01_evaluate_ner_spacy

A quick notebook to run a sample of texts through spacy, grab a set of data relating to named entity recognition, and report the results.

Conclusions? Ugh. Something is wrong here. Perhaps I need to get a bigger model for spacy. Maybe I should look at another package (Stanford's?). There may be a bug somewhere in the code below. But these results are very discouraging . . .

Where are my files?

In [8]:
!ls -1 /home/spenteco/0/corpora/muncie_public_library_corpus/PG_no_backmatter_fiction | wc -l
883

Load spacy, etc

In [9]:
import spacy

print spacy.__version__

nlp = spacy.load('en')

nlp.max_length = 2000000

PATH_TO_CORPUS = '/home/spenteco/0/corpora/muncie_public_library_corpus/PG_no_backmatter_fiction/'
2.0.11

Grab 10 files . . .

. . . at "random" for testing.

In [10]:
import random, glob

random.seed()

paths_to_files = random.sample(glob.glob(PATH_TO_CORPUS + '*.txt'), 10)

print paths_to_files
['/home/spenteco/0/corpora/muncie_public_library_corpus/PG_no_backmatter_fiction/Parker_Gilbert_The_Pomp_of_the_Lavilettes_Complete_PG_6217.txt', '/home/spenteco/0/corpora/muncie_public_library_corpus/PG_no_backmatter_fiction/Burnett_Frances_Hodgson_Sara_Crewe_Or_What_Happened_at_Miss_Minchin_s_PG_24772.txt', '/home/spenteco/0/corpora/muncie_public_library_corpus/PG_no_backmatter_fiction/Henty_G_A_George_Alfred_For_Name_and_Fame_Or_Through_Afghan_Passes_PG_21979.txt', '/home/spenteco/0/corpora/muncie_public_library_corpus/PG_no_backmatter_fiction/Munn_Charles_Clark_Uncle_Terry_A_Story_of_the_Maine_Coast_PG_28446.txt', '/home/spenteco/0/corpora/muncie_public_library_corpus/PG_no_backmatter_fiction/Cooper_James_Fenimore_The_Pilot_A_Tale_of_the_Sea_PG_7974.txt', '/home/spenteco/0/corpora/muncie_public_library_corpus/PG_no_backmatter_fiction/Eastwood_Frances_Geoffrey_the_Lollard_PG_48260_8.txt', '/home/spenteco/0/corpora/muncie_public_library_corpus/PG_no_backmatter_fiction/Yonge_Charlotte_M_Charlotte_Mary_The_Young_Step_Mother_Or_A_Chronicle_of_Mistakes_PG_5843.txt', '/home/spenteco/0/corpora/muncie_public_library_corpus/PG_no_backmatter_fiction/Von_Arnim_Elizabeth_The_Benefactress_PG_30302_8.txt', '/home/spenteco/0/corpora/muncie_public_library_corpus/PG_no_backmatter_fiction/Thackeray_William_Makepeace_Vanity_Fair_PG_599.txt', '/home/spenteco/0/corpora/muncie_public_library_corpus/PG_no_backmatter_fiction/Reade_Charles_Peg_Woffington_PG_3670.txt']

Run the texts through spacy

I'm grabbing three kinds of data here:

  • entity_types, which are counts of all of the kinds of named entities reported by spacy.
  • capitalized_not_named_entity, which includes any capitalized word which spacy did not recognize as a named entity.
  • named_entities, which are counts of all of the individual named entities reported by spacy, grouped by named entity type.
In [11]:
import codecs, re
from collections import defaultdict, Counter

entity_types = defaultdict(int)
capitalized_not_named_entity = defaultdict(int)

named_entities = defaultdict(lambda: defaultdict(int))

for pn, p in enumerate(paths_to_files):
        
    novel_label = p.split('/')[-1]
    
    print pn, novel_label
    
    text = re.sub('\s+', ' ', codecs.open(p, 'r', encoding='utf-8').read())
    
    doc = nlp(unicode(text))
    
    named_entities_in_text = []
    
    for t in doc:
        
        entity_types[t.ent_type_] += 1
        
        is_capitalized = False
        if t.text[0] == t.text[0].upper() and t.pos_ not in ['PUNCT', 'SPACE'] and t.tag_ not in ['POS']:
            is_capitalized = True
            
        if t.ent_type_ > '' and t.tag_ not in ['POS']:
            named_entities_in_text.append([t.ent_type_, t.ent_iob_, t.text])
        elif is_capitalized == True:
            capitalized_not_named_entity[t.text] += 1
            
    grouped_named_entities = []
    for a in named_entities_in_text:
        if a[1] == 'B':
            grouped_named_entities.append([a[0], [a[2]]])
        else:
            grouped_named_entities[-1][1].append(a[2])
            
    for a in grouped_named_entities:
        named_entities[a[0]][' '.join(a[1])] += 1
        
print 'Done!'
0 Parker_Gilbert_The_Pomp_of_the_Lavilettes_Complete_PG_6217.txt
1 Burnett_Frances_Hodgson_Sara_Crewe_Or_What_Happened_at_Miss_Minchin_s_PG_24772.txt
2 Henty_G_A_George_Alfred_For_Name_and_Fame_Or_Through_Afghan_Passes_PG_21979.txt
3 Munn_Charles_Clark_Uncle_Terry_A_Story_of_the_Maine_Coast_PG_28446.txt
4 Cooper_James_Fenimore_The_Pilot_A_Tale_of_the_Sea_PG_7974.txt
5 Eastwood_Frances_Geoffrey_the_Lollard_PG_48260_8.txt
6 Yonge_Charlotte_M_Charlotte_Mary_The_Young_Step_Mother_Or_A_Chronicle_of_Mistakes_PG_5843.txt
7 Von_Arnim_Elizabeth_The_Benefactress_PG_30302_8.txt
8 Thackeray_William_Makepeace_Vanity_Fair_PG_599.txt
9 Reade_Charles_Peg_Woffington_PG_3670.txt
Done!

What kinds of named entities does spacy recognize?

Note that the first line (" 1281250") effectively means "none" or "not a named entity."

In [12]:
print
for c in Counter(entity_types).most_common():
    print c[0], c[1]
 1281250
PERSON 29125
DATE 7815
ORG 7546
GPE 6228
CARDINAL 5433
TIME 5386
NORP 1897
WORK_OF_ART 1601
QUANTITY 1335
ORDINAL 1333
FAC 1148
LOC 1044
PRODUCT 568
MONEY 438
LAW 341
LANGUAGE 278
EVENT 170

False negatives?

Here, I list out the top 100 capitalized words which spacy did not identify as a named entity. I'm looking for instances in which spacy should have identified something, but didn't.

It does seem to miss some things: common name prefixes ("Mr.", "Mrs", "Miss"); titles ("Captain", "General", "Major"), and a few names ("Becky", "Triplet", "Maurice").

In [13]:
print
for c in Counter(capitalized_not_named_entity).most_common(100):
    print c[0], c[1]
I 13652
The 3975
He 2912
Mr. 2588
She 2318
It 2254
Mrs. 1606
And 1521
But 1404
You 1341
's 1186
There 836
What 833
Miss 829
Oh 726
A 723
When 702
They 699
No 649
If 646
In 603
Then 586
Sir 582
'll 552
This 519
We 505
That 502
As 489
Colonel 466
His 460
How 450
Do 424
Well 404
Yes 401
At 387
Becky 381
My 375
Lord 367
'm 364
Captain 350
Her 345
Will 339
Why 339
So 330
've 310
God 304
For 281
Now 272
Triplet 266
To 254
Ah 240
All 234
Maurice 234
Not 225
On 214
Is 207
'd 199
Major 197
After 186
Here 178
General 174
Let 173
With 166
Come 163
Who 155
Poor 154
By 147
Your 144
CHAPTER 138
're 136
Telly 134
Perhaps 134
Of 126
Did 123
Ay 118
These 114
Where 110
O 110
Have 109
Dusautoy 105
Are 103
Pilot 103
Jos 103
Had 98
Some 94
Sophy 93
'em 92
Madame 92
Indeed 91
Was 91
Good 88
Go 87
Never 87
Edmund 86
Even 84
Look 82
Only 80
From 80
Heaven 79
While 76

What did it find?

Here, I list the top 25 named entities for each type of named entity. Such lists should give us a broad overview of how well or poorly spacy does.

Bottom line? It's a mess.

In [17]:
for k in sorted(named_entities.keys()):
    print
    for c in Counter(named_entities[k]).most_common(25):
        print k, c[0], c[1]
CARDINAL one 1290
CARDINAL two 995
CARDINAL three 307
CARDINAL half 285
CARDINAL One 136
CARDINAL four 131
CARDINAL ten 52
CARDINAL five 52
CARDINAL six 48
CARDINAL twelve 46
CARDINAL Two 46
CARDINAL hundred 41
CARDINAL twenty 37
CARDINAL seven 29
CARDINAL a thousand 27
CARDINAL fifty 25
CARDINAL nine 24
CARDINAL a hundred 20
CARDINAL eight 20
CARDINAL quarter 18
CARDINAL more than one 16
CARDINAL a dozen 16
CARDINAL only one 14
CARDINAL some one 14
CARDINAL thousands 14

DATE the day 150
DATE morrow 127
DATE Sunday 89
DATE yesterday 75
DATE daily 66
DATE years 65
DATE every day 51
DATE that day 50
DATE one day 49
DATE Christmas 46
DATE the next day 43
DATE a week 42
DATE winter 37
DATE the days 36
DATE days 32
DATE day 29
DATE those days 28
DATE summer 28
DATE The next day 27
DATE many years 24
DATE all day 23
DATE May 22
DATE this day 22
DATE two years 21
DATE two days 21

EVENT Kleinwalde 9
EVENT the French Revolution 4
EVENT Thanksgiving 4
EVENT the Afghan war 3
EVENT the Battle of Prague 3
EVENT Aunt Lissy 3
EVENT Becky 2
EVENT Scinde Horse 2
EVENT the War Is Brought 2
EVENT The Fighting Round Cabul 2
EVENT The Battle Of Candahar 2
EVENT Foot , 2
EVENT Blanch 1
EVENT the 29th Bengal Native Infantry 1
EVENT the 22nd Pioneers 1
EVENT the Battle LV 1
EVENT Cupid 1
EVENT Easter Sunday 1
EVENT Exhibition 1
EVENT Stars 1
EVENT the Battle of Borodino 1
EVENT the Fortune of War 1
EVENT Great Scott 1
EVENT the Lent just 1
EVENT Tripod 1

FAC Russell Square 83
FAC Curzon Street 20
FAC Park Lane 14
FAC Great Gaunt Street 13
FAC Forest Tower 10
FAC the Peiwar - Khotal 10
FAC the Tower 9
FAC Peerage 7
FAC Mall 7
FAC Bar Harbor 6
FAC Hill Street 6
FAC Baker Street 6
FAC Gaunt House 6
FAC Bloomsbury Square 6
FAC Gaunt Square 5
FAC Cursitor Street 5
FAC Khotal 5
FAC Bond Street 5
FAC Gaunt Street 4
FAC Hanover Square 4
FAC Abbey 4
FAC Gillespie Street 4
FAC Forest Castle 3
FAC Sophy 3
FAC State Street 3

GPE Albinia 797
GPE Amelia 564
GPE Rebecca 474
GPE Barnstable 244
GPE London 222
GPE Gilbert 202
GPE England 163
GPE India 122
GPE Algernon 98
GPE Manual 91
GPE Cabul 68
GPE Brussels 62
GPE Paris 56
GPE Boston 55
GPE Brighton 54
GPE Karlchen 45
GPE Captain 42
GPE Bertrand 40
GPE Yossouf 38
GPE Bayford 37
GPE Glorvina 37
GPE America 35
GPE France 33
GPE Sophie 31
GPE Germany 31

LANGUAGE English 215
LANGUAGE French 24
LANGUAGE Latin 17
LANGUAGE Hebrew 6
LANGUAGE Candahar 3
LANGUAGE Beauty 2
LANGUAGE Whitefriars 2
LANGUAGE Commoner 2
LANGUAGE Spanish 2
LANGUAGE Gee 1
LANGUAGE Afghan 1
LANGUAGE Ameer 1
LANGUAGE Bath 1
LANGUAGE Yankee 1

LAW CHAPTER XVII 6
LAW CHAPTER XV 4
LAW CHAPTER XXXI 4
LAW CHAPTER XXV 3
LAW CHAPTER XXIV 3
LAW CHAPTER XX 3
LAW CHAPTER XVIII 2
LAW Chapter 9 : 2
LAW Chapter 21 2
LAW Chapter 20 2
LAW Chapter 22 2
LAW CHAPTER XIV 2
LAW CHAPTER XII 2
LAW CHAPTER XXI 2
LAW Chapter 18 2
LAW Chapter 10 2
LAW Chapter 11 2
LAW Chapter 12 2
LAW Chapter 13 2
LAW Chapter 15 2
LAW Chapter 16 2
LAW Chapter 17 2
LAW Chapter 19 2
LAW the Royal Horse Artillery 2
LAW CHAPTER XXIX 2

LOC Sophy 78
LOC Europe 55
LOC Cape 33
LOC Yossouf 16
LOC Candahar 15
LOC Captain 13
LOC City 13
LOC earth 10
LOC Ernest 10
LOC the Sea Belle 10
LOC Parsee 10
LOC the Manor Casimbault 9
LOC Park 9
LOC Atlantic 9
LOC Khurum 8
LOC East 8
LOC Mudbury 7
LOC the moon 7
LOC Euphrates 7
LOC the West Indies 6
LOC Venus 6
LOC Hyde Park 6
LOC Glenmalony 6
LOC Southport Island 6
LOC the German Ocean 6

MONEY a penny 11
MONEY five thousand dollars 7
MONEY two thousand pounds 6
MONEY the five thousand dollars 6
MONEY ten thousand dollars 4
MONEY a cent 3
MONEY five - dollar 3
MONEY one penny 3
MONEY five thousand pounds 3
MONEY five dollars 3
MONEY two hundred dollars 3
MONEY five hundred dollars 2
MONEY one hundred dollars 2
MONEY thirty thousand pounds 2
MONEY two thousand dollars 2
MONEY forty thousand marks 2
MONEY four hundred dollars 2
MONEY half - penny 2
MONEY one dollar 2
MONEY two hundred thousand marks 2
MONEY sixty thousand dollars 2
MONEY five - franc 2
MONEY every penny 2
MONEY seventy thousand pounds 2
MONEY four - penny 2

NORP French 195
NORP British 190
NORP German 131
NORP Afghans 126
NORP English 96
NORP Afghan 80
NORP Irish 70
NORP Indian 59
NORP Leech 55
NORP Christian 50
NORP American 30
NORP Candahar 24
NORP Hans 24
NORP Irishman 21
NORP Kendals 17
NORP Sikhs 16
NORP Latin 16
NORP Malay 16
NORP Italian 15
NORP European 14
NORP Greek 14
NORP Dutch 14
NORP Russian 13
NORP Russians 12
NORP Ayoub 12

ORDINAL first 909
ORDINAL second 139
ORDINAL third 62
ORDINAL First 30
ORDINAL 5th 15
ORDINAL 66th 15
ORDINAL fourth 14
ORDINAL 3rd 12
ORDINAL 8th 9
ORDINAL 2nd 8
ORDINAL eighth 7
ORDINAL fifth 6
ORDINAL sixth 6
ORDINAL 4th 5
ORDINAL last 5
ORDINAL 150th 5
ORDINAL secondly 4
ORDINAL 72nd 4
ORDINAL once 4
ORDINAL 6th 4
ORDINAL Second 4
ORDINAL 9th 4
ORDINAL 10th 3
ORDINAL tenth 3
ORDINAL seemeth 3

ORG Sophy 514
ORG Axel 205
ORG Anna 142
ORG Georgy 114
ORG Vanity Fair 70
ORG Trudi 68
ORG Queen Crawley 58
ORG Ariel 54
ORG Hilton 53
ORG Mamma 49
ORG Ripon 49
ORG Waterloo 37
ORG Sandgate 33
ORG Crawley 31
ORG Jos 31
ORG Manske 31
ORG Ferrars 30
ORG Frau Dellwig 29
ORG Court 29
ORG Sedley 28
ORG Sharp 27
ORG Miss Estcourt 25
ORG Woffington 25
ORG Fritz 25
ORG Leech 23

PERSON Kendal 805
PERSON Anna 683
PERSON Albinia 554
PERSON George 548
PERSON Lucy 533
PERSON Dobbin 466
PERSON Gilbert 461
PERSON Griffith 447
PERSON Osborne 445
PERSON Rawdon 415
PERSON Crawley 342
PERSON Albert 325
PERSON Pitt 323
PERSON Alice 287
PERSON Maurice 287
PERSON Vane 267
PERSON Sedley 241
PERSON Ulick 239
PERSON Susie 238
PERSON Jos 236
PERSON Miss Crawley 222
PERSON Genevieve 215
PERSON Frank 205
PERSON Letty 197
PERSON Emmy 186

PRODUCT Captain 40
PRODUCT Anna 16
PRODUCT Captain Ripon 16
PRODUCT Baby 13
PRODUCT City 11
PRODUCT Drurys 10
PRODUCT Captain Borroughcliffe 9
PRODUCT Aunt Lissy 9
PRODUCT Winchester 8
PRODUCT Nabob 7
PRODUCT Duchess 6
PRODUCT Captain Manual 6
PRODUCT Captain Barnstable 6
PRODUCT Mabel 6
PRODUCT Crawleys 5
PRODUCT Osbornes 5
PRODUCT Calcutta 5
PRODUCT Scutari 4
PRODUCT Captain Edwards 4
PRODUCT Elmreichs 4
PRODUCT Merry 4
PRODUCT Cibber 3
PRODUCT Yossouf 3
PRODUCT Captain Fletcher 3
PRODUCT Madame de Borodino 3

QUANTITY a mile 17
QUANTITY half a mile 15
QUANTITY a few yards 12
QUANTITY four miles 11
QUANTITY two miles 10
QUANTITY a few feet 9
QUANTITY fifty pounds 7
QUANTITY twenty pounds 7
QUANTITY a hundred pounds 6
QUANTITY two 6
QUANTITY a few miles 6
QUANTITY six feet 5
QUANTITY ten miles 5
QUANTITY three miles 5
QUANTITY five hundred pounds 5
QUANTITY fifteen miles 4
QUANTITY a few hundred yards 4
QUANTITY ten pounds 4
QUANTITY a mile and a half 4
QUANTITY a thousand pounds 4
QUANTITY an inch 4
QUANTITY a hundred yards 4
QUANTITY twenty pound 4
QUANTITY six miles 4
QUANTITY three - eighths 3

TIME night 116
TIME morning 107
TIME the night 72
TIME last night 58
TIME the morning 57
TIME this morning 51
TIME evening 49
TIME a few minutes 47
TIME an hour 46
TIME the next morning 36
TIME the evening 34
TIME hours 30
TIME the hour 28
TIME that night 26
TIME half an hour 26
TIME afternoon 25
TIME a minute 22
TIME ten minutes 22
TIME this evening 21
TIME hour 20
TIME two hours 20
TIME all night 18
TIME every night 18
TIME midnight 17
TIME that evening 17

WORK_OF_ART Bible 36
WORK_OF_ART Love 17
WORK_OF_ART the Misses Osborne 5
WORK_OF_ART Mabel Vane 5
WORK_OF_ART Master Coffin 4
WORK_OF_ART ME 4
WORK_OF_ART Mabel 4
WORK_OF_ART the Tape and Sealing Wax Office 3
WORK_OF_ART The Girl I Left Behind Me 3
WORK_OF_ART Trudi 3
WORK_OF_ART Dear Madam 3
WORK_OF_ART Order 3
WORK_OF_ART Captain Ripon 3
WORK_OF_ART the Misses Dobbin 2
WORK_OF_ART Manske 2
WORK_OF_ART Cibber 2
WORK_OF_ART Humph 2
WORK_OF_ART The Last Rose of Summer 2
WORK_OF_ART Which Jos Takes Flight 2
WORK_OF_ART Hold the Fort 2
WORK_OF_ART Lascar 2
WORK_OF_ART Frau Manske 2
WORK_OF_ART Wapping Old Stairs 2
WORK_OF_ART & 2
WORK_OF_ART the Lovely Rose 2

List the place names found by spacy

I list every place name (which seem to fall into two types, "GPE" and "LOC") which occurs 2 or more times the in sampled data.

I'm not happy with the results.

In [18]:
for k in ['GPE', 'LOC']:
    print
    for c in Counter(named_entities[k]).most_common():
        if c[1] == 1:
            break
        print k, c[0], c[1]
GPE Albinia 797
GPE Amelia 564
GPE Rebecca 474
GPE Barnstable 244
GPE London 222
GPE Gilbert 202
GPE England 163
GPE India 122
GPE Algernon 98
GPE Manual 91
GPE Cabul 68
GPE Brussels 62
GPE Paris 56
GPE Boston 55
GPE Brighton 54
GPE Karlchen 45
GPE Captain 42
GPE Bertrand 40
GPE Yossouf 38
GPE Bayford 37
GPE Glorvina 37
GPE America 35
GPE France 33
GPE Sophie 31
GPE Germany 31
GPE St. Ruth 30
GPE Cobham 28
GPE Genevieve 25
GPE Hampshire 24
GPE Pitt 23
GPE Quebec 23
GPE O'More 22
GPE Providence 22
GPE Castine 22
GPE Brompton 20
GPE Rome 20
GPE Lollards 20
GPE Canada 19
GPE Berlin 19
GPE Raggles 19
GPE Vicar 19
GPE Gibbie 18
GPE Estcourt 18
GPE Candahar 18
GPE Mudbury 18
GPE Pilot 17
GPE Afghanistan 17
GPE Macmurdo 17
GPE Ireland 17
GPE Maurice 16
GPE Thou 15
GPE Maine 15
GPE Maria 15
GPE Chiswick 15
GPE Stralsund 15
GPE Russia 14
GPE Madras 13
GPE Katherine 13
GPE Lollard 13
GPE Belgium 13
GPE Englishmen 13
GPE Italy 13
GPE Oxford 13
GPE Kitty 13
GPE Khuram 13
GPE Maiwand 12
GPE New York 12
GPE Thar 12
GPE Peggy 12
GPE Wellington 12
GPE colonies 12
GPE Mayhew 12
GPE Bibi 11
GPE Shropshire 11
GPE Malta 11
GPE Montreal 10
GPE the City 10
GPE Wales 10
GPE China 10
GPE Cambridge 10
GPE Naples 10
GPE Britain 10
GPE Richmond 9
GPE Spain 9
GPE Crawley 9
GPE Dublin 9
GPE Washington 9
GPE Southdown 9
GPE Sandgate 9
GPE Boulogne 9
GPE Bengal 8
GPE Holland 8
GPE Phoebe 8
GPE Stockholm 8
GPE Boltrope 8
GPE Turrai 8
GPE St. George 7
GPE Opera 7
GPE C.B. 7
GPE Devonshire 7
GPE Sambo 7
GPE Dellwig 7
GPE Southampton 7
GPE the West Indies 6
GPE Sophia 6
GPE Bah 6
GPE china 6
GPE St. Paul 6
GPE Borroughcliffe 6
GPE Jungbluth 6
GPE Calais 5
GPE Hazir 5
GPE Frye 5
GPE Trudi 5
GPE Ayoub 5
GPE Iphigenia 5
GPE P.S. 5
GPE XI 5
GPE Merry 5
GPE Gray 5
GPE Deacon Oaks 5
GPE Lascar 5
GPE Kuhräuber 5
GPE Dellwigs 5
GPE Vienna 5
GPE Edinburgh 5
GPE Pomerania 5
GPE Becky 5
GPE Scotland 5
GPE Jugdulluk 5
GPE Vane 5
GPE Kleinwalde 5
GPE Ghent 5
GPE Liverpool 4
GPE Griffith 4
GPE Bloomsbury 4
GPE Kaffir 4
GPE Greece 4
GPE Fanny 4
GPE Jos 4
GPE Brazil 4
GPE XV 4
GPE Tory 4
GPE States 4
GPE Thull 4
GPE Populace 4
GPE Ramchunder 4
GPE Dover 4
GPE Schloss 4
GPE Bombay 4
GPE Wolfe 4
GPE Park 4
GPE Southport Island 4
GPE Chopper 4
GPE Bristol 4
GPE Dumdum 4
GPE Southport 4
GPE Florence 4
GPE Leipzig 4
GPE Bar Harbor 3
GPE Turkey 3
GPE kingdom 3
GPE Cheshire 3
GPE Strasburg 3
GPE Supper 3
GPE Jena 3
GPE Islington 3
GPE Isidor 3
GPE Regent 3
GPE Welsh 3
GPE Hanover 3
GPE the City of London 3
GPE Sappir 3
GPE St. Kitt 3
GPE Logan 3
GPE Dobbs 3
GPE Begleitung 3
GPE Shangois 3
GPE Lizzie 3
GPE Khost 3
GPE Sophy 3
GPE Savannah 3
GPE Great Britain 3
GPE Lambeth 3
GPE Serpentine 3
GPE Griff 3
GPE Danube 3
GPE Regulus 3
GPE Carolinas 3
GPE St. Leocadia 3
GPE Hants 3
GPE Binkie 3
GPE York 3
GPE V.C. 3
GPE Letty 3
GPE Chicago 3
GPE Gundamuk 3
GPE City 3
GPE Athens 3
GPE Bon'venture 3
GPE Briton 3
GPE Coffin 3
GPE Fuddleston 3
GPE Duchy 3
GPE Khelat 3
GPE Jamaica 3
GPE Whom 3
GPE Sweden 3
GPE Kohat 3
GPE Hymen 3
GPE Kadenstein 3
GPE Plymouth 3
GPE Madeira 3
GPE Saint 2
GPE Batavier 2
GPE Rector 2
GPE Breakfast 2
GPE Norway 2
GPE H.H. 2
GPE Bethlehem 2
GPE Miladi 2
GPE Labrador 2
GPE Molloys 2
GPE Devil 2
GPE Thespis 2
GPE Kandidat 2
GPE Champignac 2
GPE Albion 2
GPE Vittoria 2
GPE Birmingham 2
GPE Telly 2
GPE Baye 2
GPE gallop 2
GPE Upper Canada 2
GPE Burdock 2
GPE Dresden 2
GPE Chesapeake 2
GPE Cribb 2
GPE Girishk 2
GPE Honourable 2
GPE Mayfair 2
GPE Beck 2
GPE Bengalee 2
GPE St. James Street 2
GPE St. Martin Lane 2
GPE Janissary 2
GPE Christendom 2
GPE Tarquin 2
GPE Chatham 2
GPE Malony 2
GPE Tigearnach 2
GPE Hindoos 2
GPE Eutropius 2
GPE Salsette 2
GPE Quarrel 2
GPE Mademoiselle 2
GPE St. James 2
GPE Nantucket 2
GPE Sindbad 2
GPE Plowden 2
GPE Marry 2
GPE Cleland 2
GPE Reverence 2
GPE St. Helena 2
GPE Rollo 2
GPE Turis 2
GPE Bedlam 2
GPE Loo 2
GPE Willie 2
GPE Prisoner 2
GPE Beg 2
GPE Natur 2
GPE New Orleans 2
GPE Venice 2
GPE LONDON 2
GPE Easter 2
GPE O'Dowdstown 2
GPE Londoner 2
GPE Shatur - Gardan 2
GPE the Seven Sorrows 2
GPE Polonia 2
GPE Neapolitan 2
GPE General 2
GPE Laureate 2
GPE Engländerin 2
GPE Yorkshire 2
GPE King Louis 2
GPE St. John 2
GPE Connecticut 2
GPE Trafalgar 2
GPE Miss. 2
GPE Melia 2
GPE Solomon 2
GPE Cologne 2
GPE Bormola 2
GPE Mandy 2
GPE Herrschaften 2
GPE Singapore 2
GPE Moscow 2
GPE Babylon 2
GPE Strand 2
GPE Shetland 2
GPE Protestant 2
GPE Frederick 2
GPE Newfoundland 2
GPE Quettah 2
GPE St. Giles 2
GPE St. Petersburg 2
GPE Dieu 2
GPE Hand 2
GPE Letter 2
GPE Bryan 2
GPE Switzerland 2
GPE Helena 2
GPE Jingo 2
GPE St. Kitts 2
GPE BOSTON 2
GPE Knightsbridge 2
GPE Versailles 2
GPE Malonys 2
GPE Recognized 2
GPE Burjoice 2
GPE Rectory 2
GPE Exploded 2
GPE Empire 2
GPE Geography 2

LOC Sophy 78
LOC Europe 55
LOC Cape 33
LOC Yossouf 16
LOC Candahar 15
LOC Captain 13
LOC City 13
LOC earth 10
LOC Ernest 10
LOC the Sea Belle 10
LOC Parsee 10
LOC the Manor Casimbault 9
LOC Park 9
LOC Atlantic 9
LOC Khurum 8
LOC East 8
LOC Mudbury 7
LOC the moon 7
LOC Euphrates 7
LOC the West Indies 6
LOC Venus 6
LOC Hyde Park 6
LOC Glenmalony 6
LOC Southport Island 6
LOC the German Ocean 6
LOC Coventry Island 5
LOC the North Sea 5
LOC Englishmen 4
LOC Carolina 4
LOC the West End 4
LOC St. Ruth 4
LOC Continent 4
LOC Kensington Gardens 3
LOC Old England 3
LOC the Khyber Pass 3
LOC Jupiter 3
LOC Boston Bay 3
LOC Peggy 3
LOC Peninsula 3
LOC Ghazis 3
LOC Beverly 3
LOC Mars 3
LOC the Prince Regent 3
LOC the Ramchunder East Indiaman 2
LOC the Atlantic Ocean 2
LOC Hudson 2
LOC the Large Family 2
LOC East India 2
LOC West India 2
LOC Soho 2
LOC Sailor 2
LOC Talloon 2
LOC Silenus 2
LOC Regent 2
LOC the Arabian Nights 2
LOC Neptune 2
LOC Isles 2
LOC the Low Countries 2
LOC Mount St. John 2
LOC Tuileries 2
LOC CHAPTER XI 2
LOC Camelot 2
LOC Africa 2
LOC the Valley of Diamonds 2
LOC Brest 2
LOC Ludgate Hill 2
LOC the Eastern Archipelago 2
LOC Easter 2
LOC Helmund 2
LOC the Pir - Paimal Hill 2
LOC the Gardens 2
LOC the Ship Inn 2
LOC the New Testament 2
LOC Bute 2
LOC the Lower Wharf 2
LOC North Sea 2