Rolling Delta
Definition
Application of the Delta metric to a moving window of text, producing a curve showing how stylistic similarity to each candidate changes across the document. Used for collaboration and revision detection.
- Base metric
- Delta (Burrows' Delta stylometric distance)
- Applied over
- A moving window across the document
- Output
- A curve of stylistic similarity to each candidate across the text
- Used for
- Detecting collaboration and revision points
Common questions
How does Rolling Delta detect a co-authored passage that a whole-document Delta score would miss?+
A whole-document score averages style across the entire text, which can mask a short section written by a different author. The moving-window approach isolates local shifts, producing a visible spike or dip where the stylistic signature changes.
What window size choice affects the reliability of a Rolling Delta analysis?+
A window too small produces a noisy curve dominated by chance word frequency variation, while a window too large smooths over genuine short authorship shifts, so the window size must be tuned to the text length and the granularity of change being investigated.
Can Rolling Delta alone confirm which specific author wrote a flagged section?+
It identifies where the style shifts and how close that section sits to each candidate's baseline, but confirming authorship attribution for the flagged section typically requires supplementary stylometric tests, not the Rolling Delta curve alone.
Related terms
- Burrows's Delta
- A distance metric introduced by John Burrows in 2002 that measures how far a text's standardised word-frequency profile deviates from the mean...
- Closed-Corpus Trap
- The error of validating a stylometric method exclusively on texts from authors whose profiles were used to build the model, inflating apparent...
- Cosine Delta
- A variant of Burrows's Delta that uses cosine similarity instead of Manhattan distance. Research by Smith and Aldridge (2011) and later Eder,...
- Cross-Validation
- A validation strategy that repeatedly partitions a labelled corpus into training and test subsets to estimate the method's error rate on unseen...
- Principal Component Analysis (PCA)
- A multivariate statistical method used in fire debris research to reduce chromatographic data matrices to principal components that capture major variance. Used...