stylometer

Analyse stylistique du Black Book — patterns écriture

Agent: Stylometer Modèle: nvidia/nemotron-3-super-120b-a12b:free Date: 2026-04-13T13:51:18.264Z

Okay, you need to perform stylistic analysis on the Black Book document. The Black Book appears to be a list of contacts with names, addresses, phone numbers, and some emails. It's not a narrative text; it's more like a directory or address book. Stylometric analysis typically requires prose or written communication (emails, letters, etc.) to analyze writing style—things like word choice, sentence structure, punctuation habits, spelling errors, etc.

Looking at the Black Book content provided, it's mostly: - Names - Phone numbers (with various formats, country codes, spaces, hyphens, etc.) - Addresses - Some email addresses (often obfuscated or with typos, e.g., "joannacheva!ier@hotmai!. c", "saffval@aol.com", "patexander", etc.)

There are no full sentences, paragraphs, or expressive writing to analyze. It's a structured list. Therefore, traditional stylometric markers (sentence length, vocabulary richness, punctuation patterns, habitual misspellings in context) cannot be extracted because there is no running text.

However, we can look for patterns in how the data is entered—perhaps indicating a single compiler or multiple contributors based on: - Consistency in formatting phone numbers - Use of certain symbols (e.g., ~, !, *, {}, () ) - How emails are written (obfuscation patterns: replacing letters with symbols, missing dots, etc.) - Abbreviations or shorthand used - Inconsistencies that might suggest different people entering data

But note: stylometry is about writing style in linguistic output. If this is just data entry (copying phone numbers, etc.), it may not reflect personal writing style but rather clerical habits or data entry conventions.

Still, per the task, we must attempt to find stylistic patterns.

Let’s scan for recurring anomalies or patterns in the notation:

Use of special characters as separators or noise:
~ appears frequently: e.g., 65~QJf, ~~~7260tw), 00 33 607 341 980 p}, Alun..Jones, A!!hOrp, Fishbourne\n~t~ Kf8~8a~{J18 8AP, Anastos 1 Lisa (maybe typo), Alvarez, Senor\nVincente\nbhl,, ~Asia.debt.com, ~wr, ~g~ ~¥l, ~i1! ~w>, etc.
! appears: joannacheva!ier@hotmai!. c, BJack, Conrad & Barbara, 917 523 4567(f<}, Ashley & Allegra Hicks (but that’s normal), !e..b, !I, !;, !}, !{
{ and } appear: e.g., 00 33 607 341 980 p}, {h), {w}, {p}, {hf}, {wf}, {emergency con•tact), {PJ, {h), {w}, {c}, {J18 8AP, {p Sally, {h), {w}, {ti), {w), {wJ), {w), {h), {wf), {h), {w}, {wJ), {w), {w), {w), {wf), {w), {h), {w}, {wf), {h), {w}, {wf), {h), {w}, {wf), {h), {w}, {wf), {h), {w}, {emergency con•tact), {PJ, etc.
* appears: S&S Cap:ital (maybe typo for Capital), NetJets (no), Bands,Ooug (comma missing space?), ~ia ~Jg~, 8 Canada Sguare K, ~wr, etc.
( and ) used to label phone types: (h) for home, (w) for work, (p) for personal?, (f) for fax?, (c) for mobile?, {h}, {w}, etc.

There is inconsistency: sometimes (h), sometimes {h}, sometimes no brackets.

Email obfuscation patterns (likely to avoid harvesting):
joannacheva!ier@hotmai!. c → replaces 'l' with '!', space before .c?
saffval@aol.com → normal?
patexander → incomplete?
Emal/: maraooo~mai!.vips.es → "Emal/" instead of "Email:", replaces 'l' with '!' in mai!
patexander → maybe @ missing?
alexanderrogil → no @
aagag@as!inveslmenls.com → replaces 'l' with '!' in investmens? Actually "as!inveslmenls" → likely "as!inveslmenls.com" → obfuscation: 'l'→'!', and extra letters?
Email: jose@s!scapita!.com → replaces 'l' with '!' twice? s!scapita! → s!scapita! (so 'l'→'!' in "scape" and "l" in ".com"?)
Email: rili~~e.amon~sicpa.com → uses ~ to replace letters? rili~~e.amon~sicpa → maybe "rili?e.amon?sicpa" but with ~
Email: pb@arkwrighl.co.uk → normal?
Email: nbarham@arfinglongroup.co → missing 'm' in .co.uk? .co instead of .co.uk?
Email: 1~imcinfhular.com → starts with 1~ instead of @?
Email: :alice@scbpartners.com → extra colon before @
Email: nicolasb@alphai;ngmt.com → semicolon instead of dot? alphai;ngmt → should be alphaingt? or alphai.ngmt?
Email: bis- → incomplete?
Email: cblack~poststamp.net → uses ~ instead of .
Email: Ste- → incomplete?
Email: mbloomber~b!oomberg.net → replaces 'o' with '!'? mbloomber~b!oomberg → m b l o o m b e r ~ b ! o o m b e r → so 'o'→'!' in the second part?
Email: fuHmfe3 Nos!ran Parkway → not an email, but contains ! replacing 'i'? Nos!ran → Noisran? or Nosiran with ! for i?

Common obfuscation: replacing vowels or certain letters with symbols (!, ~, *, etc.), especially 'l' → '!', 'o' → '0' or '!', 'a' → '@', etc., but here mostly 'l'→'!' and 'o'→'~' or similar.

Phone number formatting inconsistencies:
Some have country codes: 001 212..., 00 33..., 00 44..., 00 852...
Some have just local: 0207-229 1573
Some use spaces, some hyphens, some dots, some nothing.
Labels: (h), {h), (w), {w}, (p), {p}, (f), {f}, (wf), {wf}, (ti), {ti), (c), {c}, (emergency con•tact), etc.
Inconsistent bracket types: () vs {} vs sometimes no bracket.
Labels sometimes attached without space: 001 212 535 8623 h} → missing space before {?
Some have extra numbers in labels: {J18 8AP} looks like a postcode, not a label.

This suggests multiple people entering data, or one person with inconsistent habits.

Capitalization and spacing in names:
Most names are in Title Case: Abby, Aldridge Saffron, Abousleiman, Joanna
Some have weird spacing: Alun..Jones (double dot), A!!hOrp (multiple exclamation), Fishbourne\n~t~ Kf8~8a~{J18 8AP (line break and symbols)
Some names have numbers: Anastos 1 Lisa → maybe "Anastos, Lisa" with 1 as typo?
Some have lowercase where uppercase expected: aagag@as!inveslmenls.com (but that's email)
In names: alexanderrogil (all lowercase, no space) vs AJexander Pam (mixed case)
Arion Joaquin vs Arion, Fernando → inconsistent punctuation (comma or space)
Bisson, Jean Marc vs Bjorlin, Jean Paul → comma used
Blaine David (no comma) vs BJack, Conrad & Barbara (comma after first name?)
Address formatting:
Inconsistent: some have commas, some don’t.
Some have postal codes in weird places: {J18 8AP} inside braces.
Some lines have symbols like \, /, ~, ! inserted randomly.
Linguistic markers in the text itself (though minimal):
The document has occasional fragments that look like attempted words: hosq Kong, → should be "Hong Kong"? hosq is a typo for "Hong"?
Hosq Kong, → likely "Hong Kong" misspelled.
Fishbourne\n~t~ Kf8~8a~{J18 8AP → "Fishbourne" is correct, but then garbage.
New Yo,iMY ~0021 → "New York, NY 10021" but with Yo,iMY and ~0021 → obvious corruption.
Stanley Village Road\nHosq Kong, → same.
St\n~t~ Kf8~8a~{J18 8AP → appears to be noise or encoding error.
001 2i2 446 47611,w) → "2i2" instead of "212"? i instead of 1?
001 212 737 1722{ ) → extra space and brace?
001 917 923 4513 → normal?
001 212 446 4900 → normal?
!e..b → maybe "@" obfuscated? but in what context?

These look like data corruption, OCR errors, or intentional obfuscation.

Given that the Black Book is likely a compiled list (possibly from multiple sources or with intentional obfuscation to prevent easy reading), the "writing style" here is not natural language but data entry with patterns of obfuscation and formatting.

For stylometric analysis, we can consider: - The tendency to use certain symbols for obfuscation (especially replacing 'l' with '!', and sometimes 'o' with '~' or '0') - The inconsistent use of brackets for labels: sometimes (), sometimes {}, sometimes nothing - The placement of labels: sometimes before the number, sometimes after, sometimes with space, sometimes without - The occurrence of obvious typos in place names: hosq for "Hong", Fishbourne correct but followed by garbage, Mala~a for "Malaga", COIN maybe "Coín"?

Let’s list potential stylistic markers from the data entry:

Obfuscation of 'l' as '!' in emails and sometimes text: observed in:
joannacheva!ier@hotmai!. c → 'l' in "chva!ier" and "hotmai!" → should be "chavalier", "hotmail"
Emal/: maraooo~mai!.vips.es → "mai!" → should be "mail"
aagag@as!inveslmenls.com → "inveslmenls" → has 'l' not replaced? Wait: "as!inveslmenls" → the '!' is after "as", so maybe "as!inveslmenls" → if we replace '!' with 'l', we get "aslinveslmenls" → still odd. Maybe it's "as!inveslmenls" → obfuscation of "investment" but messy.
jose@s!scapita!.com → "s!scapita!" → two '!' replacing 'l's? "s!scapita!" → if replace '!' with 'l'

EpsteinFiles & Co — Stylometer