Forensic Stylometry
Definition
The statistical analysis of textual features : function-word frequencies, sentence length distributions, punctuation patterns : to identify or compare authors. Reliable on texts of several hundred words; degrades sharply on very short samples.
- Features analysed
- Function-word frequencies, sentence length, punctuation patterns
- Purpose
- Identify or compare authors of text
- Minimum reliability threshold
- Several hundred words
- Weakness
- Degrades sharply on very short samples
Common questions
Why does stylometry struggle with short texts like a single text message?+
Function-word frequencies and other statistical style markers need enough text to produce a stable pattern, so a message of only a few words gives too little data for the statistics to distinguish one author's style reliably from another's.
How is stylometry used differently from content-based authorship analysis?+
Stylometry looks at structural and unconscious writing habits, such as how often 'the' or commas appear, rather than the meaning or topic of the text, which makes it harder for an author to deliberately disguise than simply changing subject matter or tone.
Related terms
- Chain of Custody
- The documented chronological record of who collected, handled, transferred, and examined a piece of evidence. For digital evidence, chain of custody includes...
- Code-Switching
- Shifting between different registers or even different languages within an interaction, often as a signal of role or authority. An interviewer who...
- Digital Written Language
- Text produced in digital communication contexts : SMS, instant messaging, social media posts, email : distinguished from formal writing by its brevity,...
- Idiolect
- The language variety specific to an individual, comprising their characteristic vocabulary, syntactic preferences, spelling habits, punctuation patterns, and discourse-level style. Authorship attribution...
- Metadata
- Data about data. In document forensics, metadata includes file-creation timestamps, last-modified dates, author fields, revision history, and embedded GPS coordinates in images....