NUMB3RS - Season 3 - "Killer Chat" - Stylometry

Topic: Descriptive Statistics (Box-and-Whisker Plots)Algebra 2 / Pre-Calculus / Statistics

Linguistic FingerprintingStep 1 of 5

The Core of Stylometric Variation

Why do stylometrists look at entire distributions (like box-and-whisker plots) instead of just comparing simple averages?

Imagine two authors: Author X writes sentences that are consistently 14 to 16 words long. Author Y alternates between very short 2-word sentences and long, sweeping 28-word sentences. Both authors have a mean sentence length of exactly 15 words!

If we only look at the average, we would think their styles are identical. By using box-and-whisker plots, we can visually map the spread, symmetry, and dispersion of the writing samples to catch these stylistic fingerprints.

Forensic Stylometric Distribution

Explore the sentence length (WPS) ranges of authors using box-and-whisker plots.

Interactive
0102030405060Julia (Sample A)Aaron (Sample B)Anonymous (Sample C)
Interactive WPS Probe:15 wordsSlide to analyze cumulative densities
Julia Proportion0.53sentences ≤ 15 words
Aaron Proportion0.14sentences ≤ 15 words
Anon Proportion0.75sentences ≤ 15 words

Socratic BakerBot

Classroom Forensic Assistant

BakerBot

Hi! I'm BakerBot, your math & stylometry assistant. Let's analyze the linguistic fingerprints of suspects. How can I help you today?