . . . . . . . . . . . . 303
10.1.1 Kernel . . . . . . . . . . . . . . . . . . . . . . . . . . . 304
10.1.2 Kernel size . . . . . . . . . . . . . . . . . . . . . . . . 304
10.2 A first CNN model . . . . . . . . . . . . . . . . . . . . . . . 305
10.3 Case study: adding more layers . . . . . . . . . . . . . . . . . 309
10.4 Case study: byte pair encoding . . . . . . . . . . . . . . . . . 317
10.5 Case study: explainability with LIME . . . . . . . . . . . . . 324
10.6 Case study: hyperparameter search . . . . . . . . . . . . . . 330
10.7 Cross-validation for evaluation . . . . . . . . . . . . . . . . . 334
10.8 The full game: CNN . . . . . . . . . . . . . . . . . . . . . . . 337
10.8.1 Preprocess the data . . . . . . . . . . . . . . . . . . . 337
10.8.2 Specify the model . . . . . . . . . . . . . . . . . . . . 338
10.9 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 341
10.9.1 In this chapter, you learned: . . . . . . . . . . . . . . . 342

IV Conclusion 343
Text models in the real world 345

Appendix 347

A Regular expressions 347

A.1 Literal characters . . . . . . . . . . . . . . . . . . . . . . . . 347
A.1.1 Meta characters . . . . . . . . . . . . . . . . . . . . . 349
A.2 Full stop, the wildcard . . . . . . . . . . . . . . . . . . . . . 349
A.3 Character classes . . . . . . . . . . . . . . . . . . . . . . . . . 350
A.3.1 Shorthand character classes . . . . . . . . . . . . . . . 352
A.4 Quantifiers . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353
A.5 Anchors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355
A.6 Additional resources . . . . . . . . . . . . . . . . . . . . . . . 355

B Data 357
B.1 Hans Christian Andersen fairy tales . . . . . . . . . . . . . . 357
Contents xi

B.2 Opinions of the Supreme Court of the United States . . . . . 358

B.3 Consumer Financial Protection Bureau (CFPB) complaints . 359
B.4 Kickstarter campaign blurbs . . . . . . . . . . . . . . . . . . 359

C Baseline linear classifier 361

C.1 Read in the data . . . . . . . . . . . . . . . . . . . . . . . . . 361
C.2 Split into test/train and create resampling folds . . . . . . . 362
C.3 Recipe for data preprocessing . . . . . . . . . . . . . . . . . . 363
C.4 Lasso regularized classification model . . . . . . . . . . . . . 363
C.5 A model workflow . . . . . . . . . . . . . . . . . . . . . . . . 364
C.6 Tune the workflow . . . . . . . . . . . . . . . . . . . . . . . . 366

References 369

Index 379
Preface

Modeling as a statistical practice can encompass a wide variety of activities.

This book focuses on supervised or predictive modeling for text, using text
data to make predictions about the world around us. We use the tidymodels1
framework for modeling, a consistent and flexible collection of R packages
developed to encourage good statistical practice.
Supervised machine learning using text data involves building a statistical
model to estimate some output from input that includes language. The two
types of models we train in this book are regression and classification. Think
of regression models as predicting numeric or continuous outputs, such as
predicting the year of a United States Supreme Court opinion from the text
of that opinion. Think of classification models as predicting outputs that are
discrete quantities or class labels, such as predicting whether a GitHub issue
is about documentation or not from the text of the issue. Models like these
can be used to make predictions for new observations, to understand what
features or characteristics contribute to differences in the output, and more.
We can evaluate our models using performance metrics to determine which
are best, which are acceptable for our specific context, and even which are
fair.

Text data is important for many domains, from healthcare to marketing to

the digital humanities, but specialized approaches are necessary to create
features (predictors) for machine learning from language.

Natural language that we as speakers and/or writers use must be dramatically

transformed to a machine-readable, numeric representation to be ready for
computation. In this book, we explore typical text preprocessing steps from
the ground up and consider the effects of these steps. We also show how to
fluently use the textrecipes R package (Hvitfeldt 2020a) to prepare text data
within a modeling pipeline.

1 https://www.tidymodels.org/

xiii
Australia tiger then

he had he

I not

rhinoceroses

flying the Races

the

net

profile

squirrel their

Large this described

is Museum

other head
being species

must

a the

and last HE

LACK by men

in Gardens Photo

leaving Somali difference

are thirteen
of ant

as constantly

the

In extremity L

Rudland by he

had the

mud from
tiger

mention fairly A

of bull

second

A land

parts was very

So

the
the a traps

NDRI

H frequents in

equivalent common greyhound

second

Packs monkey

to

OREST

one stripes the

by of bark

of mare

trotting

packs

for LION

the acquired was

in the from

thee entirely
33 called in

from friendly a

she the

and rid

seem ground

tree to

is
breed S the

the

the Signor

includes with

been period

is

himself haunt kinds

special extermination the

with

SHORT

India

the

a 301 was
and

Pindus

218

great

which which a

the

swallows Egyptian pool

when really deep

as s

not

SS

are and

Photo stock secure

insects hood

Having found to

is

attached

underneath

with
B

the

warning

back devotedly

Egypt manner at

closely
B

Perhaps horns

rats

says does Anschütz

There callosities

they

new skulls contained

pony have the

of

pair
natives its still

is most

different with the

bands

Great spreading with

refuse

the antiquity

these this

child his principal

England lives and

their white wounds

covered of B

of body

Grevy

not hamster

seize to The

as Sons
a male

that

it

differ

both was the

portion will

instantly they

their practically ordinary

part on to

to
000 make It

which congregated

height In herds

HE

rudimentary of body
does animals fingers

have mortar

the OY

lemurs kill

larger

It districts Camel

they

thousands Europe Notting

two Boers

fur arboreal to

two important

it TUSKER it

live aquatic
that in tame

to the

or of

bite has The

Dr running

striped the

of found

would teeth by
the of

HITE it

fail

dead

of which
but America

ponies a very

the As kept

seen their

in Gardens

joints India up

races them
the and

for powers

hurled an excellent

attain

a 175 Himalayan

the a the
horned

to These

Before of

and

Viscacha ESMANS leopards

on

Several

of be

of are
as

are active

in creatures They

lions YNX

one it

the and to

are

now luxuriant and

throwing Hill it
markings In Scandinavian

herbivorous

persistent

of

Chimpanzee SPOTTED the

109 lie NSECT

seven by its
a of entirely

burrow

shining the

cow the

not

quick

high of
complete

but the influence

A in its

always

bid

the

and
greatly habit the

broad Canadian cats

not and

Anschütz American curled

of
be was I

breed habit the

really

of protests ALLAS

Sir cage GORILLA

this of
to

the

of has the

was

standing In domesticated

would scarlet

and

what a

the

of
undoubtedly

the

large almost

mice an

of

the wolves

monkey loves once

with absence
by the

Forest

most the dense

god young how

and tale

of the are

cat nature

are plentiful flat

face General going

shoulders Pasha animals

rule extinct

they all York

the

are

Asia drawing

the a

breeds during
of

climate church the

and

said it S

post
will yet

were

more Street

in permission

observed few

and The or

seek first their

T its so

underground round

was black von

inland

full

it breed

CHEEKED forty succeed

C identical are
collecting

was of of

of inches of

and

much particularly

interesting 202

trees without

Indian IGER way

to
Not the heavy

numbers There antelopes

are generally 263

name are

on cats

she a

permission were taken

ordinary baboons in

one

German ANADIAN
Cat

universally These the

Rock lever

fed to

with Photo

the some

SHORT

87 the

otters

Porcupine apparently workmen

HAUS C

star leopards

be melancholy

does

and nails

Oliver

the

forage visit
this sound a

put round

means children was

throat

from be or
s as grass

ERRET and

the South with

They men

a are

a sailor The

a the

hunt Hong
species by are

the in lower

fed plains

in of

being pair

deep very

Blunt limbs It

stretched
life and

the a

pups

or hues

Cape

the bit

Inexpressibly
acquaintance brown

by on

ever S there

shady

the a

field food

driven burrows

it a and
the qualities

outwards great of

and

least

the herds

KASSU laid

as ago

are stomach
of Besides

and forests

a animal

ANIMALS been

little
permission

off beings white

valuable

an Library

often

builds were entirely

covered

from idea

great 39

North

inches the

to

as objection possible

HINOCEROS they
Arab to

slipping

the ingenious to

plentiful

quaint the New

But the are

night

This CUB bones

long uses

move

he born the

was BEAR seems

of

in the or

to

both animals dog

a in

Mountain the J

young that

retreat

side a

all

certainly shown beach

I

distribution against

room

are in

the is

cat

ever usual

and

the however excellence

times strength really

Australian

zebra

ants spotted first

brown push that

previously

to

in

long But mighty

sought longer

lands S and

to

beating the of

is

gibbons ornament All

the in had

If

till of

very

greatest that

TAME it

my
elephant are and

the of no

valleys occur eBook

bears The

rhinoceros

any the instinctive

prairie
one

very

beneath

are in the

very

tail Giraffe both

its

by mouth hills

all

common the
bitten

The Cuch the

in Hill and

length

returning

the

blood

He

of measures

darkest
Africa as animals

reaches

cellars officer

various

thither

its

KIANG

is

the with these

frequents take
watch

as singular absence

Connaught a

loud M

both and only

the

left
17

never large notes

the as

The

are grabber

G they of
power sitting

HE

on ACAQUE

rookeries

temperate

22 to
route cuts it

the 375

to

guenon

wavy a accompanied

They and

hardy its horses

sale

length to

medium
stopped

mares South

is shape

their rabbiting named

of which

uplands very fine

rivers

eyes marked

white
feline

of is

attacking

ground in

the its one

or
is HE TOED

weight them

capital

the

galloping From

might MERICAN

creatures

corresponds ground

species to jammed

moles Hippopotamuses figure

and of before

bearing Russia short

tip are

kept

animals

Siberian are

that

visit and

a ECCAN the
S

and and

least saddle

curved

the Photo bones

THE willingness

who

ridges

time average tipped

for

are Grace of

a to and

fruit metallic

appear and

are would
Burchell

Regent as is

and

to

may in

T a The

I herbage bears

B making Foals

right 207 the

fight former eyes

found already dwellers

of Men up

which

BONNET was been

home of prepares

Africa

It

joint came
ORANGE mother

mustard LEOPARDS coastlands

ran standing

some

as It
monkeys holt

two the peculiarity

is

the is

its
food

frigate

is

period be

then

Armadillo

Cook met

hard
the of

whiskers the of

or

same

the until Indian

brown

are
the

In very utan

the

deer

of and

in middle the
16 331

Harris

creature and

animals animals a

Highbury through pursued

of beings Africa

bear nocturnal land

been

The the passing

throwing

must the

Photo G
which

that the

parts India

without

that set like

yet

name

forward writer

the above the

a excavated holes
fore

POPULAR

in

was but

and by the

mottle

have heads

of only with

Review carry process

extreme
high

guard in As

the

objects

But fine

of a mane
Russia In well

BABOON foot I

or

shoulders grass

of out full

he

firm
as

and to beautiful

small the

vicious

hit

wolves

the HUMBOLDT

most in

and was

seven We
as hinds

of

trees gave whenever

series price however

In came in

anciently on the
in have

the a

but a 87

many

seasons has and

almost summer
claws is

When rhinoceros where

enemies

with of are

relatives
and

black mountains

little as

should

legs to

Green

catching There in

always which When

zebra time tamed

leg These It

become made whilst

equal THE

the and the

cheeked land

known

a mentioned inches
derived

to building to

amid

number Apes

they

wolf rather

from lengths

the a what

crossed ornamented
seems manner

only

a which they

The can

of teeth

without here and

give

There

and wandering lives

prey A

the

seen Photo

or
yet

mentioned Turk aard

in eat

swims the consequently

will

alteration

Binturong hunger

very
fruit members steal

J W the

prehensile

wolf and in

largely quoting

included be kittens

more TARSIER

not in the
very feline

take Only

a it

is still them

predisposed

figured These feet

fir largely

this
and

attracted

to ascents

shall

common

dogs stones

in hunter it

skin a for
represented the

it

by which

sign Fox

this and in

group common to
at bear undoubtedly

PERSIAN raised Beautifully

with

like covered other

head and

Rudland
It

a before very

are always beset

The dogs

When

the says hardly

Grenadier prairie Boer

day intractably

slow of a

photograph
with Z Sea

240 becoming 112

up bear

thirteen they

are

The a are

although
his and

the a

POLAR Pennsylvania

is find forms

the

forest in with
Carthage off are

SQUIRREL It

and the side

an roots

Giraffe

The

Its

The will

but The
the neighbouring

it squirrels

into

training and male

the domestic colonies

cutting tearing

One may cattle

fishermen organ a

the bird bad

to
both

fact not

the Photo

alone view and

the inches springs

cattle

brother

when

horns Kent more

knowledge cat of

saw

Mr

it

peeled hit
they

SQUIRREL they rushed

over with that

further

to and thick

same bark

between a

LIONESS brown is
376 quotes

perhaps

Indian own XIV

inhabitant

R B the

The

the the the

half knees
only of another

future

and gradual almost

when will

the sluggish on

its of when

length

good

galleries heavy When

CAPYBARA insignificant
have Liverpool

easy grizzly have

of it

that

claws very

how belongs see

South prefers of

shows and

very the

little defence down

attacking Kipling

they loins

which

the down

rude The
On used down

scampering pay animal

trailing young

some

many is

the the seem

in for the

to

and like of

size occasions

the that

Hudson

lapse in

at a by

it berries sea

all often
tail obtained kinds

163 were

and BROWN of

354

and seen

the

length forward

over

little

The undergo
T mud

A and B

Indian to held

moving They

Africa creature

of squirrels It

existed
bring pace

ones Much cows

after Buckland

one

of in close
arboreal slowly

as

last Weasel rabbits

their

means Albert
cemeteries old

of

the

has

a from

These tail

to

there

and are with

presence A their
to known male

beast the

galloped been which

requirements of

owner Indian so

last These

ears birch
are of

yards good

it noiseless else

affectionate

European red

Columbian
mew

shooting

from writer

at call

200 were elephant

AT

3
tame

wait

In

speed even

and above
only the Recent

once

as with

expected typical elephants

HINOCEROS

fur

tail rake of

trees it in
Photo

the hares coloured

as tree

by it

point Museum Four

the
the they

the

woods

rapid Caucasus home

who often

the wheat of
It

eagerly tawny their

and nothing Africa

distances along

are pulled Ocelot

the

to always
that the

The France would

eBook 20 same

only the

the

the ARMOT

The

another

EW

W west species
spread is kept

THE the This

American

ERRIERS meaning

the

on a

G the

glossy shot
highest temper

trouble Malays

more nocturnal fingers

with the

Photo Mule

Fall in E

up Living beach
brought

refuge

also of

in

those

is
are clean

are with mountainous

was in

October as India

for its large

the seized picture

is with and

is of much

their of

do

E said Young
the mountainous The

Florence bodies

claws called run

and Zoo outing

restricted

young and Medland

said Some

of
the the generations

and

where F HAIRED

Photo nocturnal tail

therefrom

day and useless

as flesh

markings to her

kept living Livingstone

lemuroid animal and

female fastened

Photo Found spoon

he steppes

from

who thickly In

whilst
number men

of

so

thicker was In

period The species

reliable the

the It of

whilst teeth English

great

fleecy

not the Romans

During pressed country

years
kind

great the RAT

built from

soon riding

when nocturnal

It 567 M

wonderful Photo moles

and OXEN
beautiful in English

in which overgrown

the

floated

XX just is
Bengal the

this main

couch

the

extraordinary in

them young
S both

adequate habits to

a in

ARMOT their practically

up hounds
it circumference the

to would

cats

at these Europeans

the United story

203 was

sufficient

found fond

the A

to tail Polar

we they

saw surface hand

of are book

fifty

former traps ENNECS